Image Works

Image Works System

Volume Analysis (option)

Perfusion II (option)

Navigator II (option)

DentaScan (option)

Volume Rendering (option)

Advanced Vessel Analysis (option)

Blank page.
 Table of Contents
Image Works System

Chapter 1 - Introduction
• Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-1
• Hardware . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-1
• Conventions for this Manual . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-1
• Start Image Works . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-2
• Types of Windows . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-3
The Browser . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-3
Viewer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-5
Mini Viewer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-6
• Other Application Buttons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-7
VA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-7
Add/Sub . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-7
Edit Patient . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-7
Film . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-7
Perfusion 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-7
Chapter 2 - Getting Started
• Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-1
• Image Works Browser . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-1
The Browser Menus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-2
• Viewer Menus and Controls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-5
Image Control Panel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-5
Chapter 3 - Selecting, Sorting and Displaying Images
• Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-1
• Selecting and Sorting Images . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-1
Sorting Examinations and Images . . . . . . . . . . . . . . . . . . . 3-4
• Displaying Images Using the Viewer . . . . . . . . . . . . . . . . . . . . . . 3-7
• Displaying Sorted Images . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-7
Image Works System

• Paging through Displayed Images . . . . . . . . . . . . . . . . . . . . . . . . . 3-7

• Closing the Viewer or a Mini Viewer . . . . . . . . . . . . . . . . . . . . . . . 3-10
• Deleting Images from Image Disk . . . . . . . . . . . . . . . . . . . . . . . . . 3-11
Chapter 4 - Manipulating Images
• Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-1
• Setting Window Width/Levels (W/L) . . . . . . . . . . . . . . . . . . . . . . . 4-2
• Magnifying Displayed Images . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-4
• Changing Image Orientation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-4
• Scrolling an Image . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-5
• View Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-5
• Returning an Image to Normal Display . . . . . . . . . . . . . . . . . . . . . 4-6
• Saving a Manipulated Set of Selected Images . . . . . . . . . . . . . . . 4-6
• Magnifying Glass . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-6
• Changing Custom Viewing Settings . . . . . . . . . . . . . . . . . . . . . . . 4-7
Annotation level . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-8
Right mouse button . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-8
Tick marks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-8
Grid Prefs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-9
Series binding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-9
Presets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-9
(Apply) vs. (Save as defaults) vs. (Cancel) buttons . . . . . . . 4-10
• Comparing two Series (VIEWER ONLY) . . . . . . . . . . . . . . . . . . . . 4-11
• Cross Reference Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-12
Cross Reference in Compare Mode (VIEWER ONLY) . . . . 4-12
Cross Reference Not in Compare Mode . . . . . . . . . . . . . . . 4-13
Chapter 5 - Annotating Images
• Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-1
• Understanding Annotation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-1
• Adding User Annotation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-2
• Removing and Restoring Image Annotation . . . . . . . . . . . . . . . . . 5-5
• Duplicating User Annotation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-5
 Image Works System

Chapter 6 - Analyzing Images

• Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-1
• Drawing Graphics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-2
Report Cursor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-2
Straight Line Distance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-3
Angle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-3
Rectangle Area . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-4
Smooth Curve Area . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-5
Ellipse Area . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-6
Free Draw Area . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-7
• Obtaining Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-8
• Accuracy of On-View Measurements . . . . . . . . . . . . . . . . . . . . . . 6-9
Accuracy with straight line distance graphic . . . . . . . . . . . . 6-9
Accuracy with angle graphic . . . . . . . . . . . . . . . . . . . . . . . . 6-9
Accuracy with region of interest area graphics . . . . . . . . . . 6-9
• Removing and Restoring Image Graphics . . . . . . . . . . . . . . . . . . 6-10
Chapter 7 - Film Composer
• Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-1
• Laser Camera . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-2
• Paper Printer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-2
• Activating the Film Composer . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-3
• Selecting Output Device . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-4
• Setting Filming Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-4
Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-4
Slide format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-5
Greyscale . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-5
Auto printing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-6
Auto clear page . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-6
Icon labels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-6
Expose order . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-6
Number of copies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-6
Closing the Print options window . . . . . . . . . . . . . . . . . . . . 7-6
Image Works System

• Loading Images in the Film Composer . . . . . . . . . . . . . . . . . . . . . 7-7

Drag-and-drop image loading . . . . . . . . . . . . . . . . . . . . . . . 7-7
Filming keyboard accelerator (F1 key) . . . . . . . . . . . . . . . . . 7-7
Page filming keyboard accelerator (F2 key) . . . . . . . . . . . . . 7-8
MID filming keyboard accelerator (F3 key) . . . . . . . . . . . . . 7-8
• Erasing Film Frames . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-9
• Printing Images . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-9
• Clearing the Film Composer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-9
• Film Composer Status Line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-10
• Print Series Function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-10
• Putting Text Pages on Film . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-12
Chapter 8 - Networking
• Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-1
• Information Concerning Connection with DICOM Equipment . . . . 8-2
• Selecting a Remote Host . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-3
[Select remote host] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-3
[Selected host: xxxx] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-3
• Transmitting Data to a Remote Host (Network Push) . . . . . . . . . . 8-4
• Removing Host Names from the Workstation's Configuration File 8-6
• Network Queues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-7
Chapter 9 - Archiving
• Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-1
• Composition of Archiving . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-1
• Restoring Images from OD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-2
Restrictions for Restoring Images from OD . . . . . . . . . . . . . 9-2
• Labelling a MOD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-6
• Saving Images on MOD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-8
Restrictions for Saving Images on MOD . . . . . . . . . . . . . . . 9-8
Procedure: Saving Images on MOD . . . . . . . . . . . . . . . . . . 9-9
• Ejecting a Disk from the OD Drive . . . . . . . . . . . . . . . . . . . . . . . . . 9-10
• Archive Queues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-10
 Image Works System

Chapter 10 - Image Combination

• Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-1
• Save Series . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-2
• Set Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-3
• Maximum Pixel Value Extraction . . . . . . . . . . . . . . . . . . . . . . . . . 10-4
Maximum pixel value extraction on one set alone . . . . . . . 10-4
Maximum pixel value extraction between two sets . . . . . . . 10-5
• Image Addition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-6
Image addition on one set alone . . . . . . . . . . . . . . . . . . . . . 10-6
Image addition between two sets . . . . . . . . . . . . . . . . . . . . 10-7
• Image Subtraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-8
Image subtraction between two images in one set alone . . 10-8
Image subtraction between two sets . . . . . . . . . . . . . . . . . 10-9
• Minimum Pixel Value Extraction . . . . . . . . . . . . . . . . . . . . . . . . . . 10-10
• Binding Series . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-10
• Closing the Image Combination Command Window . . . . . . . . . . 10-10
Image Works System
CHAPTER 1 Image Works System

Introduction
Overview
The Image Works is a multi-tasking, multi-functional, stand-alone, single-user worksta-
tion. It has its own dedicated computer and image data base. It supports image display,
manipulation, and film composing functions.
Image Works can be networked to serve as a common workstation for CT, MR and X-
ray products. It can also be networked to Advantage Windows workstations and/or GE
Independent Consoles, and certain other equipment that uses the DICOM standard. See
your GE Support Engineer.
Note: Image Works handles Advantage and DICOM image formats. However,
there are differences in layout between these formats.

Hardware
This software application runs on a standard Image Works platform. It uses the plat-
form's system for film output. It accepts image data acquired on any CT or MR system
compatible with Image Works.

Conventions for this Manual

Note: Throughout the text in this manual, we have used certain type styles and
symbols to differentiate between one tool or graphic and another. The con-
ventions we have used are as follows:
❏ Menu titles appear in bold face (e.g., Application Menu),
❏ Menu options appear in bold face, within brackets (e.g., [Exit]),
❏ On-screen tools appear in bold face, within braces (e.g., {Scroll Bar}),
❏ Graphical buttons appear in bold face, within parentheses (e.g., (View)),
❏ On-screen prompts and messages appear in italics (e.g., Login:),
❏ User typed-in responses appear in bold face italics (e.g., root), and
❏ Keyboard hardkeys and mouse buttons are underlined (e.g., Enter or left).

1-1
Image Works System

Start Image Works

Image Works can be accessed by clicking on the [Image Works] icon on the upper left
corner of the screen in either Scan or Display mode.

Click

1-2
 Image Works System

Types of Windows
There are three major windows in Image Works. They are;
❏ Browser,
❏ Viewer,
❏ Mini Viewer,

The Browser
The Browser window, (hereafter simply called "The Browser"), is the primary tool you
will use to select images for display and manipulation.
Restart Browser

Restart Display

Image File Search

A B
Application Selection Remove Sort Network Archive PPS Queue Utilities Messages
Examinations : Exam #47, May 05 92, SMITH, JON
Exam Name Date Description Mod Fmt A Ser Type Imgs Description Mod VA

3145 J.Herman Jan 08 98 CT 1 PROSP 18 CT Add/Sub

3512 B.Fox Dec 23 97 CT

Edit Patient

Film

Mini Viewer
2 examinations one series
Series
Img Img Ctr Thck/SP Gntry Img Ctr Img Ctr SFOV DFOV Perfusion2
(deg) Res Matrix Midscn Archive
S–I (mm) (mm) R–L A–P (cm) (cm)
Viewer
1 S 50.0 10.0/ +0.0 R5.0 A0.0 43 Lung 512 No
2 S 45.0 10.0/ +0.0 R5.0 A0.0 43 Lung 512 No
3 S 40.0 10.0/ +0.0 R5.0 A0.0 43 Lung 512 No
4 S 35.0 10.0/ +0.0 R5.0 A0.0 43 Lung 512 No
5 S 30.0 10.0/ +0.0 R5.0 A0.0 43 Lung 512 No
6 S 25.5 10.0/ +0.0 R5.0 A0.0 43 Lung 512 No

60 images

C
A: Examination List
B: Series List
C: Image List
In the upper right corner of the screen there are four function buttons.
Restart Browser use (Restart Browser) to open the Browser menu when the first-time
Restart Display
attempt fails. use (Restart Display) to open the Display menu when
the first-time attempt fails.
Image File Search

1-3
Image Works System

At the top of the Browser, you will find a menu bar, which offers pull-down menus.
Refer to the CHAPTER 2 for the details.
The Browser contains three lists showing the exams, series, and images presently on the
image disk:
❏ The Examinations list displays a list of available exams,
❏ The Series list displays a list of the series contained in the selected exam,
❏ The Images list displays a list of images within the selected series. CT, MR and
X-ray specific parameters are displayed.
Note: The Date column in the Examinations list reflects the acquisition date.
The lower right corner of each list displays the total number of items (exams, series or
images) contained in the list.
Above the Series list, the currently selected exam is described by the exam number,
exam date and patient's name.
At the bottom of the Browser window is a line of text which provides you with a contin-
uous report of the number of images currently stored on the image disk. It also updates
disk space availability. This is indicated as the number of 512 x 512 or 256 x 256 matrix
images that can be added to the image disk. The exact number varies with the acquisition
matrices of all the CT, MR and X-ray images on the disk.
The Browser also contains Application Selection Buttons, located on
the right side of the Browser window, which are used to run the various
VA image display and manipulation applications that are available on your
Add/Sub
system.
Edit Patient Since (VA), (Viewer), (Mini Viewer), (Add/Sub), (Film),and (Edit
Film
Patient) are part of the basic display system, these buttons are always
present, and their use is described in the following sections.
Mini Viewer

Perfusion2

Viewer

1-4
 Image Works System

Viewer
Click on the (Viewer) button on the right side of the Browser to open the following
Viewer. The Viewer is one way to display and manipulate the images you selected from
the Browser lists. It contains an image display area and an image control panel.

Image control panel

Browser Exam - +
Film Composer Series - +
Paging Image - +
Compare
Image
lmage
Analysis
Zoom
Presets

A A
Rect. Erase
Hide
Image display area
Matte All

Image Display
Measure
Enhance Normal

Format Reference Flip/Rotate

lmage

Film Film Film Text

Series Page MlD
<F4> <F2> Page
<F3>

Save User
Screen Prefs

❏ The image display area is used for viewing images in the series previously
selected on the Browser. This entire area can be formatted to display one single
image or it can be divided up to display multiple sequential images in the series.
Note: The date annotation in the upper right corner of each view reflects the exam
acquisition date.
❏ The image control panel on the left hand side of the Viewer is used to set up and
manage images on the image display area.
To close the Viewer you have to return to the Browser.

Mini Viewer

1-5
Image Works System

Click on the (Mini Viewer) button on the right side of the Browser to open the Mini
Viewer.
The Mini Viewer is another way to display and manipulate the images you selected from
the Browser lists. It contains an image control panel and image display area

Image control panel

Exam - +
Quit Series - +
Film Composer Image - +
Paging
lmage
lmage
Analysis
Prests
Zoom
Image display area
A A
Rect. Erase
Hide
Matte All

lmage Display
Measure
Enhance Normal

Reference
Format Flip/Rotate
lmage

Film Film
Text
Page MlD
<F2> Page
<F3>

Click on the (Quit) button on the Image control panel to close the Mini Viewer.

1-6
 Image Works System

Other Application Buttons

VA
Please refer to the Volume Analysis section.

Add/Sub
Please refer to the chapter 10 [Image Combination].

Edit Patient
Please refer to the following pages.

Perfusion 2
Please refer to the CT Perfusion 2 section.

DentaScan
Please refer to the DentaScan section

Film
Please refer to the chapter 7 [Film Composer].

1-7
Image Works System

Edit Patient
This function allows you to edit patient data after examinations.
The following things should be noted.
❏ The data you wish to edit must come from the system you use.
❏ Patient weight can not be changed.
❏ Patient ID can be changed only when "trauma" or "?" is in ID data field.
❏ It may take about one minute and 45 seconds to renew the whole exam that
includes 100 images.
❏ Data edit can be repeated many times.
❏ It is highly recommended that retro recon be done before you edit the patient data.
❏ It is highly recommended that archive, networking or filming be done after you
edit the data.
❏ If the exam you wish to edit is displayed in the autofilm viewport or free view-
port, you can not enter Edit Patient function.
❏ You can not enter Edit Patient when the exam you wish to edit is being used. The
following message will appear.
Error Detected

The exam selected can not be edited

because it is currently in use.

OK

❏ If you edit the exam that includes saved 3D models, those 3D models will be
deleted from the exam. Also, you can not enter Edit Patient after you selected 3D
model. The following message will appear.
Error Detected

Edit Patient Data failed to read

the current selection: delete if 3D model

OK

1-8
 Image Works System

In order to enter Edit Patient, first select an exam on the Browser, then click on the (Edit
Patient) button.
The following menu will appear. Select (Edit Patient Data).
Edit Patient Application

Please select one of the Edit Patient Applications

Edit Patient View Edit

Cancel
Data Log

The following message will appear. Select (Accept) to continue, or select (Cancel) to
cancel.

Caution

a) Networked, archived and filmed copies

of the original exam will have to be removed.
b) Use only the latest edited version of this exam.
c) 3D Models in this exam, if any, will be
deleted in the edited exam.

Accept Cancel

If you choose (Accept), the following message will appear. Select (Accept) to continue,
or select (Cancel) to cancel.

Additional Reminders

a) Copies of the exam may have been produced

with Auto Network, Auto Archive and Auto Film.
b) All reconstructed images should be generated
before editing this exam.
c) The exam in the Patient Schedule must be
manually updated.

Accept Cancel

1-9
Image Works System

Once you accept the warning message, the following Patient Data menu will appear.
Patient Information

Exam Number
Patient ID
Patient Name
Sex
Birthdate
Age
Refeming Physician
Radiologist
Operator
History

Exam Description

Edited by

Reset Selected Value Reset All Values

Accept Cancel

Click on the data field you wish to edit, then type in new data.
You can not proceed further unless you enter at least three letters into "Edited by" field.
After the edit, if there is any field you wish to undo, click there and select (Reset
Selected Value). If you wish to undo all edited fields, select (Reset All Values).
If you are satisfied with the edit, select (Accept) or click on (Cancel) to cancel.

1-10
 Image Works System

If you select (Accept) on the previous page, the following confirmation message will
appear.

Create a new exam with

the specified changes ?

Accept Cancel

If you select (Accept), the old exam will be replaced by the edited exam.
During the process where the old exam is deleted and the newly edited exam is created,
the following percentage countdown menu appears.

Closing application......

100%

Percentage of images

Upon the completion, the newly edited exam will be listed in the Browser. The newly
edited exam will be annotated with "e+ number" in the Description area of the Browser.
The number indicates how many times the exam was edited.
When you select (View Edit Log) in the following menu, an edit log will be displayed.
Edit Patient Application

Please select one of the Edit Patient Applications

Edit Patient View Edit

Cancel
Data Log

If you wish to exit from Edit Patient Application menu, select (Cancel).

1-11
Image Works System

Blank page.

1-12
CHAPTER 2 Image Works System

Getting Started
Overview
Learning about the various menu systems and controls offered by Image Works is impor-
tant for using the system effectively. This section provides an overview of all the work-
station's menus and controls, complete with general descriptions, so that you can see at a
glance where you need to go to use a particular function.
This section also describes the methods you can use to manipulate windows. By chang-
ing a window's location on the screen (and its size and shape where applicable), you will
be able to work with a variety of windows and images simultaneously.

Image Works Browser

Image Works menu and control are shown in the Browser menu.

A B
Application Selection Remove Sort Network Archive PPS Queue Utilities Messages
Examinations : Exam #47, May 05 92, SMITH, JON
Exam Name Date Description Mod Fmt A Ser Type Imgs Description Mod VA

3145 J.Herman Jan 08 98 CT 1 PROSP 18 CT Add/Sub

3512 B.Fox Dec 23 97 CT

Edit Patient

Film

Mini Viewer
2 examinations one series
Series
Img Img Ctr Thck/SP Gntry Img Ctr Img Ctr SFOV DFOV Perfusion2
(deg) Res Matrix Midscn Archive
S–I (mm) (mm) R–L A–P (cm) (cm)
Viewer
1 S 50.0 10.0/ +0.0 R5.0 A0.0 43 Lung 512 No
2 S 45.0 10.0/ +0.0 R5.0 A0.0 43 Lung 512 No
3 S 40.0 10.0/ +0.0 R5.0 A0.0 43 Lung 512 No
4 S 35.0 10.0/ +0.0 R5.0 A0.0 43 Lung 512 No
5 S 30.0 10.0/ +0.0 R5.0 A0.0 43 Lung 512 No
6 S 25.5 10.0/ +0.0 R5.0 A0.0 43 Lung 512 No

60 images

A: Examination List
B: Series List
C: Image List

2-1
Image Works System

The Browser Menus

Application Selection Remove Sort Network Archive PPS Queue Utilities Message

A menu bar, at the top of the Browser, offers the following pull-down menus:
The Application menu enables you to refresh (update) the Browser
Application lists.

Refresh lists

The Selection menu provides access to the multiple selection functions.

Selection Please refer to Chapter 3 for each function.

Select all examinations

Select all series

Select all images

Select archived exams

Select unarchived exams

Remove
The Remove menu provides access to the remove functions. These
functions allow you to delete the currently selected exam, series or
Remove examination images from the image disk. Please refer to Chapter 3 for each function.

Remove series

Remove image

Sort
The Sort menu allows you to sort listed and displayed exami-
nations and images. Please refer to Chapter 3 for each func-
Sort examinations by exam numder
tion.
Sort examinations by patient name
❏ Exam number
Sort examinations by date

Sort examinations by modaliy ❏ Patient name (alphabetical order)

Sort examinations by archive status ❏ Date
Sort examination by PPS status
❏ Modality (CT or MR)
Sort series by series number

Sort series by PPS status ❏ Archive status (Y=Yes, N=No, U=Unknown)

Sort images by image number
❏ Image number
Sort images by location
❏ Location
Sort image by echo
Sort images by trigger ❏ Echo (MR image),Trigger (MR image)
Sort images by scan time ❏ Scan time

2-2
 Image Works System

The Network menu allows you to send or retrieve exams, series, and
Network
images to or from other workstations on your network.
Receive Please refer to Chater 8 for each function.
Ping DICOM host

Send examination

Send series

Send image

Select remote host

2-3
Image Works System

The Archive menu is used for saving and restoring items (i.e. exams,
Archive
series, and images) on an external archive device (optical disk drive,
Restore
tape cartridge drive). Please refer to Chater 9 for each function.

Save examination

Save series

Save image

Label

Detach

Selected archive device

The Queue menu is used to access the various queues managed by the
Queue
Advantage Windows workstation (network, archive if the option is
Archive
installed, filming, etc.).

Network

Filming

The Utilities menu is used to copy a certain exam, series or

Utilities
image with anonymous patient name.
Create anonymous patient by exam Select an exam (xxx) you wish to copy with anonymous
patient name, then, click on [Create anonymous patient name
Create anonymous patient by series
by exam]. The question appears : (Create anonymous patient
Create anymous patient by image by exam ?). If you choose (Yes), the exam you selected is cop-
ied and listed as "Anonymous xxx".
The same procedure can be applied to series or image.
The Message menu item is used to open a window that provides a summary
Message
list of all the messages that have arrived since the Browser was opened. The
word (updating) appears to the right of this menu title whenever Image Works
database is being updated.

Note: The word (updating) does not disappear immediately after completion of
data base updating.

2-4
 Image Works System

Viewer Menus and Controls

The Viewer is launched from the Browser window.
Select the desired exam and series (and a subset of images if you don't want to view the
entire series), then click left on the (Viewer) button on the right side of the Browser win-
dow.

Image Control Panel

The Viewer screen contains numerous image display and manipulation controls which
are grouped into an Image Control Panel. This control panel is located on the left side of
the Viewer screen. It contains, starting from the top and moving downward :

Browser Exam - +
Film Composer Series - +
Paging Image - +
Compare
Image
Image
Analysis
Zoom
Presets

A A
Rect. Erase
Hide
Matte All

Image Display
Measure
Enhance Normal

Reference Flip/Rotate
Format
Image

Film Film Film Text

Series Page MID
Page
<F4> <F2> <F3>

Save User
Screen Prefs

2-5
Image Works System

Function of each button is as follows.

(Browser)

Browser Click left on the (Browser) button at the top of the Viewer Image Control
Panel to leave the Viewer screen and return to the Browser window, while
leaving the Viewer application active.

(Film Composer)

Film Composer
Click left on this button to display the Film Composer menu which allows you
to either to call up the Film Composer window, or to print the current page, a
Multi Image Display (MID), or the current series.

(Paging)

Paging This allows you to prescribe movie loop.

(Compare)
This button is used to enter the Compare mode, which allows viewing of sev-
Compare eral series independently on the Viewer.

(Image Analysis)

Image This button allows you to enter Reformat or Add/Sub function.

Analysis

(Presets)
The Presets W/L function allows you to select one of six preset windowing
Presets
Width/Level settings.

(Exam, Series, Image) next/prior

Exam - + These buttons allow you to go directly to the next or previous exam or series in
the Browser lists without having to return to the Browser and then re-open the
Series - + Viewer screen. Click on either (-) or (+) button to go to previous or next
exam,series or image, respectively.
Image - +

Note: If you have moved down to the last series in the exam and you then click
left on the down arrow button, the Viewer automatically "loops around" to
the first series in the same exam. Likewise, if you are in the first series in
the exam and you click left on the up arrow button, the Viewer automati-
cally "loops around" to the last series in the same exam.

2-6
 Image Works System

Image slider
You can also use Image slider to go through images forward or backward by
sliding it right or left.

Zoom slider
This slider allows you to zoom an image by sliding it right or left.

Magnifying glass
Click left on the magnifying glass button to use the right mouse button to acti-
vate the magnifying glass (a movable square region on the View which pro-
vides 2X magnification).

Scroll
Click left on the image scroll button to use the right mouse button for image
scrolling.

Rectangular Matte

Rect. Use this button to eliminate certain area of an image other than the designated
Matte rectangular area.

Grid
Use this button to put a grid on an image. The grid interval is adjusted in the
User Prefs icon.

2-7
Image Works System

Annotation
This button allows you to put an annotation on an image.
A

Erase
This button allows you to erase a certain selected annotation on an image.
A

(Erase All)
This button allows you to erase all graphics and annotations on all images.
Erase
All

(Hide / Show)
This button allows you to temporarily hide any kind of graphics or annotations
Hide on an image. Once graphics or annotations are hidden, the button turns to
(Show). Click on it to show graphics or annotations again.

(Measure)
This button allows you to create and edit graphical and text annotations for
Measure some measurments.

(Format)
The Format function is used to select the number of sequential images in a
Format series to be displayed on the image display area.This button provides eight
options of display format. Default format for Viewer is four on one and one on
one for Miniviewer.

(Image Enhance)

Image This button allows you to put enhancing filter on an image and it also allows
Enhance you to enhance a gray scale.

2-8
 Image Works System

(Reference Image)
This button allows you to display a Scout image in the lower right corner that
Reference
Image shows a line indicating axial location.

(Display Normal)

Display This button allows you to restore a normal display of an image.

Normal

(Flip / Rotate)
Use this button to flip or rotate an image.
Flip/Rotate

(Film Series) <F4>

Film This button allows you to print all or some of the current series on a film.
Series
<F4>

(Film Page) <F2>

Film This button allows you to print all images currently displayed in the Viewer.
Page
<F2>

(Film MID) <F3>

Film This button allows you to print Multiple Image Display on a film.
MID
<F3>

2-9
Image Works System

(Save Screen)

Save This button is used to save a manipulated image as a new series in the current
Screen exam.

(User Prefs)

User Click left on this button to open the Preferences command window which is
Prefs used to custom-set the following functions:
❏ Amount of annotation (on the screen and on film),
❏ Function of right mouse button (scroll or magnifying glass)
❏ Scales ("tick marks") on views shown or hidden
❏ Changing preset W/L values

Command Accelerator
The keyboard-controlled command accelerator field allows you to perform
certain functions rapidly by typing commands on the keyboard followed by
pressing the Enter key.
To enter a command, first activate the command accelerator field by moving the mouse
pointer inside it and clicking left, then type the command and press the Enter key.
To open an on-screen list of all the commands, type ? in this field and press the Enter
key. Use the {Scroll Bar} on the right side of the list to move up and down in it. To close
the list, click left on the (Cancel) button at the bottom of the list.
Note: Instead of closing the list and then typing the desired command manually
from the keyboard, you can have the command entered automatically by
clicking left on it. The list closes and the command appears in the com-
mand accelerator field. Type the desired parameters (if any) from the key-
board, then press the Enter key.

2-10
 Image Works System

A table of available commands is given below.

COMMAND ABBR. DESCRIPTION

? open a list of those commands with abbreviations and descrip-
tions

MOVIE LOOP COMMANDS ------ -------------------------

paging_interval <interval> ci set interval between cine loop frames (1=all images, 2=every
second image,etc)

paging pa run a cine loop with given range (image numbers) and rate
<start> <end> <rate> (frames per second)

EXAM/SERIES SELECTION ------ -------------------------

next_exam ne go to next exam in Browser

next_series ns go to next series in Browser

previous_exam pe go to previous exam in Browser

previous_series ps go to previous series in Browser

ANNOTATION LEVEL ------ -------------------------

annotation_none an display no annotations

annotation_partial ap display partial annotations

annotation_full af display full annotations

film_annotation_none fan no annotations on film

film_annotation_partial fap partial annotations on film

film_annotation_full faf full annotations on film

(continued)

2-11
Image Works System

COMMAND ABBR. DESCRIPTION

IMAGE MANIPULATION ------ -------------------------

autofit return to autofit zoom factor

flip_left_right flr flip image horizontally

flip_top_bottom ftb flip image vetically

rotate_left rl rotate image left by 90 deg.

rotate_right rr rotate image right by 90 deg.

normal no return image to normal orientation and autofit zoom factor

scroll <x> <y> scroll image by x and y pixels

format <rows> <columns> fo set display format to specified number of rows and columns

reset rs reset to display of images as they were received on worksation

window_level <value> wl set window level value

window_width <value> ww set window width value

zoom <factor> zo set zoom factor

invert inv inverse video

COMMAND ABBR. DESCRIPTION

GRAPHICS ------ -------------------------

cross_ref s<series no.>/ xr Refer to "Cross reference Mode"

<im1>-<im2>:interval
cross_ref_off noxr No cross reference for designated image

report_cursor rc make report cursor graphics

tick_marks <on>/<off> tm turn scale(tick mark) on or off

FILMING ------ -------------------------

print_page pp capture page for filming

print_series prs print selected series

text_page_exam te open scan data text page with exam information

text_page_series ts open scan data text page with series information

(continued)

2-12
 Image Works System

COMMAND ABBR. DESCRIPTION

MISCELLANEOUS ------ -------------------------

mouse_mode mm set right mouse button mode to scroll or magnifying glass

<scroll> / <magglass>
quit quit Viewer or Mini Viewer

save_state ss store save data parameters for images from first image to last one
<image1> <image2>
select_image <image no.> si select specified image as primary

set_image_of_interest select specified image as primary and display it on top left view
<image no.>

2-13
Image Works System

Blank page.

2-14
CHAPTER 3 Image Works System

Selecting, Sorting and Displaying

Images
Overview
Now that you have a general understanding about how the Image Works oper-
ates, you are ready to begin displaying images.
In this section, we provide step-by-step instructions to guide you as you select,
sort, and display images. Topics covered include:
❏ Selecting the exams, series, and images you want to display from the
Browser lists,
❏ Calling up the Viewer or Mini Viewer application for image display,
❏ Paging through displayed images - manually or using a movie loop,
❏ Sorting images according to echo sequence, location, image number, date
of scan, trigger, or patient name (vs. scan order), and then displaying them in
that order, and
❏ Removing images from your system disk.

Selecting and Sorting Images

When selecting images for display, you will be working from the Browser win-
dow.
This Browser window is displayed automatically when you log onto the Image
Works.
You will note that, in the Examinations list of the Browser, exams without patient
names are displayed as no-name-xxx. The xxx is a numerical suffix the worksta-
tion applies to distinguish different exams.
In the Series list, both MR and CT series show the series number, total number
of images, scanning range, and type of series (prospective, retrospective, etc.).
In addition, CT images indicate reference, while MR images show the acquisi-
tion plane and pulse sequence used.
In both the Series and Image lists, pertinent data acquisition parameters are dis-
played. These parameters are annotated in the same way that they are on the
CT HiLight or HiSpeed Advantage and MR Signa Advantage systems.
When you enter the Browser, the workstation automatically highlights (selects)
the first exam in the list, the first series of that exam, and the first image of that
series.

3-1
Image Works System

Procedure: Selecting Images for Display

To select a new exam, place the cursor on the exam of your choice. Then click
once with the left mouse button. The series list displays a list of series within the
newly selected exam.
Place your cursor on the series of your choice; click once with the left mouse
button to highlight it.
The third list gives the images within the selected exam and series. Using the left
mouse button, click on the image you would like to display.
If the currently highlighted options are not the ones you would like to
display, you can scroll through the listed exams, series and images to
find another selection, using the {scroll bars}.

3-2
 Image Works System

Procedure: Multiple Selection

Note: Multiple selection of exams, series and images on the Browser is useful
for item management (Remove, Archive Save, Network Push, etc.)
whereas multiple selection of images can also be used to select a
reduced group of images in a series to be used by the viewing applica-
tion.
To select more than one exam, hold the Control key down and click the left
mouse button on each exam you wish to select. The same procedure can be
used to select multiple series or multiple images.
To deselect a previously selected item, again hold the Control key down and
click the left mouse button on the item you wish to deselect.
If you select multiple exams or multiple series, the list of their components
(series or images) will not be displayed in the Browser. For example, if multiple
exams are selected in the Browser, the series and image list displays will appear
empty.
Selection of a new item with the left mouse button will reset the current selection.
For example, if more than one exam is currently selected and you click the left
mouse button on an exam (without pressing the Control key), that exam will be
the only one selected.
You can select a sequential range of exams, series or images using the Shift
key. To do so, select the first item you desire (click on the item with the left
mouse button). Next move the cursor to the last item you wish to select. Hold the
Shift key down and click the left mouse button. All the items between the first
and last items that you selected are now highlighted (selected).
Or, another way to select a sequential range of items is to press and hold left on
the first desired item, drag to the last desired item, and release the left mouse
button.

3-3
Image Works System

Procedure: Selecting All Exams, Series or Images

To select all the images, series or exams in the Browser, follow this procedure:

Selection
Click left on the Selection menu in the Browser menu bar.
Select one of the following three options:
Select all examinations
[Select all examinations] selects all the examinations in the
Select all series list,

Select all images [Select all series] selects all the series in the currently selected
exam,
Select archived exams [Select all images] selects all the images in the currently
selected exam and series.
Select unarchived exams
[Select archived exams] selects all archived exams that are
marked with "Y (Yes)".
[Select unarchived exams] selects all unarchived exams that
are marked with "N (No)".
Note: Archive status "U (Unknown)" is reported when that exam
originates from other station.

Sorting Examinations and Images

For ease of access, you can sort examinations by exam number,
Sort patient name, date and modality(CT or MR).
Sort examination by
examination number
Images appear in the Browser lists (by default) by increasing image
Sort examination by number. Most of the time, that is probably the way you will want them
patient name sorted.
Sort examination by
date But there may be exceptions. For instance, with MR images, you
Sort examination by might want to display a sequence of all second echo images in a
modality given series. The workstation can do that for you, sorting and dis-
Sort examination by playing MR images in order of echo.
archive status
Sort examination by Note: This option is only available for MR images with more than
PPS status
one echo. See your MR product manual for a detailed descrip-
Sort images by
image number tion of sorting by echo vs. sorting by scan sequence.
Sort images by
location
It is also possible to sort MR images by trigger delay.
Sort images by echo

Sort images by
trigger
Sort images by
scan time

3-4
 Image Works System

Procedure: Sorting Examinations By Exam Number, Patient

Name, Dat, Modality,Archive status and PPS status.
To sort examinations by examination number, patient name, date and modality,
follow this procedure:
Click left on the Sort menu in the Browser menu bar.
Sort
Select one of the following four options:
Sort examination by
examination number [Sort examinations by exam number] sorts entries in the Examina-
Sort examination by tions list of the Browser in the order of the examination numbers in
patient name the first column in the Browser list.
Sort examination by
date [Sort examinations by patient name] sorts entries in the Examina-
Sort examination by tions list of the Browser by alphabetical order of patient names. This
modality is the default sort mode for the Examinations list when the Browser is
Sort examination by first activated,
archive status
Sort examination by [Sort examinations by date] sorts entries in the Examinations list of
PPS status the Browser in reverse chronological order according to exam start-
ing time and date,
[Sort examinations by modality] sorts entries in the Examinations
list of the Browser by CT or MR.
[Sort examinations by archive status] sorts entries in the Exami-
nations list of the Browser by archive status.
[Sort examinations by PPS status] sorts entries in the Examina-
tions list of the Browser by PPS status.
Y(Yes) indicates that examination is archived. N(No) indicates that
examination is not archived. U(Unknown) indicates whether that
examination is archived is unknown since the examination originates
from other station.
Click left on one of these four options.

3-5
Image Works System

Procedure: Sorting Images by Number, Location, Echo, Trigger

Delay, or Scan Time
First select the series of which images you wish to sort.
Then click left on the Sort menu in the Browser menu bar.
Sort
[Sort images by echo] (MR series only) sorts images in an MR
Sort images by
image number series by increasing echo number, then increasing image number. All
Sort images by first echos are listed first, second echos listed next, etc.
location
[Sort images by trigger] (MR series only) sorts images in an MR
Sort images by
echo
series by trigger delay.
Sort images by [Sort images by image number] sorts images in the series by
trigger
increasing image number. This is the default sort mode when a
Sort images by series is first selected on the Browser.
scan time
[Sort images by location] sorts images in the series by increasing
location.
[Sort Images by scan time] sorts images in the series by scan time.
Click left on one of these two (or four for MR series) options.
Note: Once you select an image sorting mode and open the Viewer or a Mini
Viewer, that Viewer or Mini Viewer displays images in that order only. In
other words, when you open a Viewer or Mini Viewer, it is locked into the
current sort mode.
When you need to return to the default sort mode, go back to Sort menu, then
select [Sort images by image number] again.
You may then re-open the Viewer or open another Mini Viewer. The series will
now be sorted (and displayed) by image number.

3-6
 Image Works System

Displaying Images Using the Viewer

Once you have chosen the desired exam, series and image on the Browser, you
can use either the Viewer or a Mini Viewer to display the images.
Once the Viewer has come up on the screen, you can select the
desired display format by clicking left on the Format button in the
Formats
image control panel and then clicking left on the desired format from
the twelve options in the drop-down menu.
Note: You can also use the keyboard to obtain other display formats,
by means of the format command in the command accelera-
tor field. Type format followed by the number of rows and the
number of columns, separated by spaces, then press the
Enter key.
Note: The annotation level is automatically reduced when the
selected display format does not allow space for full annota-
tion. If all available annotation is not visible when you're view-
ing images, changing to a larger display format allows you to
obtain more annotation.

Displaying Sorted Images

Once you have started the Viewer, you can display and page through these
sorted images, using the methods described earlier.

Paging through Displayed Images

Once you have opened the Viewer and selected a format with one or more dis-
played images, the workstation allows you to page through the series of dis-
played images.
You can page backward or forward by clicking left on the left-arrow and right-
arrow buttons in the image paging control panel (in the center of the image con-
trol panel), by using the {scroll bar} or by using the Prior /Next keys on the key-
board.

3-7
Image Works System

Paging Button
You can open the cine loop control panel window by clicking left on the
Paging (Paging) button located at the upper left of the image control panel. A
sequential "projection" of a selected range of images in the series at a
speed of up to 30 images (frames) per second can be set up and exe-
cuted using the cine loop control panel.

Paging

Start

End

FPS

Temporal Spatial

Select Series Cancel

The various buttons and controls in this control panel are described below:

Start slider
This slider allows you to choose the first image in the series for paging. Clicking
left on the right or left arrow increases or decreases the number, respectively.

End slider
This slider allows you to choose the last image in the series for paging. Clicking
left on the right or left arrow increases or decreases the number, respectively.

FPS slider
This slider allows you to set the desired movie frame rate.
Hold left on the slider and drag left or right to the desired rate in frames per sec-
ond. This can be done prior to or during loop execution.
Note: If you leave the speed slider on 0, it will go to 10 images/second by
default upon loop start-up.

3-8
 Image Works System

Spatial Button
Click left on this button to obtain a repeating cine loop from front to back, then
back to front. An image set consisting of four images, for example, appears in
the following order: 1, 2, 3, 4, 4, 3, 2, 1, etc.
Click left on this button to obtain this type of movie loop.

Temporal Button
Click left on this button to obtain a continuous cine loop. For example, an image
set consisting of four images appears in the following order: 1, 2, 3, 4, 1, 2, 3, 4
etc.
Click left on this button to obtain this type of movie loop.

GO! Button
This button allows you to start the cine loop prescribed by the settings of the
above-mentioned controls.
Click left on this button to start the cine loop.
Before the cine loop starts running at the selected rate, there is a time delay
while the workstation loads the sequence of images.
Note: In the case of multiple image format display of a single series, the cine
loop always appears in the upper left view.
Note: When the cine loop is running, an fps field appears on the left of the fps
slider. It provides the true cine loop rate in frames per second, which may
be less than the speed selected with the fps slider.

Select Series Button

This button allows you to select another series for paging.

Cancel Button
This button allows you to exit from paging.

3-9
Image Works System

Closing the Viewer or a Mini Viewer

When you are finished working with an image, and you would like to close the
Viewer or Mini Viewer, follow the appropriate procedure below:

Procedure: Closing the Viewer prior to selecting another series

Click left on the (Browser) button in the lower left corner of the Viewer screen.
The workstation iconifies the Viewer application and returns you to the Browser.

Procedure: Closing a Mini Viewer

Click left on the (Quit) button in the lower right corner of the Mini Viewer win-
dow. The workstation quits the Mini Viewer application.

3-10
 Image Works System

Deleting Images from Image Disk

Occasionally, you may want to remove an image, series, or entire exam from the
image disk to make room for others. In this section, we describe how you can do
this using the Remove menu (located on the Browser menu bar).
Caution: Once you have deleted an exam, series, or image from disk, it is
gone for good. Use care in making your decisions. It is recom-
mended that you not delete images which have been networked or
filmed until you verify that these tasks have been successfully com-
pleted.

Procedure: Deleting Images, Series, or Exams

First select the exam, series, or image you want to remove using the left mouse
button to click on the desired selection to highlight it.
Click and hold left on the Remove menu. A pull-down menu appears.
Remove

Remove
This menu is for making your delete request. Do you want to delete
examination images from the workstation disk by exam, by series, or by image?
Drag the cursor to highlight your selection and release left.
Remove series
The Remove menu title appears in gray and cannot be used if a
Remove image selected item is in use by the Viewer or a Mini Viewer, if it is being used
by a network "Push" operation, or if it is being used by an archiving

A confirmation window asks you to verify your selection. To proceed with the
delete process, click left on [OK]. Or, to cancel the delete process, click left on
[Cancel].
The workstation removes the highlighted item from system disk, and automati-
cally updates the Browser lists.

3-11
Image Works System

Blank page.

3-12
CHAPTER 4 Image Works System

Manipulating Images
Overview
This section explains how to change the way your images are displayed in the
Viewer or a Mini Viewer. Just a few of the possibilities:
❏ Scrolling an image around within the Viewer or Mini Viewer,
❏ Using the view selection feature to perform most display functions (win-
dow width/level, magnification/reduction, image scroll, etc.) on selected
images within a series, without affecting the other images in that series,
❏ Removing user-applied annotation, graphics, min/mag factor, and image
orientation functions, using the Display Normal function,
❏ Viewing two series independently on the Viewer (compare mode),
❏ Viewing the cross reference of one series on another within the Viewer.
Note: When you apply the image manipulation functions explained in this chap-
ter (Manipulate, Flip, Rotate, Window W/L, etc.) to an image in a movie
loop, the system applies them to all the images in the image set.

4-1
Image Works System

Setting Window Width/Levels (W/L)

You can adjust window W/L to control which pixel values are visible in your
images, with what degree of contrast, in three ways:
❏ By making a selection from a palette of preset w/l values,
❏ By using the middle mouse button to make interactive adjustments, or,
❏ By using the keyboard up-, down-, left-, and right-arrow keys.

MR Images: WW WL
Minimum value 0 -32 767
Maximum value +65 535 +32 767

CT Images: WW WL
Minimum value 0 -32 767
Maximum value +65 535 +32 767

Procedure: Using Preset Selections to Adjust W/L

MR Images: CT Images:
W/L Button W/L Value W/L Button W/L Value
1 40/20 Abdomen 400/40
2 100/50 Head 100/35
3 200/100 Lung 1 000/-700
4 1000/500 Mediastinum 350/40
5 2000/1000 Spine 300/35
6 4000/2000 Vertebrae 2 000/350

To change window W/L values using the preset W/L buttons:

Click left on the preset W/L button of your choice.The workstation auto-
Presets
matically applies these values to your displayed image(s).
Abdomen
Head
Lung
Mediastinum
Spine
Vertabra

4-2
 Image Works System

Procedure: Using the Mouse to Adjust W/L Interactively

Increase level

Browser _ +
Exam
Film Composer Series _ +
Paging Image _ Decrease Increase
+
Width Width
Compare
Image 1
Image
Analysis
Zoom
Presets
Decrease level

A
Rect. Erase Hide
Matte All

Image Display
Measure Enhance Normal

Format Reference Flip/Rotate

Image

Film Film Film

Text
Series Page MID
Page
<F4> <F2> <F3>

Save User
Screen Prefs

To use the mouse to change W/L values in an image or group of images on the
Viewer or Mini Viewer:
Select the view(s) whose windowing you wish to modify. See "View Selection"
below.
Place your cursor on the view you wish to use as a real-time "monitor" of your
windowing changes.
Then, using the middle mouse button, press and drag on the image as illustrated
above. Note that the cursor changes into the graphic shown above.
Release the mouse button once you obtain the values you want. All selected
images are updated.
Note: If you perform windowing with the mouse on an unselected view, all unse-
lected views will be affected by the windowing, and all selected views will
NOT be affected.
Window/Level sensitivity increases as the cursor moves away from the point
where the middle mouse button was depressed. To decrease sensitivity at any
point, simply release and re-press the middle mouse button.

4-3
Image Works System

Procedure: Using Keyboard Arrow Keys to Adjust W/L

To adjust W/L using the keyboard arrow keys:
❏ To increase width, press the keyboard right-arrow (→) key,
❏ To decrease width, press the keyboard left-arrow (←) key,
❏ To increase level, press the keyboard up-arrow (↑).
❏ To decrease level, press the keyboard down-arrow (↓) key,
Note: Pressing an arrow key briefly provides a single increase/decrease by two,
whereas holding an arrow key down continuously provides accelerated
increase/decrease until the key is released.
Note: The change is first applied to the primary view (blue border) and then
propagates to the other selected views when you stop pressing the arrow
keys for a few moments.

Magnifying Displayed Images

The Image Works allows you to enlarge or reduce the size of your
image(s) using the Viewer or Mini Viewer zoom controls.

Note: The zoom factor value displayed above the zoom slider is with respect to
the number of pixels in the original image as it arrived on the workstation.
Note: When you reduce an image, the workstation automatically reduces the
quantity of annotations, if necessary, to avoid overflowing the display
area. For example, if all available annotation is not visible when you're
viewing a CT scout image, magnifying the image allows you to obtain
more annotation.
When you magnify an image, you can use the cursor to move the image around
within its frame. See "Scrolling an Image", in this chapter.

Changing Image Orientation

Using the (Flip/Rotate) button in the Viewer or Mini Viewer, you can
Flip/Rotate
change the orientation and size of images in the Viewer or a Mini
Viewer.
❏ Flipping the image(s) from left to right, or from top to bottom

❏ Rotating the image(s) by 90 degrees, in either a clockwise (right)

or counter-clockwise (left) direction

4-4
 Image Works System

Scrolling an Image
Sometimes an image is not displayed in its entirety, for instance when you have
magnified (i.e. zoomed in on) an image.
Whenever a portion of an image is outside the view, you can scroll the image
around in one of two ways:
❏ Click left on the image scroll button(palm icon), then scroll the desired
image by pressing right on it and dragging (if the image is selected, all
selected images will propagate when you release right, whereas if it is not
selected, there is no propagation to other images),
❏ While pressing and holding the Shift key on the keyboard, scroll the pri-
mary view (blue border) by pressing the keyboard arrow keys corresponding
to the desired scrolling direction (when you release the Shift key, the scrolling
propagates to all selected images).
Note: To scroll on a movie loop, the loop must first be running.

View Selection
Views can be selected for creation of text or graphical annotations, or for win-
dowing independently of the other images in the series.
To select a view, simply click left anywhere on it. Its border becomes blue, and it
is called the primary view. Only one view can be the primary view at a time.
The primary view (blue border) is the view that receives text or graphical annota-
tions when such annotations are called up from the Draw panel.
If you then click left on another view, it in turn becomes the primary view (blue
border) and the previous primary view becomes what is called a secondary view
(yellow border).
If you successively click left on various images, you will build up a group consist-
ing of secondary views and one primary view.
To select a view as primary and deselect all other views, double-click left on the
desired view.
To select a view as primary and all other views as secondary, triple-click left on
the desired view.

Procedure: Independent Windowing on Selected Views

Once you have selected the desired views as described above, you can perform
windowing with the middle mouse button as described above in "Procedure:
Using the Mouse to Adjust W/L Interactively" above. Only the selected views will
be affected by your windowing adjustments.

4-5
Image Works System

Returning an Image to Normal Display

The workstation allows you to restore normal image orientation and autofit zoom
factor following orientation and zoom operations.
This is done by clicking left on the (Display Normal) button.
Display
Normal

Saving a Manipulated Set of Selected Images

Once you have manipulated or analyzed an image, you can save it in its updated
form.
When you do so, the selected image is stored in a new series in the same exam-
ination and is listed in the Browser with the same series number but with SSAVE
marked in the Type column of the Series list.
This is done by selecting the desired image as primary (blue border),
Save
Screen
and then clicking left on the (Screen Save) button. The new series/
image will be available on the Browser within a short time.
Note: If you subsequently save another image from the same original series, it
is added to the end of the existing "SSAVE" series, and the value in the
Imgs column for the existing series increments by one.

Magnifying Glass
The magnifying glass function provides a moveable square zone on a view in
the Viewer or Mini Viewer which gives a 2X magnification of the view.
There are two ways to activate the magnifying glass:
❏ Move the mouse pointer to the part of the view you're interested in, and
press and hold the Shift key on the keyboard, then activate the magnifying
glass by pressing the middle mouse button (to deactivate magnifying glass,
release Shift key and middle mouse button)
or,

❏ If the magnifying glass button is not already activated, click left

on it, then move the mouse pointer to the part of the view you're
interested in, and activate the magnifying glass by pressing the right
mouse button (to deactivate magnifying glass, release right mouse
button).

4-6
 Image Works System

Changing Custom Viewing Settings

The following functions can be customized via the (User Prefs) button:
User
Prefs ❏ Amount of annotation (on the screen and on film),
❏ Function of right mouse button: image scrolling or magnifying
glass,
❏ Tick marks on views shown or hidden,
❏ Grid preference setting
❏ Series binding On/Off
❏ Square viewports On/Off
❏ Changing preset W/L names and values.
Click left on the (User Prefs) button in the Viewer or Mini Viewer to open the
command window which gives access to these functions.

Annotation level

Screen Film

No annotation No annotation
Partial annotation Partial annotation
Full annotation Full annotation
Custom annotation Custom annotation

Customize Customize

Tick Marks
Grid Prefs
Vertical
Customize
Horizontal
Right mouse button Series binding
Scrolling Series binding ON
Magnify glass Series binding OFF
Square viewports
Square viewports On
Square viewports Off
Presets

Modality CT

title w/width w/level

Abdomen 400 40 Set current
Head 100 1059 Set current
Lung 1000 324 Set current
Mediastinum 350 1064 Set current
Spine 300 1059 Set current
Vertebrae 2000 1374 Set current

Apply Save as defaults Cancel

4-7
Image Works System

Annotation level
This part of the command window allows you to set the amount of medical anno-
tation information on the screen and on film.
Click left on (No annotation), (Partial annotation), or (Full annotation) under
the Screen heading to set the amount of annotation on screen. Do the same
under the Film heading to set the amount of annotation on film.
You can also customize annotations for your own site.Click left on (Customize)
icon, then the following menu opens. Select your own choices and click on (OK).
You can access those annotations by selecting (Custom annotation).

Customize

1. hospital
2. patient
3. patient id
4. esi
5. scan time
6. scan
7. geometry
8. top
9. bottom
10. left
11. right
W. Windowing
G. Graphic results

OK Cancel

Right mouse button

This part of the command window allows you to customize the use of the right
mouse button on views.
Click left on (Scrolling) to use the right mouse button for image scrolling. This is
the standard use for the right mouse button. See "Scrolling an Image Around"
above in this chapter.
Click left on (Magnify glass) to use the right mouse button to activate the mag-
nifying glass. See "Magnifying Glass" above in this chapter.

Tick marks
This part of the command window allows you to turn the on-view tick marks on or
off.
Click left on (Shown) to turn on the tick marks, or on (Hidden) to turn them off.

4-8
 Image Works System

Grid Prefs
This part of the command window allows you to view the grid setteing.
Click left on (View) to look at the grid.

Series binding
This part of the command window allows you to turn the series binding on or off.
Click left on (Series binding ON) to activate the function, or on (Series binding
OFF) to deactivate it.

Presets
This part of the command window allows you to custom-set the six w/l preset
buttons in the Viewer or Mini Viewer.
First, click left on the (Modality) button and select [CT] or [MR] from the list that
pops up, according to which modality you wish to custom-set.
To store the current w/l values being used by the primary view in the Viewer or
Mini Viewer (the view with a blue border), click left on the (Set current) button
corresponding to the preset button desired. The w/width and w/level fields
change to the values used by the primary view.
To enter a w/width or w/level value manually, move the mouse cursor into the
field you want to change, double-click left to select it and enter the numeric value
you want from the keyboard.
To change a preset button title that will appear on the Viewer or Mini Viewer,
move the mouse cursor into the title field you want to change, double-click left to
select it and enter the alphanumeric title you want from the keyboard.

4-9
Image Works System

(Apply) vs. (Save as defaults) vs. (Cancel) buttons

Once you have set everything in the command window as you want, you can
make use of your new settings in one of three possible ways:
❏ Click left on the (Apply) button to apply the new settings ONLY to the
Viewer or Mini Viewer from which you called up the command window. In this
case, the new settings are enabled only for that Viewer or Mini Viewer, and
only until you quit that Viewer or Mini Viewer. The new settings are not saved
for any future use of the the Viewer or Mini Viewers.
❏ Click left on the (Save as defaults) button to replace the old stored set-
tings with the new settings. In this case, your new settings will be used as
default values by all uses of the Viewer or Mini Viewer.
❏ Click left on the (Cancel) button to completely discard your new settings.
In all three cases given above, the command window is closed as soon as you
click on the button.

4-10
 Image Works System

Comparing two Series (VIEWER ONLY)

To set up the compare mode, which allows you to view any two series indepen-
dently on the Viewer, proceed as follows:
❏ Select the desired first set of image(s) on the Browser and open the
Viewer normally,
❏ Click left on the (Compare) button in the Viewer.The Browser window re-
appears,
❏ Select the desired second set of image(s) on the Browser and click left on
the (Viewer) button. The Viewer screen returns with the compare mode acti-
vated.
When the compare mode is activated, the Viewer displays the two selected
series. Click left on the desired format button in the upper left corner of the
Viewer screen: either one or two images can be displayed from each series, and
the two series can be displayed side by side or one above the other.
Note: A separate set of image scroll (Prior/Next) buttons appears at the bot-
tom of the Viewer for each of the two series. However, the Prior/Next
keys on the keyboard affect both series simultaneously.
To quit the compare mode, click left on the (Cancel Compare) button at the bot-
tom of the Viewer. The Viewer returns to normal operation with the first series
displayed.

4-11
Image Works System

Cross Reference Mode

The Viewer or a Mini Viewer can be set up to display cross reference lines on a
view which represent the slices in a different series.
Note: Cross reference lines may overlap annotations. For this reason, when
filming, it is recommended to film the reference view twice: once with
cross reference deactivated and once with it activated.
For the Viewer, there are two ways to set up cross reference lines, depending on
whether the compare mode is activated. The following two sections describe
these two cases.

Cross Reference in Compare Mode (VIEWER ONLY)

Once you have set up the Viewer in compare mode (see "Comparing two Series
(VIEWER ONLY)" above), you can activate the cross reference mode as follows:
❏ Choose the view on which you would like to display cross reference lines.
Select this view as the primary view (blue border) by clicking left on it,
❏ Type the command xr on the keyboard and press Enter. The cross refer-
ence lines appear on the primary view. They are numbered according to the
slice numbers of the corresponding series,
❏ To view a particular slice represented by a cross reference line on the pri-
mary view, click left on that line; the corresponding slice appears on the part
of the Viewer displaying the corresponding series.
To deactivate the cross reference mode, type the command noxr on the key-
board and press Enter. The cross reference lines disappear.

4-12
 Image Works System

Cross Reference Not in Compare Mode

When using the Viewer without the compare mode (i.e. when using the Viewer
normally) or a Mini Viewer, you can activate the cross reference mode as fol-
lows:
Choose the view on which you would like to display cross reference lines. Select
the view as the primary view (blue border) by clicking left on it,
Type the command:
xr s<series no.>/<im1>-<im2>:<interval>
where <series no.> is the desired series number as displayed on the Browser,
<im1> is the first image number in the range, <im2> is the last image number in
the range, and <interval> is the increment between cross referenced images
represented. Press Enter to validate the command. The cross reference lines
appear on the primary view. They are numbered according to the slice numbers
of the corresponding series.
To deactivate the cross reference mode, type the command noxr on the key-
board and press Enter.
For example:
xr s3
activates the cross reference mode with series no. 3 (all images, interval of one
image),
or:
xr s3/10-60:5
activates the cross reference mode with series no. 3, from image no. 10 to
image no. 60, at and interval of 5 images.

4-13
Image Works System

Blank page.

4-14
CHAPTER 5 Image Works System

Annotating Images
Overview
You will work with two kinds of annotation on the Image Works Workstation: the
information which the workstation provides about exams, series, and images,
and the annotation that you add to images.
This section discusses both types of annotation, including:
❏ How to add your own annotation to an image, and
❏ How to remove annotation temporarily (i.e. hide it), so that you can view
an image more clearly,
❏ How to remove annotation permanently.

Understanding Annotation
Much of the annotation on both CT and MR images is self-explanatory: data
identifying the patient, hospital, physician(s), and image parameters. See your
CT or MR product manual for an explanation of the abbreviations used in image
annotation.
Note: If the caret ( ^ ) character is entered into the patient name field on the
medical imaging system, it is replaced with a space on the Image Works
workstation.

5-1
Image Works System

Adding User Annotation

The Image Works allows you to add your own annotations, for example com-
ments or observations, to an image to draw attention to a specific region of anat-
omy.
By using this user annotation capability, you can add multiple areas of annota-
tion to an image and then delete that annotation later, if you choose. An annota-
tion can also include an on-screen arrow for pointing out specific structures.

Procedure: Adding User Annotation

Select the image onto which you want to add annotation by clicking left on it (it
becomes the primary view and has a red border around it
Click left on the (A) button in the Viewer or Mini Viewer.
A A white box and arrow graphic appear on the selected image.

The box graphic provides the field for entering text, while the arrow allows you to
point out a specific structure.
You are ready to begin entering the text for your annotation. To do so, first move
the mouse pointer onto the view and type the text from the keyboard.
Note: The mouse pointer MUST stay on the view for the duration of annotation
text entry.
As you type, the box expands to accommodate the text.

5-2
 Image Works System

You can start a new line of text by pressing the Enter key.

this is

this is sample text

this is the 2nd line

To move the box and/or arrow after entering your annotation, follow the proce-
dure given below.
Click and hold left on the annotation and drag it to the new location. Release the
mouse button to deposit the box.
The arrow's stem rotates and changes in length ("rubber band" effect) as it fol-
lows the annotation, while its point remains in its original location.

5-3
Image Works System

To move the arrow's point, click and hold left on it. Then drag it across the image
until it is pointing at the structure that you are interested in. Release the mouse
button to deposit the point in its new location.

Note: If you want a freestanding text annotation, without an arrow, press and
hold left on the arrow's point and drag it into the box. When you release
the mouse button, the arrow is no longer visible.
When you have finished adding your annotation and positioning it, you can
deselect it by pointing outside it and clicking left. The box disappears and the
entire annotation turns brown.

5-4
 Image Works System

Procedure: Deleting User Annotation

Once you have added annotation, you can then remove it as follows:
Click left on the annotation that you wish to remove. The entire annotation
becomes red and the box reappears around the text.
Click left on the eraser button in the Draw panel of the Viewer or
A Mini Viewer. The selected annotation disappears.

Procedure: Deleting All User Annotation

You can also remove ALL added annotation from ALL images:
Click left on the (Erase all) button in the Viewer or Mini Viewer.
Erase
All annotations disappear from all images.
All

Removing and Restoring Image Annotation

You can remove image annotation temporarily by clicking left on
Hide the (Hide) button. All annotations and graphics disappear from
the image and the (Hide) button becomes the (Show) button.
To restore hidden annotations and graphics to the image, click left
on the (Show) button. All hidden annotations and graphics are
restored to the image and the (Show) button again becomes the
(Hide) button.

Duplicating User Annotation

If you need to repeat the same user annotation in several places, you can do so
as follows:
❏ Click left on the annotation that you wish to duplicate. It becomes red and
the box reappears around the text,
❏ To create a duplicate of the annotation and leave the original in place,
press the Copy key on the left side of the keyboard, or, to create a duplicate
and delete the original, press the Cut key on the left side of the keyboard,
❏ Select the desired target view for the duplicate annotation by clicking left
on it (the target view can be any view in the Viewer or in any Mini Viewer),
❏ Press the Paste key on the left side of the keyboard to place the duplicate
of the annotation on the target view.

5-5
Image Works System

Blank page.

5-6
CHAPTER 6 Image Works System

Analyzing Images
Overview
Sometimes it is important to know exact measurements; for instance, the dis-
tance between two points on an image, the size of an angle, or the area of a par-
ticular region of interest. The Image Works allows you to analyze images in a
variety of ways to obtain the measurements you need. Your options include:
❏ Defining a Region of Interest (ROI),
❏ Measuring the area, mean, standard deviation, and minimum and maxi-
mum pixel values within an ROI,
❏ Measuring the distance between two points,
❏ Measuring an angle,
❏ Measuring the pixel intensity value at a specific point on the image,
❏ Removing and restoring image graphics,
❏ Duplicating image graphics.
Note: ROI, distance and angle measurements are ONLY possible if the viewing
format is 2 x 2 or less for the Viewer, and 1 x 1 for the Mini Viewer.
Note that once you have analyzed an image, you can save it under a new head-
ing.
Each graphic on a view is assigned a number. This number also appears to the
left of its statistics data (distance, area, mean, etc.) in the lower right corner of
the view. In this way, several graphics can be defined on a view and the statistics
of each one can be easily distinguished from the others.
Note: The black background outside the reconstruction circle of CT images
uses a value of -1024 for all calculations.

6-1
Image Works System

Drawing Graphics
To draw a graphic on a view, first select the view as primary by clicking left any-
where on it.
Click left on the (Measure) button in the Viewer or Mini Viewer, then
Measure click left on one of the seven graphic buttons (straight line distance,
angle, rectangular area, smooth curve area, free draw area, or ellipse
area). The corresponding default graphic appears in blue on the view,
and its statistics appear in the lower right corner of the view.

Report Cursor
To obtain a report cursor on the selected view, click left on the cor-
responding button in the Draw panel. The report cursor appears as
an "X" symbol in the center of the view.
In the upper left corner of the view the pixel value (V), and the anatomical coordi-
nates (Left, Right, Superior, Inferior, Anterior and Posterior) in mm, correspond-
ing to the current position of the report cursor, are displayed.
To move the report cursor, click and hold left on it and drag it to the desired posi-
tion. The V and anatomical coordinate values in the upper right corner of the
view are updated in real time.
Note: Regardless of the zoom factor being used to view images, report cursor
information is calculated based on the pixel from the original,
UNZOOMED image data as they arrived on the Image Works worksta-
tion.

6-2
 Image Works System

Straight Line Distance

To obtain a straight line distance graphic on the selected view,
click left on the corresponding button in the Draw panel.

To resize the line, drag left on either endpoint (box) to move it.
The other endpoint stays fixed and the segment resizes as you
drag (rubber band).
The statistics data for the straight line appear in the lower right corner of the view
and include distance in mm and angle in degrees of the line with respect to the
vertical axis.
Note: See "Accuracy of On-View Measurements" below in this chapter for infor-
mation about measurement accuracy.
To move the entire line without resizing, drag left anywhere on it except the end-
points to move it.

Angle
To obtain an angle graphic on the selected view, click left on the
corresponding button in the Draw panel.

To resize the angle, drag left on any one of the three points
(boxes) defining the angle to move it. The other two points stay
fixed and the segment(s) of the angle resize as you drag (rubber
band).
The statistics data for the angle appear in the lower right corner of the view and
include the angle in degrees.
Note: See "Accuracy of On-View Measurements" below in this chapter for infor-
mation about measurement accuracy.
To move the entire angle without resizing, drag left anywhere on it except the
three defining points to move it.

6-3
Image Works System

Rectangle Area
To obtain a rectangle area graphic on the selected view, click left on
the corresponding button in the Draw panel.

To resize the rectangle, drag left on one of the four corner points
(boxes) of the rectangle to move it. The other three points stay fixed
and the rectangle resizes as you drag (rubber band).

The statistics data for the rectangle appear in the lower right corner of the view
and include:
❏ Minimum and maximum pixel values [MIN, MAX] within the rectangle,
❏ Mean pixel value ("mean") within the rectangle,
❏ Standard deviation ("sd") of the pixel values within the rectangle (this pro-
vides a measure of variability according to pixel values),

❏ Area ("area") enclosed by the rectangle, in mm2.

To rotate the rectangle, drag left on one of the its four rotation points
(small dashes located in the middle of each side).

Note: The ROI area measurement can vary by up to 5% as a result of this rota-
tion.
To move the rectangle, drag left on the rectangle's center point
(small "+" symbol located at the center of the rectangle) or anywhere
on the sides of the rectangle except its four corner points and four
rotation points.

6-4
 Image Works System

Smooth Curve Area

To obtain a smooth curve area graphic on the selected view, click
left on the corresponding button in the Draw panel.

This graphic appears initially with one point (box).

To add a point to the end of the curve, move the mouse pointer to the desired
location for the new point (it must NOT be on the curve), hold down the Shift key
and click left.
To add a point between two existing points on the curve, move the mouse
pointer to the desired location for the new point (it must be ON the curve), hold
down the Shift key and click left.
To move a point on the smooth curve, drag left on it. All other points
stay in their original position and the curve resizes as you drag.

Note: To close the endpoints of a smooth curve, first add the last point near (but
not on top of) the first point, then move it on top of the first point.
The statistics data for the smooth curve appear in the lower right corner of the
view and include:
❏ Minimum and maximum pixel values [MIN, MAX] within the smooth curve,
❏ Mean pixel value ("mean") within the smooth curve,
❏ Standard deviation ("sd") of the pixel values within the smooth curve (this
provides a measure of variability according to pixel values),

❏ Area ("area") enclosed by the smooth curve, in mm2.

To move the smooth curve, drag left anywhere on it except the points (boxes).
To delete a point on the smooth curve, while holding down the Shift key, click left
on the point to be deleted.

6-5
Image Works System

Ellipse Area
To obtain an ellipse area graphic on the selected view, click left on
the corresponding button in the Draw panel.

To resize the ellipse, drag left on one of the four boxes just outside the ellipse.
The ellipse resizes as you drag (rubber band).
The statistics data for the ellipse appear in the lower right corner of the view and
include:
❏ Minimum and maximum pixel values [MIN, MAX] within the ellipse,
❏ Mean pixel value ("mean") within the ellipse,
❏ Standard deviation ("sd") of the pixel values within the ellipse (this pro-
vides a measure of variability according to pixel values),

❏ Area ("area") enclosed by the ellipse, in mm2.

To rotate the ellipse, drag left on one of its four rotation points (small dashes
located in the middle of each side) to rotate the ellipse.
Note: The ROI area measurement can vary by up to 5% as a result of this rota-
tion.
To move the ellipse, drag left on the its center point (small "+" symbol located at
the center of the ellipse) or anywhere on the sides of the ellipse to move the
ellipse.

6-6
 Image Works System

Free Draw Area

To obtain a free drawn area graphic on the selected view, click left
on the corresponding button in the Draw panel.

The graphic appears initially with one point (box).

Drag left on this box to move it to the desired starting point.
To start drawing, while holding down the Shift key, drag left on the box.
Or, to define a series of segments, while holding down the Shift key, click left on
each desired segment endpoint location.
To remove the most recently created segment, press the Back Space key. The
workstation may take a few moments to delete the segment.
The statistics data for the free drawn curve appear in the lower right corner of
the view and include:
❏ Minimum and maximum pixel values [MIN, MAX] within the free drawn
curve,
❏ Mean pixel value ("mean") within the free drawn curve,
❏ Standard deviation ("sd") of the pixel values within the free drawn curve
(this provides a measure of variability according to pixel values),

❏ Area ("area") enclosed by the free drawn curve, in mm2.

To move the free drawn curve, drag left anywhere on it.

6-7
Image Works System

Obtaining Measurements
Once you place a graphic on a view, the workstation automatically computes the
appropriate measurements and statistics and displays them in the lower right
corner of the view. The measurements and statistics computed and displayed
depend on the type of graphic being used.
When you use a report cursor, the workstation computes the pixel intensity value
V and the report cursor's anatomical coordinates.
When you use a straight line distance graphic, the workstation computes both
the segment's length in mm and the angle the segment creates in relation to the
horizontal axis, in degrees.
When you use an angle graphic, the workstation computes the size of the
defined angle, in degrees.
When you use a rectangle, smooth curve, ellipse, or free drawn graphic, the
workstation computes the following measurements:

❏ The AREA enclosed by the graphic in mm2,

❏ The MEAN value of the pixels enclosed by the graphic,
❏ The minimum and maximum [MIN, MAX] pixel values enclosed by the
graphic,
❏ The standard deviation (SD) of the pixel values enclosed by the graphic,
which provides a measure of variability according to pixel values.
The workstation automatically updates each of these measurements anytime
you modify the graphics.
Note: When a Screen Save image is being viewed, the report cursor is accom-
panied by pixel intensity value V only, and all other graphics are accom-
panied by NO measurements and statistics.

6-8
 Image Works System

Accuracy of On-View Measurements

This section provides information about the accuracy of measurements obtained
with on-screen graphics.
Note: Regardless of the zoom factor being used to view images, region of inter-
est statistics are calculated based on the pixels from the original,
UNZOOMED image data as they arrived on the Image Works worksta-
tion.

Accuracy with straight line distance graphic

Measurement accuracy using the straight line distance graphic is equal to the
displayed length +/- image pixel size.

Accuracy with angle graphic

Measurement accuracy using the angle graphic is equal to the displayed angle
value +/- 10 degrees for an angle measured between segments which are five
times larger than the image pixel size.
Accuracy improves as the length of the segments increases.

Accuracy with region of interest area graphics

Area measurement accuracy using a region of interest graphic (rectangle,
smooth curve, ellipse or free draw) is equal to the displayed area value +/- the
circumference of the region multiplied by (image pixel size)2 / 2. Mean and stan-
dard deviation values for the intensity of the pixels in the region are also affected
by this accuracy.
Region of interest statistics are based on the pixels INSIDE the graphic defining
the region.

6-9
Image Works System

Removing and Restoring Image Graphics

The Image Works workstation allows you to remove image graphics temporarily,
or permanently, depending on your needs. Here's how:

Procedure: Removing and Restoring Image Graphics

f you wish to remove graphics from all views temporarily, click left
Hide on the (Hide) button. All graphics disappear from all views and
the (Hide) button becomes the (Show) button.
To restore hidden graphics to all views, click left on the (Show)
button.
All graphics re-appear on all views and the (Show) button
becomes the (Hide) button again.
f you wish to remove a graphic from the view permanently, select
A it by clicking left on it, then click left on the eraser button.
Or, press the Cut key on the left side of the keyboard. See "Dupli-
cating Image Graphics" below in this chapter.
Note: f you click left on the eraser button without first selecting a graphic object,
the report cursor, if there is one, is erased.

Erase Or, to remove all graphics from all views permanently, click left on
All the (Erase All) button in the Draw panel.
This removes all graphics from all views. (You cannot restore
graphics which you have removed via the (Erase All) button.)

6-10
CHAPTER 7 Image Works System

Film Composer
Overview
The Image Works workstation allows you to send the images you have dis-
played, analyzed, and manipulated to a laser camera for film output or to a
printer for paper output. To do so, your workstation must be linked to an appro-
priate output device. See your GE Sales Representative and Service Engineer
for details.
In this chapter, the word "film" or "sheet of film" means either a sheet of film or a
sheet of paper, depending on the output device being used.
The Image Works workstation gives you a choice of print formats so that you can
lay out your images in different ways on each film, depending upon your needs.
Your format choices typically include 1x1, 2x2, 2x3, 3x4 and 4x5, although oth-
ers are possible depending on the type of output device being used. The work-
station also allows you to place images from multiple viewing applications (i.e.
different Mini Viewers, 3D, Reformat, etc.), and thus multiple studies, on the
same film.
There are three primary steps to the filming process: setting filming parameters,
loading the images, and printing.
Setting filming parameters means telling the workstation:
❏ How many images (typically up to twenty) you would like printed on each
sheet of film,
❏ How many copies (up to ten) you would like of each film,
❏ Whether you would like to activate printing of your films manually (by click-
ing on a button) or automatically (as soon as all film frames in the Film Com-
poser are full).
Loading simply means displaying your image(s) within one or more viewing
applications, and then telling the workstation where you want each image to be
placed on the film.
Printing is the act of transferring the image(s) onto a sheet of film once you have
completed the first two steps.
There is also a "Print Series" function available in the Viewer or Mini Viewer. It
allows you to set up printing of all or part of the series currently being viewed.
See "Print Series Function" later in this chapter.

7-1
Image Works System

Laser Camera
See the documentation provided by the manufacturer of the laser camera for a
description of the parts and controls on this camera and for general operating
instructions.
Note that there are adjustments you can make on the camera to tailor the pre-
sentation of your images. These can include density and contrast adjustments,
your choice of black or clear for blank frames, and borders between frames.
These values are typically set by your GE Service Engineer and are not adjusted
during routine filming.

Paper Printer
See the documentation provided by the manufacturer of the paper printer for
general operating instructions.

7-2
 Image Works System

Activating the Film Composer

If you are using the Viewer or a Mini Viewer, activate the Film Composer by
clicking left on the (FilmComposer) button in the Viewer or Mini Viewer. The
Film Composer window appears in the lower right corner of the monitor.
Or, go to the Browser and click left on the (Film) button.

Film Composer

Formats Laser Camera

Options
Current status is :
Clear

Print

7-3
Image Works System

Selecting Output Device

Before setting film parameters or composing a film, you must select the desired
output device (typically a choice between a laser camera and a paper printer).
To do this, press and hold left on the current output device name at
Laser Camera
the top of the Film Composer window, then select the desired device
from the list that pops up.
The output device name, as well as the available (Formats) buttons, are
updated according to your selection.
Note: The output device can be changed when images are loaded in the Film
Composer if the currently selected (Formats) button is compatible with
the new output device being selected.

Setting Filming Parameters

Format
The film format is set by clicking left on the desired (Formats) button in
Formats the Film Composer window.
Note: The available formats may vary depending on the type of laser
imager or printer being used.
The number of filming frames in the Film Composer is updated accord-
ing to your choice of format. (For example, the drawing above left
shows a 3 x 4 film format selected.)

7-4
 Image Works System

Options
Next click left on the (Options) button in the Film Composer window.
A window entitled "Print options" pops up:

Print options
Slide format: Greyscale:
Off Off

Auto printing: Auto clear page:

Off Off

Icon labels: Expose order:

Image Left/Right
Top/Bottom
Number of copies:
1
Done

Slide format
The Film composer can be set up to produce images in slide format.
Slide format: Press and hold left on the (Slide format) button in the Print options
window and select [On] or [Off] from the list that pops up.
Off

Note: If the selected output device does not support the slide format function,
this function is not available and the (Slide format) button appears in
grey.

Greyscale
Greyscale: If you want a greyscale printed along the side of every sheet of film,
click left on the (Greyscale) button in the Print options window and
Off select [On] from the list that pops up.
To deactivate the gray scale, repeat as above, but select [Off] from the
list that pops up.
Press and hold left on the (Greyscale) button in the Print options win-
dow and select [On] or [Off] from the list that pops up.

Note: If the selected output device does not support the greyscale function, this
function is not available and the (Greyscale) button appears in grey.

7-5
Image Works System

Auto printing
The Film Composer can be set up to print your film automatically as soon as all
film frames in it are full.
Auto printing: Press and hold left on the (Auto printing) button in the Print
options window and select [On] or [Off] from the list that pops up.
Off

Auto clear page

The Film Composer can be set to automatically clear all its film frames immedi-
ately following printing.
Auto clear page: Press and hold left on the (Auto clear page) button in the Print
options window and select [On] or [Off] from the list that pops up.
Off

Icon labels
Icon labels: Press and hold left on the (Icon labels) button in the Print options
window and select [E/S/I] or [Image] from the list that pops up.
Image

Expose order
Expose order: Press and hold left on the (Expose order) button in the Print
Left/Right options window and select [Left/Right, Top/Bottom] or [Right/Left,
Top/Bottom Bottom/Top] from the list that pops up.

Number of copies
Number of copies: To set the number of copies to be printed of each film, click left on
the (+) or (-) button to the right of the Number of copies field in the
1
Print options window to increase or decrease the number by one,
respectively.
Or, move the mouse cursor into the field, double click left to select it and enter
the desired number from the keyboard.
Note: The maximum possible number of copies is 10.

Closing the Print options window

To close the Print options window, click left on the (Done) button.
Done
The window closes and all the settings in it are enabled.

7-6
 Image Works System

Loading Images in the Film Composer

This section describes the various ways of loading images onto the Film Com-
poser.
Bear in mind that you must perform all image manipulation or analysis functions
on these images before filming them.
Note: Annotations on filmed images appear according to the film annotation
level selected in the User Prefs command window, should filmed annota-
tions overlap or run off the edge of film frames, either reduce the film
annotation level or change the film format to accommodate the film anno-
tations.
Note: The appearance of images in film frames in the Film Composer is not
identical to the appearance of those images in the viewing application.
Note: If an image is reduced to a size smaller than the available viewing area
on the Viewer or Mini Viewer, and then filmed, it is enlarged so as to com-
pletely fill the film slot that it occupies.
Caution: Verify that films are acceptable before deleting the images used to
produce them.

Drag-and-drop image loading

When you are ready to load an image being viewed into the Film Composer,
move the mouse pointer onto it, press and hold left and drag to the desired avail-
able film frame in the Film Composer. Release left to deposit the image. A few
seconds later a miniature reproduction of the image appears in the film frame.
Repeat this drag-and-drop procedure as many times as you wish to complete
the film grid. It is not necessary to fill all the film frames (unless you activated
(Auto printing), nor to place images into consecutive frames.

Filming keyboard accelerator (F1 key)

Instead of dragging and dropping an image into the Film Composer as described
above, you can load it by moving the mouse pointer onto it and pressing the F1
function key on the keyboard. The image is placed in the next available film
frame in the Film Composer.
The advantage of this method is that it can be quicker than dragging and drop-
ping. However, you cannot skip film frames using this method.

7-7
Image Works System

Page filming keyboard accelerator (F2 key)

The page filming feature allows you to load a whole page (i.e. all displayed
images in the viewing application) of images into the Film Composer.
The Film Composer must be empty before this function can be used. If it isn't,
use the (Clear) button in the Film Composer window. See "Clearing the Film
Composer" below.
Move the mouse pointer onto any image in the viewing application and press the
F2 function key on the keyboard. The images are loaded into the frames in the
Film Composer in the order in which they are displayed in the viewing applica-
tion.
Note: The viewing application format and Film Composer format must be identi-
cal for page filming to function. If they are different when you press F2,
the Film Composer automatically changes format to match that of the
viewing application. If the viewing application format is unavailable on the
Film Composer, a message pops up and informs you of this when you
press the F2 key.
Note: If you are using the Viewer, you can also activate page filming by clicking
left on the (Capture Page <F2>) button in the lower left corner of the
monitor.

MID filming keyboard accelerator (F3 key)

Multi Image Display (MID) filming allows you to put the entire displayed image
content of the viewing application into one film frame in the Film Composer.
Note: Since all the displayed images are placed into a single film frame in the
Film Composer, the resolution of an MID film is not as good as the resolu-
tion obtained when the same set of images are filmed using page filming.
Move the mouse pointer onto any image in the viewing application and press the
F3 function key on the keyboard. All the currently displayed images are placed
into the next available film frame in the Film Composer.
The images are displayed in the frame in the same order as they appear in the
viewing application. No changes are made to either the image or user annotation
when the images are loaded as an MID frame in the Film Composer.
Note: MID frames and single image frames can be freely mixed on the same
sheet of film.
Note: If you are using the Viewer, you can also perform MID filming by clicking
left on the (Film MID <F3>) button in the lower left corner of the monitor.

7-8
 Image Works System

Erasing Film Frames

To erase the image in a film frame in the Film Composer, click left on it. The
message "Do you really want to discard this image" pops up. Click left on (Yes)
to erase, or (No) to cancel the erase operation.

Printing Images
To print the content of the Film Composer, click left on the (Print) but-
Print ton in it.

The status line at the bottom of the Film Composer changes to "Printing..." and
the image data are sent to the print queue. The output device will print your film
when your data reach the top of the queue. See "Print Queue" below.
Note: This procedure is only necessary if either Auto printing is not activated in
the Print options window (see "Auto Printing" above) or you want to print
the Film Composer content before its film frames are all full.

Clearing the Film Composer

To erase all the images in the Film Composer, click left on the (Clear)
Clear button in it. The message "Clear the current page" pops up. Click left on
(OK) to proceed with the clear operation, or (Cancel) to abort the clear
operation.
Note: If Auto clear page is activated in the Print options window (see "Auto clear
page" above), the Film Composer is automatically cleared immediately
after printing.

7-9
Image Works System

Film Composer Status Line

The status line at the bottom of the Film Composer keeps you informed of vari-
ous conditions in the filming process (print queue empty, printing, film supply
low, output device not connected, etc.).
Status messages are displayed in different colors depending on their nature:
❏ Green status message indicates a normal condition (e.g. printing),
❏ Yellow status message indicates a warning condition (e.g. film supply
low),
❏ Red status message indicates a blocking condition (e.g. film tray empty or
output device not connected)
Some status conditions are accompanied by an arrow button. If you click left
on this button, a window pops up giving more information about the condi-
tion.

Whenever the output device status changes (e.g. from idle to active, active to
idle, disconnected, etc.) a message is sent to the Messages window. To open
this window, click left on the Message Menu on the main Browser.

Print Series Function

The print series function allows you to set up printing of all or part of the series
currently being viewed with the Viewer or Mini Viewer. To do this, proceed as fol-
lows:
❏ Select the desired series on the Browser and open the Viewer or a Mini
Viewer,
❏ Insure that the Film Composer is activated (it does not have to be visible
on the screen),
Open the Print Series command window by clicking left on the (Film
Film Series Series <F4>) button in the Viewer, pressing the F4 function key on
<F4> the keyboard or typing either print_series or prs in the command
accelerator field in the Viewer or Mini Viewer and pressing the Enter
key,

Format
In the Format box in the Print Series command window, click left on
the (Use Film Composer) button to print using the currently
Used Film Compose
selected format on the Film Composer, which allows you to keep
.Viewer Format
any images already in it, or on the (Viewer Format) button to print
using the current Viewer or Mini Viewer format, which requires the
Film Composer to be initially empty, i.e. cleared,

7-10
 Image Works System

Image Selection
❏ In the Image Selection box, hold and drag left on each end
of the slider to define the range of images in the series to be
1 printed,
1 9

❏ In the Interval box, click left on (Print all images) to print all
Print all images
images in the selected range, (1 / 2) to print every other image
1/2 only, or (1 / 3) to print every third image only,
1/3

❏ Click left on the (Print Series) button to launch printing, or

Print Series the (Close) button to abandon printing and close the Print
Series command window.
Close
Once you have launched printing, the Current Print Jobs field
shows the status of the job(s) launched.

To halt all printing launched from the Print Series command win-
Cancel All
dow, click left on the (Cancel All) button.

Print Series

Format Image Selection

Use Film Composer 1 9
Viewer Format 1 9

Interval Current Print Jobs

Print all images No Current Job
1/2
Cancel All
1/3

Print Series Close

Note: The Film Composer does not have to be called up or visible on the screen
for the Print Series function to operate.

7-11
Image Works System

Putting Text Pages on Film

You can generate an exam or series text page and send it to the Film Composer
for printing. The procedure for doing this is given below:
❏ In the command accelerator field in the Viewer or Mini Viewer, type the
command te for exam text page, or ts for series text page (if only certain
images in the current series were selected on the Browser for viewing, data
from those images only will appear on the text page) and press Enter. The
Text Page command window pops up,
❏ If the text page data occupy more than one page, you can scroll through
them by clicking left on the up- and down-arrows in the Text Page command
window,
❏ To send the text page to the next available frame(s) in the Film Composer,
click left on the (Manual Film) button in the Text Page command window.
Note: Each Film Composer frame containing a text page appears as a stylized
black-text-on-white-background page.
❏ To close the Text Page command window, click left on the (Quit) button.
At this point you can continue filling the frames of the Film Composer as
described above in this chapter; frames containing text are handled in exactly
the same way as any other image-containing frame in the Film Composer.

7-12
CHAPTER 8 Image Works System

Networking
Overview
Networking gives you (or anyone else on your network) the ability to transfer
images between workstations and/or medical imaging systems quickly and eas-
ily.
Note that we use "transfer" as a generic term for moving image data between
stations. When we want to be more specific, we say "transmit" or "push" for the
action taken when you send images from your station to another; and "receive"
or "get" for the act of retrieving images from another station onto your station.
But there are other advantages as well. For instance, you can routinely off-load
display, manipulation, and filming activities from a medical imaging system to the
Image Works workstation. This improves the productivity of the medical imaging
system. It also eliminates the problems associated with the hand-delivery of
films to the people who need to work with them.
From the Image Works workstation, you can send examinations, series, and
images to, or retrieve them from, other systems on your network. The worksta-
tions and imaging systems networked to the Image Works workstation are
termed "hosts." Currently, the Image Works workstation may be networked to CT
Advantage-level, Pace, CT9800, Sytec 6000/8000 (Prospeed) and 3000i con-
soles/products as well as MR Advantage-level and Vectra consoles/products via
the GE Medical Systems Network. For more information regarding networking
your Image Works products with other medical imaging systems or products,
contact your local GE Field Service Engineer.
Caution: The workstation will not always inform you if the system has failed to
completely transfer all the data requested. Therefore, checking with
the source or destination host is recommended to verify that the
data transferred successfully. (The system will tell you if the net-
work is active however.)
Note: The host names stored in the configuration file of your workstation's
computer software are entered by your GE Service Engineer. Con-
sult your GE Service Engineer for your list of hosts and their
assigned code names.
While the Image Works workstation allows you to remove host names from net-
work memory, use caution when doing so. If you inadvertently delete a host
name from your configuration file, only your GE Service Engineer should restore
it.
Note: Image Works handles Advantage and DICOM image formats. However,
there are differences in annotation and Browser list data presentation
between these formats.

8-1
Image Works System

Information Concerning Connection with DICOM

Equipment
When a network connection between an Image Works workstation and DICOM
equipment is being configured, the DICOM equipment has to be configured with
an "application name" and "DICOM port number". Use the Image Works work-
station's host name as this "application name" for the DICOM port number.
[Ping DICOM host] enables you to confirm the physical link
Network
between the Image Works system and DICOM equipment.
Receive When the physical link is complete, the pop-up message
appears. Click on [OK].
Ping DICOM host

Send examination Attention

Send series ! xxxx Dicom Host Alive

Send image
OK
Select remote host

When the physical link is not complete, the pop-up menu

appears as follows. Then, click on [OK] to once exit the proce-
dure and reconfirm the link.

Attention

! Connection error : xxxx

OK

8-2
 Image Works System

Selecting a Remote Host

Click left on the Network menu in the Browser menu bar and
Receive
select [Select remote host] or [Selected host: xxxx] from the list
Ping DICOM host that pops up:

Send examination [Select remote host]

Send series If this is the first time that you are selecting a remote host during
this session (since starting up the main application), [Select
Send image remote host] appears as the bottom option in the list.
Select remote host
[Selected host: xxxx]
On the other hand, if you have already selected a remote host dur-
ing this session, this option will read [Selected host: xxxx].
Note: If "xxxx" is the name of the host that you want, it is already selected,
so you can skip the rest of this section and continue with your net-
working operation.
Regardless of the option name, selecting it does the same thing: open the
Remote host selection window.

Remote host selection

Remote host list
MCT_OC0
NMR1_OC0
ZNR2_IC0

Remote host selection ZNR2_IC0

Ok Add Remove Update Cancel

This window provides a list of the workstations and medical imaging systems
currently networked to your workstation.
Click left on the name of the host in the list with which you want to transfer data.
The host name becomes highlighted.
Click left on (Ok) to have the current selection become the selected host. The
Remote host selection window closes.
Or, click left on (Cancel) to cancel your selection and exit the Remote host
selection window without making any change to the current selection.

8-3
Image Works System

Transmitting Data to a Remote Host (Network Push)

Note: This function is not supported for a Signa 4.x, CT9800 or any GE-
YMS system as remote host.
To send exams, series, or images from your Image Works workstation to another
workstation or imaging system on your network, first check that there is enough
room on the remote system's disk to accommodate the images being sent. See
the user documentation provided by the manufacturer of the remote system.
On the Browser, select the desired exam(s), series or image(s) to be sent.
Select the remote host to which you want to send data.
Click left on the Network menu on the Browser, then click left on
Receive
[Send examination], [Send series], or [Send image] depending
Ping DICOM host on what you want to send.

Send examination

Send series

Send image

Select remote host

A confirmation window pops up and prompts you "Push selected examination(s)/

series/image(s)?". Click left on (OK) to proceed, or click left on (Cancel) to can-
cel the push operation.
The items selected on the Browser for sending are transferred from your Image
Works workstation to the remote host.
Note: During the sending process, the Network menu title on the main
Browser becomes Network (active). It returns to Network when the
sending process is finished.
Note: If the remote host cannot be reached, a pop-up message will report
that fact. Verify the remote host is operational, check the network
cabling, and call your GE Field Engineer if the problem persists.

8-4
 Image Works System

To verify that your items have arrived on the remote host, you can look for them
on the Remote Browser as follows:
❏ Wait for the sending process to finish, which is indicated by the Network
(active) menu title on the main Browser returning to Network (or check the
network queue; see "Network Queues" below),
❏ If the Remote Browser is not already open, click left on the Network
menu in the main Browser menu bar and select [Query remote host] from
the list that pops up,

Application
❏ Update the display of items listed in the Remote Browser lists
by clicking left on the Application menu in its menu bar and
Refresh Lists selecting [Refresh Lists] from the list that pops up (note that the
time of last refresh shown at the bottom of the Remote Browser
Quit updates accordingly),
❏ Now examine the Remote Browser for the items you sent,
❏ When you've finished using the Remote Browser, close it by
Application
clicking left on the Application menu in its menu bar and select-
Refresh Lists ing [Quit] from the list that pops up.

Quit

8-5
Image Works System

Removing Host Names from the Workstation's

Configuration File
Note: Use care when removing host names via the workstation's Remote
host selection window. If you inadvertently delete a host name, your
GE Field Service Engineer can restore it.
Click left on the Network menu in the Browser menu bar and select [Select
remote host] or [Selected host: xxxx] from the list that pops up.
The Remote host selection window pops up.

Remote host selection

Remote host list
MCT_OC0
NMR1_OC0
ZNR2_IC0

Remote host selection ZNR2_IC0

Ok Add Remove Update Cancel

Click left on the name of the host that you want to remove. The host name
becomes highlighted.
Click left on (Remove) at the bottom of the Remote host selection window. A
confirmation message pops up.
Click left on (OK) to remove the host from the configuration file. The host is
removed from the Remote host selection window.
Or, click left on (Cancel) to cancel the remove operation.
(Note that the Remote host selection window contains buttons labeled (Add)
and (Update). These buttons are for use by your GE Field Service Engineer to
add new nodes to the host list or update existing ones.)

8-6
 Image Works System

Network Queues
It is not necessary to wait for one push or get operation to be completed before
you set up the next one, because when you set up a push operation, it is placed
into the push queue, and when you set up a get operation, it is placed into the
pull queue.
These two queues are "waiting lines" for requests to be carried out.
To view the requests waiting in the queues, click left on the Queue
Queue
menu on the main Browser, then select [Network] from the list that
Archive pops up.
A window pops up for each of the three network servers currently per-
Network
forming transfers: "Advantage" and "Dicom". No window pops up for
network servers that are not performing transfers.

Note: The network server used for a transfer depends on the type of
equipment (i.e. the type of remote host) with which the transfer is
being done, and was configured by your GE Field Service Engineer
when your Advantage Windows workstation was installed.
If none of the network servers is performing transfers, the message All Network
queues are empty pops up. Click left on (OK) to remove the message.
Note: When images are being "pushed" from one Advantage Windows
station to another Advantage Windows station, the receiving station
shows the job on its SdC NET pull queue.

8-7
Image Works System

Refresh To subsequently update the status of the two queues in a particular

window, click left on its respective (Refresh) button.
Each line in a queue represents a job to be executed. Jobs consisting of an
entire exam take on the form "exam no.", those consisting of an entire series
take on the form "exam no./series no." and those consisting of a single image
take on the form "exam no./series no./image no." followed, in all cases, by the
remote server name in parentheses [(ic_imo) in the figure on the previous page
is an example] and one of the following three status messages:
❏ (Active) for the job currently being executed,
❏ (Pending) for jobs waiting to be executed,
❏ (Paused) for temporarily paused jobs.
Note: If an entire exam or series is being pushed or pulled, only one job is
executed for the entire exam or series, respectively. But, if only cer-
tain series in an exam, or certain images in a series are being
pushed or pulled, a separate job is executed for each series or
image, respectively.
Note: Limiting the number of jobs in any one queue to 100 is recom-
mended.

Pause To temporarily suspend execution of requests in a queue, click left on

its respective (Pause) button. All pending jobs now have the
(Paused) status, but the active job continues until completion.
Note: If you pause the Pull queue and then initiate more network get jobs,
those jobs are placed in the queue in front of the paused jobs and
executed immediately.

Resume To continue execution of requests in a paused queue, click left on its

respective (Resume) button. All jobs return to their original status
and execution continues.

Clear To permanently remove a job, select it by clicking left on it, then click
left on the queue's respective (Clear) button. A confirmation window
pops up. Click left on (OK) to proceed, or click left on (Cancel) to
abort the clear operation.
Quit To close a queue window, click left on its (Quit) button.

8-8
CHAPTER 9 Image Works System

Archiving
Overview
The archiving option allows you to save and/or restore large amounts of image
data on optical disks (ODs). ODs can either be of the Write Once Read Many
(WORM) or Magnetic Optical Disk (MOD) type.
Note: With the archiving option, data can only be restored from a WORM disk.
This option can be used to import data from ODs coming from a large part of the
GEMS MR/CT installed base, as well as from GE-YMS systems (equipped with
Pioneer brand 5-1/4" OD drives only).
The archiving option provides full read/write inter-operability with Signa 5.X, CT
Advantage or CT/MR Independent Console systems.

Composition of Archiving
An Image Works workstation equipped with the archiving has a 5-1/4" OD drive
connected to it. Also, the archiving software is installed on the workstation,
which is indicated by the presence of Archive on the Browser menu bar.
Note: We strongly recommend reading the user documentation provided by the
manufacturer of the OD drive prior to using it.

9-1
Image Works System

Restoring Images from OD

Before attempting to restore images from OD to the Image Works workstation,
be sure that your are in compliance with the restrictions given in the following
section:

Restrictions for Restoring Images from OD

The ability to restore images from OD to the Image Works workstation depends
on two factors:
❏ OD type,
❏ Image format.
Details of OD types and image formats supported by the Image Works worksta-
tion restore functions are given in the next two sections:

OD Types Readable by Image Works

The following OD types can be used to restore image data from the OD to the
Image Works workstation:
❏ 5-1/4" MOD labelled on any Genesis OC/IC,
❏ 5-1/4" WORM disk written on any Genesis OC/IC,
❏ 5-1/4" MOD labelled with Image Works,
❏ 5-1/4" MOD labelled on a GE-YMS system equipped with a Pioneer brand
OD drive.

Image Types Readable from OD to Image Works

Images of the following types can be restored from OD to the Image Works
workstation:
❏ All Genesis images,
❏ GE-YMS,
❏ Image Works

9-2
 Image Works System

Procedure: Restoring Images from OD

After ensuring that the OD and images comply with the restrictions given above,
proceed as given below to restore images.
First, check that there is enough room on your image disk(s) to accommodate
the images being restored. Information about disk space is displayed at the bot-
tom of the Browser. Bear in mind however that this information is approximate
and that it may take a few minutes to update following addition or deletion of
images.
Next, determine which side of the MOD contains the images you want to restore
(side A or side B) and insert it in the OD drive accordingly (see user documenta-
tion provided by the manufacturer of the OD drive).
Click left on the Archive menu on the Browser, then click left on
Archive [Restore].
Restore An Archive Browser appears on the screen. It displays the exams,
Save series and images on the OD side in the OD drive, in much the same
examination way as the main Browser displays the available items on the Image
Save series Works workstation.
Save image Note: The Archive menu appears as Archive (active) if other items
are already in the process of being saved or restored from the
Label
current OD side in the drive, but the menu is still available.
Detach Your request will simply be saved in a queue, and executed as
Selected archive soon as the other requests ahead of it are completed. See
device "Archive Queues" below.

Selection Remove Sort Network Archive Queue Message

Examinations : Exam #47, May 05 92, SMITH, JON
Exam Name Date Description Mod Fmt A Ser Type Imgs Description Mod
3145 J.Herman Jan 08 98 CT 1 PROSP 18 CT
3512 B.Fox Dec 23 97 CT

2 examinations one series

Series
Img Img Ctr Thck/SP Gntry Img Ctr Img Ctr SFOV DFOV Res
S–I (mm) (mm) (deg) R–L A–P (cm) (cm) Matrix Midscn Archive

1 S 50.0 10.0/ +0.0 R5.0 A0.0 Lung 512 No

43
2 S 45.0 10.0/ +0.0 R5.0 A0.0 Lung 512 No
43
3 S 40.0 10.0/ +0.0 R5.0 A0.0 Lung 512 No
43
S 35.0 R5.0 A0.0 Lung 512 No
4 10.0/ +0.0 43
R5.0 A0.0 Lung 512 No
5 S 30.0 10.0/ +0.0 43
R5.0 A0.0 Lung 512
6 S 25.5 10.0/ +0.0 43 No

60 images
Image Database on host ic_imo (09:52AM)

9-3
Image Works System

Now you must select the examinations, series or images that you want to restore
from the disk. To help you do this, you can use the Sort menu and Selection
menu on the Archive Browser as described below:
To select all items in a list on the Archive Browser, click left on the
Selection Selection menu on the Archive Browser, then click left on [Select
all examinations], [Select all series], or [Select all images].
Select all
examinations All the items in the chosen list on the Archive Browser are high-
Select all lighted.
series
Select all
images

To sort items on the Archive Browser according to certain criteria,

Sort click left on the Sort menu on the Archive Browser, then click left
on [Sort examinations by patient name], [Sort examinations by
Sort examinations trigger], [Sort images by location], or [Sort images by image
by patient name
number].
Sort images by
trigger The items will reappear in the list, sorted according to your choice.
Sort images by
location
Sort images by
image number

Once all the items that you want to restore are selected on the
Restore Archive browser, click left on the Restore menu on the Archive
Browser, then click left on [Restore examination], [Restore
Restore
examination series], or [Restore image] depending on what you want to
restore.
Restore series
The items selected on the Archive Browser for restoring are read
Restore image from the OD in the OD drive to the Image Works workstation.

Note: The word (updating) appears to the right of the Message menu title in
the main Browser menu bar whenever the Image Works data base is
being updated. (The word (updating) does not disappear immediately
after completion of data base updating.)
Note: All restored images appear on the main Browser images list with "Yes" in
the Archive column.

9-4
 Image Works System

To eject the MOD from the OD drive, it must first be "detached" via the [Detach]
option in the Archive menu on the main Browser. See "Ejecting a Disk from the
OD Drive" below for details.
To quit the Archive Browser, click left on the Application menu on
Application
the Archive Browser, then click left on [Quit].
Quit

Note: The Archive Browser is quit automatically when the [Detach] option in the
Archive menu of the main Browser is executed.
See "Ejecting a Disk from the OD Drive" below.

9-5
Image Works System

Labelling a MOD
Brand new MODs must be labelled before they can be used with the Image
Works workstation. Also, you can use this procedure to re-label a MOD which
contains data that you don't need anymore.
Caution: When a MOD is re-labelled, all existing data on it are permanently
lost.
Caution: When a MOD is labelled on an Image Works workstation, it can only
be read and written with a Signa 5.X, CT Advantage or CT/MR Inde-
pendent Console with Genesis software version 4.1 or later. A MOD
labelled on an Image Works workstation CANNOT be read on such
systems with Genesis software versions earlier than 4.1.
To label a MOD, first decide which side you want to label (side A or side B) and
insert it in the OD drive accordingly (see user documentation provided by the
manufacturer of the OD drive).
Click left on the Archive menu on the Browser, then click left on
Archive
[Label]. A window entitled "Format Window" pops up.
Restore

Save
examination

Save series

Save image

Label

Detach

Selected archive
device

Format Window

Device PIONEER

Device Identifier 01

Comments MR images Side A

LABEL CANCEL

Note: The Device field shows the OD drive type. This is set up during installa-
tion, and requires no intervention from the user.

9-6
 Image Works System

Click left on the Device Identifier field and enter a value of your choice. This
value identifies the MOD side being labelled.
Click left on the Comments field and enter your comments from the keyboard.
Typically, information about the type of images, MOD side, or other pertinent
information of your choice are used. (The text in this field will appear at the bot-
tom of the Archive Browser during selection of images to be restored from this
MOD side.) The maximum number of characters allowed in this field is 160.
Click left on the (LABEL) button to begin the labelling process. Click left on
(Yes) to proceed with labelling or (No) to abandon.
To abort the labelling process, click left on the (CANCEL) button in the Format
Window.

9-7
Image Works System

Saving Images on MOD

Before attempting to save images on MOD with the Image Works workstation,
be sure that your are in compliance with the restrictions given in the following
section:

Restrictions for Saving Images on MOD

The ability to save images on MOD with the Image Works workstation depends
on two factors:
❏ MOD type,
❏ Initial source of images.
Details of MOD types and image sources supported by the Image Works work-
station save functions are given in the next two sections:

MOD Types Writeable by Image Works

The following MOD types can be used to save image data from the Image Works
workstation:
❏ 5-1/4" MOD labelled on a Signa 5.X, CT Advantage or CT/MR Indepen-
dent Console,
❏ 5-1/4" MOD labelled with Image Works.

Image Types Writeable to MOD by Image Works

Images from the following sources, once on the Image Works workstation, can
be saved on MOD:
❏ Signa 4.x,
❏ Signa 5.X, CT Advantage or CT/MR Independent Console,
❏ CT9800,
❏ GE-YMS CT scanners with Advantage Net,
❏ Image Works 1.2 (except saved 3D objects as indicated by 3D OBJ in the
Type column of the series list on the main Browser).

9-8
 Image Works System

Procedure: Saving Images on MOD

After ensuring that the MOD and images comply with the restrictions given
above, proceed as given below to save images.
Note: If your MOD is brand new, or if you want to re-label it (and consequently
lose all the data on it permanently), follow the instructions for labelling
given in "Labelling a MOD" below before proceeding.
First, decide which side of the MOD you want to save images on (side A or side
B) and insert it in the OD drive accordingly (see user documentation provided by
the manufacturer of the OD drive).
On the Browser, select the desired exam(s), series or image(s) to be saved.
Finally, click left on the Archive menu on the Browser, then click left
Archive
on [Save examination], [Save series], or [Save image] depending
Restore
on what you want to save.
The items selected on the Browser for saving are written onto the
Save
examination MOD in the OD drive.
Save series Note: The Archive menu appears as Archive (active) if other items
are already in the process of being saved or restored, but the
Save image menu is still available. Your request will simply be saved in a
queue, and executed as soon as the other requests ahead of it
Label
are completed. See "Archive Queues" below.
Detach Note: Items being saved cannot be deleted from the Image Works
Selected archive
workstation until their saving is completed. This is seen by the
device gray Remove menu title on the Browser whenever items being
saved are selected.
Once an image has been saved, the word "Yes" appears in the Archive column
of its respective entry in the images list of the main Browser. When archived
images are networked directly to another Image Works station, they also appear
as archived on that station's Browser. This allows all Image Works users to keep
track of which images have and have not been saved on MOD.
However, when images are networked from a non-Image Works workstation, the
word "No" appears in the Archive column of the Browser of the receiving Image
Works workstation, regardless of whether or not they have been archived previ-
ously.
To eject the MOD from the OD drive, it must first be "detached" via the [Detach]
option in the Archive menu on the Browser. See "Ejecting a Disk from the OD
Drive" below for details.

9-9
Image Works System

Ejecting a Disk from the OD Drive

Once an OD has been used for save or restore operations, it cannot be ejected
from the OD drive unless you first perform a detach operation as described
below:
First, click left on the Archive menu on the main Browser, then click
Archive left on [Detach]. A confirmation window pops up and asks you Detach
Restore
media? (If images are in the process of being restored from the OD,
the message is Detach media: restore images process will be
Save stopped!.) Click left on (OK) to proceed, or click left on (Cancel) to
examination
abort the detach operation.
Save series
You can now eject the OD from the OD drive by pressing the eject but-
Save image
ton on its front panel. See the OD drive manufacturer's documentation
Label for details.

Detach If the [Detach] option in the Archive menu appears in gray lettering,
this means that the OD is already detached, and you can proceed
Selected archive
device directly with ejecting it from the OD drive.

Archive Queues
It is not necessary to wait for one save or restore request to be carried out
before you set up the next one, because when you set up a save request, it is
sent to the Archive Save queue, and when you set up a restore request, it is sent
to the Archive Restore queue.
These two queues are "waiting lines" for requests to be carried out.
To view the requests waiting in the queues, click left on the Queue
Queue
menu on the main Browser, then click left on [Archive].
Archive An "Archive" window pops up, displaying both the Archive Save
Queue and the Archive Restore Queue.
Network

9-10
 Image Works System

Archive

Archive Save

65/3/1 (Pending)
65/3/2 (Pending)
65/3/3 (Pending)
2899 (Pending)
273/3 (Pending)

Pause Resume Clear

Archive Restore

238/7 (Active)
344/10/9 (Pending)

Pause Resume Clear

Refresh Quit

The content of the Archive Save and Archive Restore Queues shown in this win-
dow is up to date at the moment when it pops up. To subsequently update the
status of the two queues, click left on the (Refresh) button.
Note: Jobs in the Restore queue have priority over those in the Save queue.
Save jobs are only serviced when the Restore queue is empty.

9-11
Image Works System

Each line in a queue represents a job to be executed. Jobs consisting of an

entire exam take on the form "exam no.", those consisting of an entire series
take on the form "exam no./series no." and those consisting of a single image
take on the form "exam no./series no./image no." followed, in all cases, by one of
the following three status messages:
❏ (Active) for the job currently being executed,
❏ (Pending) for jobs waiting to be executed,
❏ (Paused) for temporarily paused jobs.
Note: If an entire exam or series is being saved or restored, only one job is exe-
cuted for the entire exam or series, respectively. But, if only certain series
in an exam, or certain images in a series are being saved or restored, a
separate job is executed for each series or image, respectively.
Note: Limiting the number of jobs in any one queue to 100 is recommended.
To temporarily suspend execution of requests in a queue, click left on its respec-
tive (Pause) button. All jobs now have the (Paused) status. To continue execu-
tion of requests in a paused queue, click left on its respective (Resume) button.
All jobs return to their original status and execution continues.
To permanently remove a job, select it by clicking left on it, then click left on the
queue's respective (Clear) button. A confirmation window pops up. Click left on
(OK) to proceed, or click left on (Cancel) to abort the clear operation.
To close the queue window, click left on the (Quit) button.
Note: The Archive Restore Queue is deleted on system shutdown or following
an Archive - [Detach] operation (see "Ejecting a Disk from the OD Drive"
above), whereas the Archive Save queue is maintained under either of
these conditions.

9-12
CHAPTER 10 Image Works System

Image Combination
Overview
The image combination function allows you to perform various operations on
one or two sets of images selected on the Browser and place the resulting
images in a new series.
To activate this function, click left on the (Add / Sub) button in the Browser. The
Image Combination command window pops up.

Image Combination

Either one or two sets of images must be selected before any image combina-
tion functions can be performed. See "Set selection" below.
If only one set is selected, each operation performed produces ONE resulting
image.
If two sets are selected, images in the two sets are paired according to physical
location in the patient's body. Unpaired images in either set are ignored. Each
operation performed produces ONE resulting image PER PAIR. See the appro-
priate sections below for details about each type of operation.
The images resulting from each operation are generated in the exam defined by
the left-hand (Select Set) button.

10-1
Image Works System

Save Series
Resulting images are assigned a series number and starting image number
given by the Current Save Series field in the upper right corner of the Image
Combination command window. By default, images resulting from subsequent
operations are added onto the end of the same series.
To place images resulting from subsequent operations into a new series, click
left on the (New Save Series) button. This increments the save series number
by one and displays the Save Series and Save Image fields in the upper right
corner of the Image Combination command window. You can then either accept
the numbers in these two fields or click left on either of them, use the Back
Space key to delete them, and enter the desired numbers from the keyboard.
The series resulting from image combination operations are distinguished from
other series in the Browser via one of the following two indications in the
Browser series list Type column:
❏ "Proc" appears in the Browser series list Type column if the images in the
series are the result of processing pairs of images having identical locations
in the patient's body,
❏ "Comb" appears in the Browser series list Type column if the images in
the
series are the result of a combination of images having different locations in the
patient's body.
Since "Proc" series contain images resulting from processing pairs of images
having identical locations in the patient's body, such series can be used like any
other series of acquisition images, i.e. geometrical measurements, reformatting,
3D reconstruction, etc.
Caution: Since "Comb" series contain images resulting from a combination of
images having different locations in the patient's body, the absolute
anatomical coordinates accompanying these series (shown both in
the Browser and on the displayed images) are not accurate. Only
relative geometrical measurements (i.e. distance, angle or area)
made within a resulting image are accurate.
Note: Even though "Proc" and "Comb" series can, in certain cases, be "pushed"
via the network, or transferred via archive MOD disk, to other console
types, they can be displayed, analyzed and manipulated on Image Works
workstations ONLY.

10-2
 Image Works System

Set Selection
With the Image Combination command window open as described above, select
the image(s) on the Browser that you wish to define as Set 1. (The images do
not have to be consecutive in the Browser list.)
Now click left on the left-hand (Select Set) button in the Image Com-
bination command window. Data about the selected images replace
the title of the button, and appear in the form exam no./series no./
image no(s).
If applicable, repeat the procedure for the second set, but use the right-hand
(Select Set) button.
To change a set selection, repeat the procedure with a different selection on the
Browser and the set selection updates automatically.
To clear a set selection, click left on the (Clear Selection) button in
the Image Combination command window.The titles of the (Select
Set) buttons are restored.

10-3
Image Works System

Maximum Pixel Value Extraction

Maximum pixel value extraction is an operation that consists of finding maximum
image intensity values pixel by pixel. Two types of maximum pixel value extrac-
tion are possible:
❏ Extraction on one set alone,
❏ Extraction between two sets.

Maximum pixel value extraction on one set alone

This type of extraction finds the maximum pixel intensity, pixel by pixel, from
among all the images selected via the left-hand (Select Set) button.
The resulting image is placed in a series bearing the series number and image
number given by the Current Save Series field in the upper right corner of the
Image Combination command window.
To perform this operation, proceed as follows:
❏ Insure that both (Select Set) buttons are cleared. If necessary, click left
on the (Clear Selection) button to clear all selections,
❏ Define the first set as described in "Set selection" above,
❏ Check that the series number and image number given by the Current
Save Series field are as you like; if necessary, change them as described
above,
Click left on the (MAX) button to select the maximum pixel value
extraction mode,
Click left on the (=) button to perform the operation and generate the
new image.

The new image is available shortly in the Browser for use by viewing applica-
tions.

10-4
 Image Works System

Maximum pixel value extraction between two sets

This type of extraction finds the maximum pixel intensity values, pixel by pixel, of
all pairs of images between the two selected sets corresponding to the same
physical location in the patient's body.
The resulting images are placed in a series bearing the series number and
image number given by the Current Save Series field in the upper right corner of
the Image Combination command window.
To perform this operation, proceed as follows:
❏ Define the two sets as described in "Set selection" above,
❏ Check that the series number and image number given by the Current
Save Series field are as you like; if necessary, change them as described
above,
Click left on the (MAX) button to select the image addition mode,

Click left on the (=) button to perform the operation and generate the
new image.
The new images are available shortly in the Browser for use by viewing applica-
tions.

10-5
Image Works System

Image Addition
Image addition is an operation that consists of adding image intensity values
pixel by pixel. Two types of image addition are possible:
❏ Image addition on one set alone,
❏ Image addition between two sets.

Image addition on one set alone

SUM OF ALL IMAGES SELECTED VIA THE LEFT-HAND (Select Set) BUT-
TON ONLY
This type of addition adds the image intensity values, pixel by pixel, of all the
images selected via the left-hand (Select Set) button. For normalization pur-
poses, each resulting pixel value is divided by the number of images in the
selected set.
The resulting image is placed in a series bearing the series number and image
number given by the Current Save Series field in the upper right corner of the
Image Combination command window.
To perform this operation, proceed as follows:
❏ Ensure that both (Select Set) buttons are cleared. If necessary, click left
on the (Clear Selection) button to clear all selections,
❏ Define the first set as described in "Set selection" above,
❏ Check that the series number and image number given by the Current
Save Series field are as you like; if necessary, change them as described
above,
Click left on the (+) button to select the image addition mode,

Click left on the (=) button to perform the operation and generate the
new image.

The new image is available shortly in the Browser for use by viewing applica-
tions.

10-6
 Image Works System

Image addition between two sets

[ FIRST SET + SECOND SET ] IN PAIRS
This type of addition adds the image intensity values, pixel by pixel, of all pairs of
images between the two selected sets corresponding to the same physical loca-
tion in the patient's body. By default, equal weighting is applied to the two pixels
in each pair, but you can change the weighting via the Ratio slider. In all cases,
for normalization purposes, each resulting pixel value is divided by two.
The resulting images are placed in a series bearing the series number and
image number given by the Current Save Series field in the upper right corner of
the Image Combination command window.
To perform this operation, proceed as follows:
❏ Define the two sets as described in "Set selection" above,
❏ Check that the series number and image number given by the Current
Save Series field are as you like; if necessary, change them as described
above,
Click left on the (+) button to select the image addition mode,
If necessary, drag left on the Ratio slider to obtain other than the
default equal weighting of pixel values between the two series; drag
leftward to increase the weighting on the images selected on the left-
hand (Select Set) button or drag rightward to increase the weighting
on the images selected on right-hand (Select Set) button (if you
move the Ratio slider to one extreme, the images selected on the
opposite (Select Set) button have zero weight in the addition opera-
tion),
Click left on the (=) button to perform the operation and generate the
new images.

The new images are available shortly in the Browser for use by viewing applica-
tions.

10-7
Image Works System

Image Subtraction
Image subtraction is an operation that consists of subtracting image intensity
values pixel by pixel. Two types of image subtraction are possible:
❏ Image subtraction between two images in one set alone,
❏ Image subtraction between two sets.
A selectable "Accept Negative Pixels" function is available for either type of sub-
traction. This function, when enabled, allows negative pixel values in the result-
ing image(s). (If this function is not enabled, all negative pixel values are set to
zero.)

Image subtraction between two images in one set alone

[ FIRST IMAGE - SECOND IMAGE IN SELECTED SET ]
This type of subtraction subtracts the image intensity values, pixel by pixel, of
the second image selected in the Browser defined via the left-hand (Select Set)
button from the first image selected in the Browser defined via the same (Select
Set) button.
The resulting image is placed in a series bearing the series number and image
number given by the Current Save Series field in the upper right corner of the
Image Combination command window.
To perform this operation, proceed as follows:
❏ Ensure that both (Select Set) buttons are cleared. If necessary, click left
on the (Clear Selection) button to clear all selections,
❏ Define the first set as described in "Set selection" above (be sure that only
TWO images are selected on the Browser),
❏ Check that the series number and image number given by the Current
Save Series field are as you like; if necessary, change them as described in
"General information" above,
❏ Click left on the (Accept Negative Pixels) button if you want the resulting
image to contain negative values,
Click left on the (-) button to select the image subtraction mode,

Click left on the (=) button to perform the operation and generate the
new images.

The new image is available shortly in the Browser for use by viewing applica-
tions.

10-8
 Image Works System

Image subtraction between two sets

[ FIRST SET - SECOND SET ] IN PAIRS
This type of subtraction subtracts the image intensity values, pixel by pixel, of all
pairs of images between the two selected sets corresponding to the same phys-
ical location in the patient's body. For each pair, the pixel values from the set
defined by the right-hand (Select Set) button are subtracted from the pixel val-
ues from the set defined by the left-hand (Select Set) button. By default, equal
weighting is applied to the two pixels in each pair, but you can change the
weighting via the Ratio slider.
The resulting images are placed in a series bearing the series number and
image number given by the Current Save Series field in the upper right corner of
the Image Combination command window.
To perform this operation, proceed as follows:
❏ Define the two sets as described in "Set selection" above,
❏ Check that the series number and image number given by the Current
Save Series field are as you like; if necessary, change them as described
above,
Click left on the (-) button to select the image subtraction mode,

❏ If necessary, drag left on the Ratio slider to obtain other than the default
equal weighting of pixel values between the two series; drag leftward to
increase the weighting on the images selected on the left-hand (Select Set)
button or drag rightward to increase the weighting on the images selected on
right-hand (Select Set) button (if you move the Ratio slider to one extreme,
the images selected on the opposite (Select Set) button have zero weight in
the addition operation),
Click left on the (=) button to perform the operation and generate the
new images.
The new images are available shortly in the Browser for use by viewing applica-
tions.

10-9
Image Works System

Minimum Pixel Value Extraction

Minimum pixel value extraction is an operation that consists of finding minimum
image intensity values pixel by pixel. Two types of minimum pixel value extrac-
tion are possible:
❏ Extraction on one set alone,
❏ Extraction between two sets.
The procedure for performing these extractions is identical to that
given above for maximum pixel value extraction. The only difference
is that you must click left on the (MIN) button instead of on the
(MAX) button.

Binding Series
You can create a new series which consists of copies of selected images from
one or more existing series.
Save State information is not maintained in the new series generated in
this way.
To perform this operation, proceed as follows:
❏ Ensure that both (Select Set) buttons are cleared. If necessary, click left
on the (Clear Selection) button to clear all selections,
❏ Define the first set as described in "Set selection" above,
❏ Check that the series number and image number given by the Current
Save Series field are as you like; if necessary, change them as described
above,
Click left on the (BIND) button to select the series binding mode,

Click left on the (=) button to perform the operation and generate the
new images.
The new images are available shortly in the Browser for use by viewing applica-
tions.
You can concatenate copies of images from other series to the end of this new
series by following the same procedure, being sure to set up the Save Series
and Save Image field values correctly prior to clicking left on the (=) button.

Closing the Image Combination Command Window

To close the Image Combination command window, click left on the
(QUIT) button.

10-10
 A I
Annotation level, 4-8 Image Combination, 10-1
Image Addition, 10-6
Archiving, 9-1
Image Subtraction, 10-8
Set Selection, 10-3
C
Changing Image Orientation, 4-4 M
Magnifying Displayed Images, 4-4
D
Drawing Graphics, 6-2 N
Angle, 6-3
Networking, 8-1
Ellipse Area, 6-6
DICOM, 8-2
Free Draw Area, 6-7
Rectangle Area, 6-4
Report Cursor, 6-2 S
Smooth Curve Area, 6-5
Straight Line Distance, 6-3 Selecting and Sorting Images, 3-1
Format, 3-7
scroll bars, 3-2
F Selection menu, 3-4
Film Composer, 7-1 Setting Window Width/Levels, 4-2
Auto clear page, 7-6
Auto printing, 7-6
Expose order, 7-6 U
Format, 7-4
Greyscale, 7-5 User Annotation, 5-2
Icon labels, 7-6 User Prefs, 4-7
Number of copies, 7-6
Slide Format, 7-5
2271012-100.book Page 1 Thursday, May 8, 2003 11:30 AM

Volume Analysis

REGULATORY REQUIREMENTS
This product complies with the regulatory requirements of the following:

• Council Directive 93/42/EEC concerning medical devices: the label

affixed to the product testifies compliance to the Directive.

European registered place of business:
GE Medical Systems Europe
Quality Assurance Manager
BP 34
F 78533 BUC CEDEX France
Tel: +33 (0)1 30 70 40 40
• Medical Device Good Manufacturing Practice Manual issued by the FDA
(Food and Drug Administration, Department of Health, USA).
• Underwriters’ Laboratories, Inc. (UL), an independent testing laboratory.
• Canadian Standards Association (CSA).
• International Electrotechnical Commission (IEC), international standards
organization, when applicable.
• USA/HHS:

United States Federal law restricts this product to use

CAUTION by or on the order of a physician.

• General Electric Medical Systems is ISO 9001 and EN 46001 certified.

• The original document was written in English.

1
2271012-100.book Page 2 Thursday, May 8, 2003 11:30 AM

Volume Analysis

LEGAL NOTICE

Copyright 1994-2001 World Wide Web Consortium, (Massachu

setts Institute of Technology, Institut National de Recherche en Informatique

et en Automatique, Keio University).

All Rights Reserved. http://www.w3.org/Consortium/Legal/

THIS SOFTWARE AND DOCUMENTATION IS PROVIDED “AS IS,” AND COPY-

RIGHT HOLDERS MAKE NO REPRESENTATIONS OR WARRANTIES,
EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO, WARRANTIES OF
MERCHANTABILITY OR FITNESS FOR ANY PARTICULAR PURPOSE OR THAT
THE USE OF THE SOFTWARE OR DOCUMENTATION WILL NOT INFRINGE
ANY THIRD PARTY PATENTS, COPYRIGHTS, TRADEMARKS OR OTHER
RIGHTS.
COPYRIGHT HOLDERS WILL NOT BE LIABLE FOR ANY DIRECT, INDIRECT,
SPECIAL OR CONSEQUENTIAL DAMAGES ARISING OUT OF ANY USE OF
THE SOFTWARE OR DOCUMENTATION.

Copies of the LibXML library can be found at the following:

http://www.w3.org/status
http://www.xmlsoft.org

2
2271012-100.book Page 3 Thursday, May 8, 2003 11:30 AM

Table of Content Volume

Analysis (optional)

Chapter 1 - GENERAL
• SYSTEM REQUIREMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-2
• HOW TO USE THIS DOCUMENT . . . . . . . . . . . . . . . . . . . . . . . . 1-3
• CONVENTIONS FOR THIS MANUAL . . . . . . . . . . . . . . . . . . . . . 1-4
• PIXELS AND VOXELS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-5
Chapter 2 - SAFETY
• Intented Use . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-2
• Measurements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-2
• Building the 3D Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-3
• Annotations on Filmed or Saved Images . . . . . . . . . . . . . . . . . . . 2-7
Chapter 3 - INTRODUCTION
• OVERVIEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-2
• SUMMARY OF OPERATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-3
Image Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-3
Building the 3D Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-4
Display and Controls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-4
Processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-5
Reformatting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-6
3D Imaging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-7
Navigator (option) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-8
Volume Rendering (option) . . . . . . . . . . . . . . . . . . . . . . . . . 3-8
Advanced Vessel Analysis (option) . . . . . . . . . . . . . . . . . . . 3-9
Annotations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-9
Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-10
Recording . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-10
Chapter 4 - IMAGE REQUIREMENTS
• IMAGE REQUIREMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-2
2271012-100.book Page 4 Thursday, May 8, 2003 11:30 AM

Volume Analysis (optional)

• IMAGE QUALITY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-4

Resolution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-4
Spatial Accuracy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-5
Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-5
Chapter 5 - BUILDING THE 3D MODEL
• INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-2
• STARTUP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-3
• PRE-PROCESSING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-4
• PROTOCOL PARAMETERS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-5
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-5
First Slice / Last Slice . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-5
Thresholds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-6
Model Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-7
High Res. On / Off . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-7
Filter Floaters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-8
Close Holes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-8
View Layout . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-8
• LOADING A 3D MODEL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-9
Chapter 6 - DISPLAY AND CONTROLS
• INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-2
• PANEL CONTROLS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-3
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-3
Control Panels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-3
Main control panel . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-3
Secondary control panels . . . . . . . . . . . . . . . . . . . . . . . 6-4
Undo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-4
Close . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-4
Layout presets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-5
Buttons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-8
Data Entry Fields. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-8
Sliders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-8
Thumbwheels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-8
2271012-100.book Page 5 Thursday, May 8, 2003 11:30 AM

Volume Analysis (optional)

• ON-VIEW CONTROLS AND OPERATIONS . . . . . . . . . . . . . . . . 6-9

View Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-9
3D Cursor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-10
General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-10
Active Annotaions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-12
Modifying Window Width and Level. . . . . . . . . . . . . . . . . . . 6-13
Image Location . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-15
Zoom . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-16
Roam (Scroll). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-16
View Type . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-17
Patient Name . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-17
On-View Traces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-18
Other On-View Functions . . . . . . . . . . . . . . . . . . . . . . . . . . 6-21
• THE REVIEW CONTROLLER . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-22
Reviewing the Slices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-24
Managing Bookmarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-25
Shortcuts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-26
Cardiac . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-27
• MAIN ON-VIEW MENU . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-28
• SMART CURSOR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-30
• REFERENCE IMAGE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-31
View Orientation of A Alice Plane . . . . . . . . . . . . . . . . . . . . 6-32
Paging and Adjustment of Orientation . . . . . . . . . . . . . . . . . 6-33
Chapter 7 - PROCESSING
• OVERVIEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-2
• USING THE VOLUME ANALYSIS 2 TOOLS . . . . . . . . . . . . . . . . 7-3
• USING THE VOLUME ANALYSIS A PROTOCOLS . . . . . . . . . . . 7-4
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-4
Guide Protocols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-5
Chapter 8 - REFORMATTING
• BASELINE REFORMATTING . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-2
• OBLIQUE REFORMATTING . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-4
2271012-100.book Page 6 Thursday, May 8, 2003 11:30 AM

Volume Analysis (optional)

Controls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-4
• CURVED REFORMATTING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-9
• MPVR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-11
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-11
MPVR Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-11
Rendering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-12
• BATCH REFORMATTING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-13
Chapter 9 - 3D IMAGING
• 3D OBJECT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-2
• RENDERING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-4
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-4
Rendering Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-5
Volume Rendering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-7
• VOLUMN SEGMENTATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-8
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-8
Thresholding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-8
Scalpel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-11
Paint Tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-13
Quick Paint . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-13
Paintbrush . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-15
Special Tools. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-19
Filter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-23
Advanced Processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-24
3D Object Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-27
• 3D MODEL MANAGEMENT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-28
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-28
Icon Area . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-28
Storing a 3D Model as an Icon View . . . . . . . . . . . . . . . . . . . 9-29
Recalling a 3D Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-29
Deleting an Icon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-29
Applying Colors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-30
Transferring 3D Models Between Views . . . . . . . . . . . . . . . . 9-31
2271012-100.book Page 7 Thursday, May 8, 2003 11:30 AM

Volume Analysis (optional)

Transparency Status Annotation (Merged 3D Views Only) . 9-33

• DISPLAY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-35
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-35
Rotating the 3D Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-37
Cut Planes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-41
Volume of Interest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-42
Using Colors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-43
3D Shading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-43
Chapter 10 - NAVIGATOR
• INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-3
• START NAVIGATOR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-4
• CONTROLS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-5
Active Annotations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-5
On-View Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-8
Protocol and Control Panels . . . . . . . . . . . . . . . . . . . . . . . . 10-9
• SET UP THE IMAGE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-10
• EXPLORE WITH THE “NAVIGATOR” . . . . . . . . . . . . . . . . . . . . . 10-13
Navigator Markers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-14
The Navigator Symbol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-15
• ADJUSTING THE NAVIGATORE VIEW . . . . . . . . . . . . . . . . . . . . 10-18
Rendering Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-18
Cut Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-19
Field of View . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-19
Reference Image . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-20
• SYNCHRONIZED NAVIGATION MODE. . . . . . . . . . . . . . . . . . . . 10-21
• DEFINE THE FLYTHROUGH PATH. . . . . . . . . . . . . . . . . . . . . . . 10-22
Insert Step . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-22
Insert & Seek Step (recommended) . . . . . . . . . . . . . . . . . . 10-23
Other Controls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-23
• SET MOVIE PARAMETERS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-24
Sampling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-24
Display Speed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-25
2271012-100.book Page 8 Thursday, May 8, 2003 11:30 AM

Volume Analysis (optional)

Multiview Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .10-25

Round Trip . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .10-26
• VIEW SEQUENCE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .10-27
View Sequence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .10-27
Viewing Controls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .10-28
• SAVE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .10-30
• USING NAVIGATOR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .10-31
• VIRTUAL DISSECTION OF THE COLON . . . . . . . . . . . . . . . . . . .10-33
Starting the Lumen Function . . . . . . . . . . . . . . . . . . . . . . . . .10-34
• ERROR MESSAGES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .10-37
Chapter 11 - VOLUME RENDERING
• HOW TO USE THIS CHAPTER . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-2
• INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-3
Rendering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-3
Rendering Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-4
Volumn Rendering, Overview . . . . . . . . . . . . . . . . . . . . . . . . 11-5
Volume Rendering, Principles. . . . . . . . . . . . . . . . . . . . . . . . 11-7
Comparison with Other Rendering Techniques . . . . . . . . . .11-10
Partial Volume Effect. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11-11
Transferring VR Models Between Views. . . . . . . . . . . . . . . .11-13
Focus Status Annotation (Merged VR Views Only) . . . . . . .11-15
• USING VOLUME RENDERING . . . . . . . . . . . . . . . . . . . . . . . . . .11-16
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11-16
Volume Rendering on 3D Views . . . . . . . . . . . . . . . . . . . . . .11-16
Volume Rendering on MPVR Oblique Views . . . . . . . . . . . .11-17
• CONTROLS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11-18
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11-18
VR Presets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11-18
• VR OPACITY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11-20
VR Opacity Control Panel . . . . . . . . . . . . . . . . . . . . . . . . . . .11-20
Define Opacity Curve . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11-20
Curve Type . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11-21
2271012-100.book Page 9 Thursday, May 8, 2003 11:30 AM

Volume Analysis (optional)

Low and High Control Points . . . . . . . . . . . . . . . . . . . . . . . . 11-23

Max. Opacity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-24
Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-25
Save Preset . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-27
• VR COLORS, COLOR SHADING MODE . . . . . . . . . . . . . . . . . . . 11-28
Shading Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-28
Color Sliders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-29
Brightness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-30
Color Style . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-30
Creating Color Sliders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-31
Deleting Color Sliders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-31
• VR COLORS, DEFINE OBJECTS MODE. . . . . . . . . . . . . . . . . . . 11-32
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-32
Defining an Object . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-32
Adjusting the Range of Visible Voxels . . . . . . . . . . . . . . . . . 11-33
Adjusting the Opacity of an Object . . . . . . . . . . . . . . . . . . . 11-33
Detaching an Object . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-34
Chapter 12 - ANNOTATIONS
• TYPES OF ANNOTATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-2
System Annotations (Including Active Annotations) . . . . . . 12-2
Text Annotations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-2
Measurement Annotations. . . . . . . . . . . . . . . . . . . . . . . . . . 12-2
• TEXT ANNOTATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-3
Text Entry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-4
• MANAGING ANNOTATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-5
General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-5
Preset Annotations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-5
Show/Hide Annotations . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-6
• SAVING ANNOTATIONS AS PRESETS . . . . . . . . . . . . . . . . . . . 12-7
• RECORDING THE ON-VIEW ANNOTATIONS . . . . . . . . . . . . . . 12-8
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-8
Show/Hide Annotations . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-8
2271012-100.book Page 10 Thursday, May 8, 2003 11:30 AM

Volume Analysis (optional)

Patient Name . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-8

3D Cursor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-9
Reformatted Images . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-9
Color Save . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-9
Chapter 13 - MEASUREMENTS
• ON-VIEW MEASUREMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-2
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-2
Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-3
Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-5
Editing Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-8
Measurement Annotations . . . . . . . . . . . . . . . . . . . . . . . . . . 13-9
• RULER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13-10
• PROFILE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13-11
• HISTOGRAMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13-13
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13-13
Smoothing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13-14
Cross-Section Histogram . . . . . . . . . . . . . . . . . . . . . . . . . . .13-14
Volume Histogram. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13-16
Voxel Reference Line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13-18
Voxel Class Boundaries . . . . . . . . . . . . . . . . . . . . . . . . . . . .13-19
• MEASUREMENT ACCURACY . . . . . . . . . . . . . . . . . . . . . . . . . . .13-21
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13-21
Measurement Resolution . . . . . . . . . . . . . . . . . . . . . . . . . . .13-21
Geometrical Accuracy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13-22
Image Set Resolution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13-22
Acquisition Accuracy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13-23
Display Settings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13-23
Partial Volume Averaging . . . . . . . . . . . . . . . . . . . . . . . . . . .13-23
Chapter 14 - RECORDING
• FILMING, MOVIE LOOP, SCRAPBOOK AND SCREENSAVE . . . 14-2
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-2
Annotations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-3
2271012-100.book Page 11 Thursday, May 8, 2003 11:30 AM

Volume Analysis (optional)

Image Size on Film . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-3

• BATCH . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-4
Batch Types. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-4
Batch Protocols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-5
Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-6
Custom Protocols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-12
• MOVIE LOOP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-13
Previewing a Film Batch or Movie Loop . . . . . . . . . . . . . . . 14-17
• LINK WITH DATA EXPORT APPLICATION . . . . . . . . . . . . . . . . . 14-19
Recording Images as JPEG Images . . . . . . . . . . . . . . . . . . 14-19
Recording Batych Sequences as JPEG Series or MPEG Movies
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-19
Recording Movie Loops as JPEG Series or MPEG Movies 14-19
Recording Fly-Through Sequences as MPEG Movies . . . . 14-19
Managing Data Export Sessions . . . . . . . . . . . . . . . . . . . . . 14-20
• SAVING THE 3D MODEL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-21
GLOSSARY
2271012-100.book Page 12 Thursday, May 8, 2003 11:30 AM

Volume Analysis (optional)

2271012-100.book Page 1 Thursday, May 8, 2003 11:30 AM

CHAPTER 1 Volume Analysis

CHAPTER 1 - GENERAL
Volume Analysis 2 package is an optional software package that provides
advanced functions for the analysis of 3-dimensional image data sets on Image
Works systems.
The application “builds” a 3D model of the image data in the workstation mem-
ory, from which you can generate and display reformatted views and 3D objects.
3D models generated by another application such as Advantage 3DXR can also
be loaded and processed by the application..

The equipment on which this application runs includes

one or more hard disk drives which may hold medical
CAUTION data related to patients. Such equipment may in some
countries be subject to regulations concerning the pro-
cessing of personal data and the free circulation of such
data.
It is strongly recommended that access to patient files be
protected from all persons not in medical attendance.

The physician is responsible for keeping written track of

transformations made on the model, using text annota-
CAUTION tions, for example.
See Chapter 12 “Annotations”.

1-1
2271012-100.book Page 2 Thursday, May 8, 2003 11:30 AM

Volume Analysis

1 SYSTEM REQUIREMENTS
This software application runs on a standard Image Works platform. It uses the
platform’s system for film output. It accepts image data acquired on any CT system
compatible.

2 HOW TO USE THIS DOCUMENT

• For a general overview of the features and functions of the Volume Analysis 2
software, read Chapter 3 “Introduction”.
• Datasets to be processed with Volume Analysis 2 must meet certain basic
requirements. See Chapter 4 “Image Requirements”.
• To select an exam, start Volume Analysis 2 from the Image Works Browser and
build the 3D model in the workstation memory (or load an existing 3D model),
refer to Chapter 5”Building the 3D Model”.
• To view and adjust the images displayed by Volume Analysis 2, see Chapter 6
“Display and Controls”.
• For a brief discussion of processing techniques and the use of protocols in the
Volume Analysis 2 application, see Chapter 7 “Processing”.
• To add text annotations and perform measurements on the views, refer to Chap-
ter 12 “Annotations” and Chapter 13”Measurements”.
• To film and/or save the results, see Chapter 14 “Recording”.
• For a list of the terms in this manual that are specific to the application, refer to
the Glossary at the back of the manual.

1-2
2271012-100.book Page 3 Thursday, May 8, 2003 11:30 AM

Volume Analysis

3 CONVENTIONS FOR THIS MANUAL

Throughout the text in this manual, certain type styles and symbols are used to dif-
ferentiate between one tool or graphic and another:
• Menu and control panel titles appear in bold face: Application menu.
• Menu options appear in bold face, within square brackets: [Exit]
• Graphical buttons appear in bold face, within parentheses: (View)
• On-screen tools appear in bold face, within braces: {Scroll Bar}
• On-screen prompts and messages appear in italics: Login.
• User typed-in responses appear in bold face italics: sdc.
• Keys on the workstation keyboard appear within angle brackets: <Enter>.
• Mouse buttons are underlined: left.

Whenever this manual refers to “clicking”, “selecting”, “pressinga abutton”, etc. with
the mouse, this always means using the left mouse button, unless the middle or
right mouse button is specifically mentioned.

1-3
2271012-100.book Page 4 Thursday, May 8, 2003 11:30 AM

Volume Analysis

4 PIXELS AND VOXELS

The documentation and literature concerning 3D image processing is not always
fully consisten in the use of the terms “pixel” and “voxel”. This paragraph attempts
to clarify the use of these terms in the context of this User Guide.

The definitions are simple enough:

• A “pixel” is a “picture element”, the basic element from which a two-dimensional
picture is constructed.
• A “voxel” is a “volume element”, the basice elemetnt from which a three-dimen-
sional volume is constructed.
However, an ambiguity occurs in 3D processing, because the 3D volume is defined
by a stack of images (slices). If we consider each image as a “picture” we could say
it is made up of “pixels”. But, strictly speaking, each image ia a representation of a
3D volume (the slice), hence made up of “voxels”.
• This UserGuide will use “voxel” when referring to the three-dimensional elements
that make up the images (slices) from a 3D image set.
The basic elemetns that make up the display on the workstation monitor screen are
also referred to as “pixels”. The typical workstation color monitor display consists of
1280 by 1024 pixels, and when the viewing area is divided into four views (as is
mostly the case) each view consists of 512 by 512 pixels.
• This User Guide will use “pixel” only when referring to the screen display.

1-4
2271012-100.book Page 1 Thursday, May 8, 2003 11:30 AM

CHAPTER 2 Volume Analysis

CHAPTER 2 - SAFETY
To assure an efficient and safe use of Volume Analysis 2, it is essential to read
this chapter before attempting to use the package.
This chapter is extremely important. It contains safety information that you must
thoroughly understand before you begin to use the software.
Volume Analysis 2 is an application which runs in the Image Works environ-
ment. Therefore, the user MUST have a good knowledge and understanding of
both the Image Works Basic Display.
The software is intended for use by qualified and trained personnel only. If you
need additional training, seek assistance from your GE applications specialist by
contacting the appropriate number near the end of the manual.
Make certain that the correct version of your operator manuals are readily avail-
able at all times. Make a point to review the procedures and safety precautions
periodically.

2-1
2271012-100.book Page 2 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Intended Use

The Volume Analysis 2 package is an optional software

CAUTION package that provides advanced functions for the analy-
sis of 3-dimensional image data sets on CT Workstation
systems.
These tools provide additional supplemental information,
complementing diagnosis based on classical techniques.

Measurements

The software calculates and displays measurements with a

CAUTION resolution of one decimal place (such as 0.1 mm, 0.1
degree, etc.).
You should be aware that the real measurement accuracy is
generally considerably less for a number of different rea-
sons.
To assess the accuracy of the measurements performed
with Volume Analysis 2, you should be fully familiar with
the section “Measurement Accuracy” in the chapter “Mea-
surements” of the Volume Analysis User Guide.

Measurements are more reliable when done on 2D views;

CAUTION we recommend you to at least always check on the 2D
reformatted views where exactly the points have been
deposited.

2-2
2271012-100.book Page 3 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Building the 3D Model

A 3D VIEW IS A TWO-DIMENSIONAL PROJECTION ON

WARNING! THE SCREEN OF THE 3D VOLUME. THERE IS NO INDI-
CATION ON A 3D VIEW OF HOW “DEEP” INSIDE THE
3D VOLUME THE 3D CURSOR IS LOCATED.
TO PREVENT MISINTERPRETATION OF THE CURSOR
LOCATION ON THE AXIS PERPENDICULAR TO THE
VIEWING POSITION, THE USER SHOULD ALWAYS
VERIFY THE CURSOR POSITION BY CORRELATION
WITH THE BASELINE AND REFORMATTED VIEWS.

THRESHOLDING FOR THE BUILDING OF THE 3D

WARNING! MODEL EXCLUDES ALL VOXEL VALUES OUTSIDE
THE SELECTED RANGE FROM THE 3D VOLUME.
BEFORE APPLYING THE THRESHOLD(S), MAKE SURE
THAT THE SELECTED THRESHOLD SETTINGS WILL
NOT RESULT IN REMOVING PATHOLOGIES OR OTHER
ESSENTIAL ANATOMICAL STRUCTURES.

THE USE OF FLOATER FILTERING FOR THE BUILDING

WARNING! OF THE 3D MODEL REMOVES ALL OBJECTS FROM
THE 3D MODEL THAT HAVE A SIZE EQUAL TO OR
SMALLER THAN THE SELECTED FILTER SIZE.
BEFORE APPLYING A FILTER, MAKE SURE THAT THE
SELECTED FILTER SIZE WILL NOT RESULT IN
REMOVING PATHOLOGIES OR OTHER ESSENTIAL
ANATOMICAL STRUCTURES FROM THE 3D MODEL.

2-3
2271012-100.book Page 4 Thursday, May 8, 2003 11:30 AM

Volume Analysis

THE WINDOW WIDTH AND LEVEL (W/L) SETTINGS

WARNING! DETERMINE HOW CLEARLY PATHOLOGIES AND
OTHER ANATOMICAL STRUCTURES PRESENT IN THE
CURRENT 3D MODEL CAN BE DISCERNED ON THE
VIEWS.
INCORRECT W/L SETTINGS MAY RESULT IN PATHOL-
OGIES AND OTHER ESSENTIAL ANATOMICAL STRUC-
TURES NOT BEING DISPLAYED CORRECTLY, OR
EVEN NOT BEING DISPLAYED AT ALL.
A SINGLE W/L SETTING CANNOT ALWAYS CLEARLY
DISPLAY ALL FEATURES PRESENT IN AN EXAM.
WHERE NECESSARY, USE SEVERAL DIFFERENT SET-
TINGS TO EXPLORE THE EXAM DATA.
ALSO NOTE THAT THRESHOLDING (SEE E.G., CHAP-
TER 5 AND CHAPTER 9) REMOVES ALL VOXELS WITH
VALUES OUTSIDE THE SELECTED THRESHOLD
RANGE FROM THE 3D VOLUME.
ANY PATHOLOGIES OR OTHER ANATOMICAL FEA-
TURES REMOVED IN THIS MANNER CAN NO LONGER
BE DISPLAYED, IRRESPECTIVE OF THE W/L SET-
TINGS.

A CURVED VOI VIEW CAN INTRODUCE DISTORSIONS

WARNING! IN THE SHAPE OF OBJECTS.
TO PREVENT MISINTERPRETATION OF THE SHAPE
OF AN OBJECT, THE USER SHOULD ALWAYS VERIFY
THE CURSOR POSITION BY CORRELATION WITH THE
BASELINE AND REFORMATTED VIEWS.

2-4
2271012-100.book Page 5 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Annotations on Filmed or Saved Images

When filming or saving images for diagnostic purposes,

CAUTION always make sure the patient name is displayed on all
views.

2-5
2271012-100.book Page 6 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Blank page..

2-6
2271012-100.book Page 1 Thursday, May 8, 2003 11:30 AM

CHAPTER 3 Volume Analysis

CHAPTER 3 - INTRODUCTION
Volume Analysis 2 has been developed as a full set of “tools” to process, ana-
lyze and display three-dimensional (3D) image sets on the CT workstation.
Reformatting allows you to define and display cross-sections of the 3D volume
that are oriented differently from the original acquisition images.
3D imaging allows you to extract a region or anatomical feature of interest from
the 3D volume and view, tilt and rotate the resulting “3D object” (or objects) on
the screen.
Processing can be performed either by using the “tools” of the application
directly, or (in particular for routine exams) by using the pre-defined protocols
and guides supplied with the application.
This chapter provides you with a general overview of the operations and func-
tions available in the Volume Analysis 2 software.

WHEN USING Volume Analysis 2 FOR 3D IMAGE

WARNING! DATA PROCESSING, SEVERAL CONTROLS (SUCH
AS WINDOW WIDTH AND LEVEL, THRESHOLD SET-
TINGS, ETC.) WILL AFFECT HOW ANATOMICAL FEA-
TURES AND PATHOLOGIES ARE DISPLAYED.
INCORRECT SETTINGS OF THESE CONTROLS CAN
RESULT IN PATHOLOGIES AND/OR OTHER ESSEN-
TIAL ANATOMICAL STRUCTURES NOT BEING DIS-
PLAYED, OR EVEN BEING REMOVED FROM THE
DATA BEING PROCESSED.
IT REMAINS THE RESPONSIBILITY OF THE USER TO
MAKE SURE THE DATA ARE PROPERLY PRO-
CESSED, AND WHERE NECESSARY TO CORRELATE
THE PROCESSED IMAGES WITH THE ORIGINAL
ACQUISITION DATA.

3-1
2271012-100.book Page 2 Thursday, May 8, 2003 11:30 AM

Volume Analysis

1 OVERVIEW
Volume Analysis 2 has been developed as a full set of “tools” to process, analyze
and display three-dimensional (3D) image sets on the CT workstation.
We can consider a CT exam as the representation in digital form of a three-dimen-
sional volume that can be loaded into computer memory. Various 3D processing
techniques can now be used to “look” at this three-dimensional volume:
• Reformatting: to define and display cross-sections of the 3D volume that are ori-
ented differently from the original acquisition images.
• 3D imaging: to extract a region or anatomical feature of interest from the 3D vol-
ume (volume segmentation), then view, tilt and rotate the resulting “3D object” (or
objects) on the screen.
Processing can be performed by using the “tools” (controls and functions) available
in the application directly.
However, many pre-defined protocols are supplied with the application (which com-
bine detailed processing instructions with the necessary controls and functions in
the same control panel) and can help you to eliminate much of the data entry, com-
mand selection, etc. required for processing an individual exam.

3-2
2271012-100.book Page 3 Thursday, May 8, 2003 11:30 AM

Volume Analysis

2 SUMMARY OF OPERATIONS
When using Volume Analysis 2 you will be going through the following steps:
• Selecting the exam to be processed from the Image Work Browser,
• Click the button of “VA” and starting Volume Analysis 2 then selecting a proto-
col
(for some types of exam and image data the protocol is selected automatically,
i.e. 3DXA images),

Restart Browser

Restart Display

Image File Search

Application Selection Remove Sort Network Archive PPS Queue Utilities Messages
Examinations : Exam #47, May 05 92, SMITH, JON
Exam Name Date Description Mod Fmt A Ser Type Imgs Description Mod VA

3145 J.Herman Jan 08 98 CT 1 PROSP 18 CT Add/Sub

3512 B.Fox Dec 23 97 CT

Edit Patient Click

Film

Mini Viewer
2 examinations one series
Series
Img Img Ctr Thck/SP Gntry Img Ctr Img Ctr SFOV DFOV Perfusion2
(deg) Res Matrix Midscn Archive
S–I (mm) (mm) R–L A–P (cm) (cm)
Viewer
1 S 50.0 10.0/ +0.0 R5.0 A0.0 43 Lung 512 No
2 S 45.0 10.0/ +0.0 R5.0 A0.0 43 Lung 512 No
3 S 40.0 10.0/ +0.0 R5.0 A0.0 43 Lung 512 No
4 S 35.0 10.0/ +0.0 R5.0 A0.0 43 Lung 512 No
5 S 30.0 10.0/ +0.0 R5.0 A0.0 43 Lung 512 No
6 S 25.5 10.0/ +0.0 R5.0 A0.0 43 Lung 512 No

60 images

• Displaying the exam,

• Processing the exam, using either the Volume Analysis 2 “tools” directly, or
using the pre-defined processing protocols and guides,
• If required, adding text annotations and performing measurements,
• Recording the results,
• Closing Volume Analysis 2.

3-3
2271012-100.book Page 4 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Image Requirements
To perform 3D image processing on an image set, the software builds a “3D model”
of the image set in the workstation memory. To allow this, the image set should
meet certain basic requirements.
While 3D image processing is a powerful tool to extract and display the maximum
of information from an image set, the user should always keep in mind that the
quality of the resulting information (images, measurements) can never be better
than that contained in the original image set.
See Chapter 4.

3-4
2271012-100.book Page 5 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Building the 3D Model

The first steps when using 3D processing are to select the image data you want to
work with from the Image Works Browser, click the button of “VA” and start Volume
Analysis 2.
You then select a protocol (for some types of exam and image data the protocol is
selected automatically).
A protocol controls how the 3D model in the workstation memory is built from the
data in the image set, and defines the initial view layout on the screen.
Note: Some (but not all) protocols also contain a sequence of processing steps to
perform a particular task, such as selecting and displaying a particular ana-
tomical feature. See paragraph “Processing” below, and Chapter 7.
You will notice we refer to “building a 3D model” rather than “loading the image set”,
because the image set is not simply loaded into memory. A certain amount of “pre-
processing” is performed, to improve the response time during the actual image
processing that will follow, and to select specific structures.
If you use an existing protocol, the pre-processing is completely transparent. You
can, however, create your own custom protocol.
Volume Analysis 2 can also directly load and process 3D models that you have
saved earlier, or that were generated by other applications such as Advantage
3DXR.
See Chapter 5.

Display and Controls

The Volume Analysis 2 application uses the same display layout as other applica-
tions, with a control panel on the left of the screen and a viewing area on the right.
The viewing area normally consists of four views, although, if required, you can use
the entire viewing area for a single view.
You control the Volume Analysis 2 application through a graphic user interface.
Part of the image processing functions of the Volume Analysis 2 application are
accessed via the control panel. Many of the basic viewing controls, such as those
that you use to zoom in and out on the images, or adjust window width and level
(contrast and brightness), are accessed directly on the views. Most of the review
tools such as those that you use to move through the image sets or manage book-
marks are accessed via the Review Controller, a square displayed on the views.
See Chapter 6.

3-5
2271012-100.book Page 6 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Processing
Processing can be performed either by using the “tools” available in the application
directly, or, in particular for routine exams, by using the pre-defined protocols sup-
plied with the application.
Within the Volume Analysis 2 application, you will encounter:
- Build protocols: these define how the 3D model is built and how it will be
displayed initially,
- Processing protocols: these consist of a build protocol combined with a
sequence of processing steps to perform a standard task,
- Guides: these are small additional protocols used for specific tasks such
as switching the view layout or performing certain measurements,
- Batch protocols: these define an image set that can be filmed or saved
on the image disk for later viewing, or displayed as a “movie” (animated
sequence).
Note: The “Volume Rendering” option uses its own specific set-up protocols.
Note: The “Advanced Vessel Analysis” option uses its own specific set-up proto-
cols.

3-6
2271012-100.book Page 7 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Reformatting
A CT image set represents a 3D volume defined by successive cross-sections
through the anatomy of the patient. For 3DXR image data, the 3D volume is
defined by a 3D model, but the principle and the subsequent processing are the
same.
Reformatting allows you to define and display cross-sections of the 3D volume that
are oriented differently from the original acquisition images.
A “baseline” view is a basic axial, coronal or sagittal view. Of these, the acquisition
view displays the images in the acquisition plane of the original image set, the other
two are the corresponding orthogonally reformatted views. They can be moved to
show any location in the 3D volume, but remain aligned parallel to the three main
axes of the RAS coordinate system. An oblique view is a plane reformatted view
that can be both moved and rotated to any location and orientation within the 3D
volume.
If a feature of interest extends significantly in three dimensions, a single standard
baseline or oblique view cannot show the full extent of the feature. To overcome
this, you can use curved reformatting, curved VOI or MPVR; alternatively you can
create a batch of regularly spaced reformatted images.
With a curved view, instead of using a plane cross-section, you “unfold” a curved
cross-section of the 3D volume, that conforms as much as possible to the shape of
the feature of interest. With MPVR (Multi-Planar Volume Reconstruction), instead of
using baseline and oblique views representing slices that are only one voxel thick,
you define a “thick” slice that encompasses the feature of interest.
See Chapter 8.

3-7
2271012-100.book Page 8 Thursday, May 8, 2003 11:30 AM

Volume Analysis

3D Imaging
3D imaging allows you to extract a region or anatomical feature of interest from the
3D volume (this is referred to as volume segmentation) and display the result as
one or more 3D objects.
You use thresholding to extract a 3D object by selecting a range of voxel values
that represents a specific tissue or anatomical feature. The scalpel tool allows you
to perform “cuts” in the 3D volume to define a region of interest. The paint tools
allow you to mark a region of interest with colored “paint”, then display only this
region.
Different rendering modes define how the resulting 3D object will be displayed:
− Surface shading treats the 3D object as if it were fully opaque.
− Projection shading treats the 3D object as if it were translucent. You can select
how the voxel values across the depth of the 3D object will be displayed (maxi-
mum intensity, minimum intensity, average, etc.).
− Volume rendering (option) treats the voxel values along the the depth of the 3D
object as representing a varying degree of opacity.
You can rotate a 3D object to examine it from all angles, define cut planes, use a
spherical “shutter” to show only an essential part of the object, merge several sepa-
rately defined 3D objects into a composite view, and use color to distinguish sepa-
rate objects.
See Chapter 9.

Annotations
As with Image Works Browser, we can distinguish three main types of annotations
on the Volume Analysis 2 views:
system annotations, user text annotations and measurements.
See Chapter 12.

3-8
2271012-100.book Page 9 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Measurements
You can measure voxel value, distance, angle, area and total volume on the views,
in a manner similar to the AW Basic Display application. Measurement annotations
are displayed directly on the views, and can be filmed or saved with the views like
any other annotation. You can also create measurement annotations combined
with explanatory text.
You can display a ruler (grid or tick marks) to better appreciate dimensions on the
views.
You can define and display profile graphs and histograms. These are displayed on
separate views, that also can be filmed or saved as required. A profile graph
shows the voxel value along a 3D trace (profile). Histograms graphically show the
voxel value distribution and associated statistics within an area (cross-section) or
within a volume.
Measurement accuracy depends on various factors. To assess the accuracy of any
measurements you want to perform, refer in particular to paragraph 5 “Measure-
ment Accuracy” in Chapter 13.
See Chapter 13.

Recording
You can record the results of your work by printing them on film or paper , or by sav-
ing them on the image disk of the workstation. For this, you use the same Scrap-
Book applications and Screen Save function that you also use with Image Works
Basic Display.
When working with three-dimensional image sets, you will often have a requirement
for recording a set of regularly spaced images (such as a set of reformatted
images, or a set of images showing the rotation of a 3D object in successive steps).
The Batch function of Volume Analysis 2 allows you to set up such a set of
images, then preview the set as an animated sequence (movie loop), film it, and/or
save it on the image disk.
A batch setup can be saved and re-used as a Batch protocol. Several successive
batch setups can be combined into a single protocol.
You can also save the current 3D model “as is” on the image disk, to be reloaded
and further processed at a later time. A saved 3D model cannot be displayed or
processed by other applications.
See Chapter 14.

3-9
2271012-100.book Page 10 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Blank page.

3-10
2271012-100.book Page 1 Thursday, May 8, 2003 11:30 AM

CHAPTER 4 Volume Analysis

CHAPTER 4 - IMAGE REQUIRE-

MENTS

To perform 3D image processing on an image set, the software either builds a

“3D model” of the image set in the workstation memory, or loads a 3D model
from the workstation image disk.
Either way, the image data should meet certain basic requirements listed in this
chapter.
3D image processing is a powerful tool to extract and display the maximum of
information from an image set. However, the user should always keep in mind
that the quality of the resulting information (images, measurements) can never
be better than that contained in the original image set.
Resolution of the 3D images and measurement accuracy are usually limited by
the inter-slice distance of the image set.

4-1
2271012-100.book Page 2 Thursday, May 8, 2003 11:30 AM

Volume Analysis

1 IMAGE REQUIREMENTS
It is not possible to perform 3D and reformatted plane reconstruction on just any set
of image data available on the Image Works Browser.
Volume Analysis 2 accepts CT image sets, 3D models (type 3D OBJ) generated
with Advantage 3DXR or saved with Volume Analysis 2, and Reformat images
(type RFMT) saved with Volume Analysis 2.
A CT image set to be used with Volume Analysis 2 should meet at least the follow-
ing requirements:
• Field of view, matrix size and display center must be the same for all images in
the set,
• Orientation and gantry tilt should be the same for all images in the set,
• Tilted acquisitions are not supported for right/left decubitus patient orientations,
• There MUST NOT be two images corresponding to the same location in the set,
• The set should include only AXIAL, SAGITTAL, CORONAL or OBLIQUE images.
Other types such as screen saves, etc. are not supported,
• Inter-slice distance MUST be less than 10 mm.
• The series must contain at least 5 images.
The Volume Analysis 2 software uses the FIRST selected image in the Browser
as a basis for using/discarding the other images selected for building the 3D model.
For example, any images having a matrix size or gantry tilt different from that of the
first selected image in the Browser are discarded.
A 3D model to be used with Volume Analysis 2 should be fully compatible with the
3D OBJ format recognized by Volume Analysis 2.
This is the case for 3D models generated with Advantage 3DXR, and 3D models
saved with Volume Analysis 2 for further processing.
Note: A 3DXR image set (i.e., the original 2D images used by Advantage 3DXR to
generate a 3D model) can be viewed using the Display Viewer or Mini-
Viewer, but it cannot be selected or processed with Volume Analysis 2.
When using Volume Analysis 2 to load, display and process a 3D model gener-
ated with Advantage 3DXR, certain features such as annotations are specific to
3DXR.

4-2
2271012-100.book Page 3 Thursday, May 8, 2003 11:30 AM

Volume Analysis

It remains the responsibility of the physician to determine

whether the maximum inter-slice distance of 10 mm is
CAUTION acceptable for the exam.
Always bear in mind that, within an exam, details with
dimensions in the order of or less than the inter-slice dis-
tance cannot be identified with an acceptable degree of reli-
ability.

4-3
2271012-100.book Page 4 Thursday, May 8, 2003 11:30 AM

Volume Analysis

2 IMAGE QUALITY
3D image processing can be a powerful tool to extract and display the maximum of
information from an image set. However, the quality of the resulting information
(images, measurements) can never be better than that contained in the original
image set.

Resolution
The resolution of reformatted and 3D images is primarily determined by the resolu-
tion of the original acquisition images and the inter-slice distance.
For images in the acquisition plane, resolution is limited by the resolution of the
original image set (typically 512x512 for CT). For images in a plane that is not par-
allel to the original acquisition plane, resolution is determined by the inter-slice dis-
tance.
Note: The viewing area on a typical workstation platform color monitor display con-
sists of 1024x1024 pixels, or 512x512 pixels when the viewing area is
divided into four views. Hence display resolution is normally not a limiting
factor.
Ideally, to obtain the same resolution in all planes, the voxels in the image set
should be isotropic, i.e., have the same dimensions in all three axes.
As an example, in a typical image with a field-of-view of 25cm and a 256x256
matrix the individual elements have a size of 1mm x 1mm (approx.). To obtain the
same resolution in all three axes, the inter-slice distance should also be 1mm.
Practical considerations such as type of exam, scan time, or patient radiation dose,
may lead to the choice of a larger inter-slice distance. Satisfactory 3D images can
be obtained with larger inter-slice distances (oblong voxels), up to several mm in
the above example.
An important point to remember is that it is not possible to identify details within the
exam with dimensions in the order of or less than the inter-slice distance with any
degree of reliability.
The Volume Analysis 2 application will accept inter-slice distances of up to 10 mm.
However, with such a large inter-slice distance, the 3D reconstruction of the image
set may be less than satisfactory because of the lack of resolution of the resulting
3D or reformatted images along the axis perpendicular to the acquisition plane.

At all times, it remains the responsibility of the physician

CAUTION to determine whether the inter-slice distance used for a
particular exam is acceptable.

A discussion of the optimum value of inter-slice distance for a given type of exam is
obviously outside the scope of this manual. Consult the appropriate literature on
the subject.

4-4
2271012-100.book Page 5 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Spatial accuracy
Spatial accuracy is limited by the accuracy with which the acquisition system has
been set up and calibrated, and may be degraded by such effects as motion and
streak artifacts.
The spatial accuracy of MR images is dependent on the patient, the pulse
sequence and the MR system itself. Features such as metallic implants or air-bone
interfaces may lead to susceptibility artifacts and spatial distortions that tend to
degrade the overall spatial accuracy to a value less than that observed with a Qual-
ity Assurance phantom, even on a perfectly tuned MR system.

Measurements
The Volume Analysis 2 application normally displays measurement values to one
decimal place (0.1mm for inear measurements, 0.1 for angle measurements and
0.1mm2 for area measurements). Measurement accuracy s obviously less than
this. The main factors determining measurement accuracy are:
• Acquisition accuracy,
• Resolution and inter-slice distance of the image set,
• Partial volume effects,
• Display settings.
See Chapter 13 “Measurements”, paragraph 5 for details.

4-5
2271012-100.book Page 6 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Blank page.

4-6
CHAPTER 5 Volume Analysis

CHAPTER 5 -BUILDING THE 3D

MODEL

The first steps when using 3D processing are to select the image data you want
to work with from the Image Works Browser, and start Volume Analysis 2. You
then select a protocol (for some types of exam and image data the protocol is
selected automatically).
A protocol controls how the 3D model in the workstation memory is built from the
data in the image set, and defines the initial view layout on the screen.
A certain amount of “pre-processing” is performed before the actual loading of
the image set to improve theresponse time during the actual image processing
and to select structures of interest. This is why we refer to “building a 3D model”
rather than “loading the image set”.
If you use an existing protocol, the pre-processing is completely transparent.
You can, however, creat your own custom protocol. The parameters that define
the building of the 3D model are listed and described in this chapter.
The above applies when you process standard CT image sets. Volume Analy-
sis 2 also allows you to reload and process 3D models that you have saved ear-
lier, or that were generated by Advantage 3DXR. In those cases, data are
loaded directly and protocol selection is automatic.
Volume Analysis 2 also allows you to reload Direct 3D presets. In this case,
data are loaded directly and protocol selection is automatic.
Note: Some protocols also contain a sequence of processing steps to perform a
particular task, such as selecting and displaying a particular anatomical
feature. See Chapter 7.

5-1
Volume Analysis

1 INTRODUCTION
For the purpose of 3D image processing and display, the workstation constructs a
“3D model” in the workstation memory from the data in an image set. This means
that the data is loaded and arranged in the computer memory in such a way that it
can be accessed in the most efficient manner during the subsequent image pro-
cessing. This is also referred to as “pre-processing”.
Secondly, the display of 3D objects with surface shading requires that for each
voxel the orientation of the iso-value plane (the plane in which surrounding voxels
of the same value lie) is known. Computing this information anew each time for
each image would make the display response time inacceptably slow. Hence, if 3D
objects are to be displayed with surface shading, the necessary data are computed
for each voxel during the pre-processing and added to the “3D model”.
When loading an existing 3D model that was either saved earlier, or generated by
Advantage 3DXR, the data are transferred directly into the workstation memory and
no pre-processing takes place.
When loading an existing Direct 3D that was either saved earlier, or generated by
the Direct 3D software, the data are transferred directly into the workstation mem-
ory and no pre-processing takes place.
When clicking on one of the preprogrammed buttons of the Control Panel, the data
are transferred directly into the workstation memory and no pre-processing takes
place.

5-2
 Volume Analysis

2 STARTUP
You select the patient, exam, exam series and images in the Image Works Browser
and click the button of “VA” then start Volume Analysis 2.

Restart Browser

Restart Display

Image File Search

Application Selection Remove Sort Network Archive PPS Queue Utilities Messages
Examinations : Exam #47, May 05 92, SMITH, JON
Exam Name Date Description Mod Fmt A Ser Type Imgs Description Mod VA

3145 J.Herman Jan 08 98 CT 1 PROSP 18 CT Add/Sub

3512 B.Fox Dec 23 97 CT

Edit Patient Click

Film

Mini Viewer
2 examinations one series
Series
Img Img Ctr Thck/SP Gntry Img Ctr Img Ctr SFOV DFOV Perfusion2
(deg) Res Matrix Midscn Archive
S–I (mm) (mm) R–L A–P (cm) (cm)
Viewer
1 S 50.0 10.0/ +0.0 R5.0 A0.0 43 Lung 512 No
2 S 45.0 10.0/ +0.0 R5.0 A0.0 43 Lung 512 No
3 S 40.0 10.0/ +0.0 R5.0 A0.0 43 Lung 512 No
4 S 35.0 10.0/ +0.0 R5.0 A0.0 43 Lung 512 No
5 S 30.0 10.0/ +0.0 R5.0 A0.0 43 Lung 512 No
6 S 25.5 10.0/ +0.0 R5.0 A0.0 43 Lung 512 No

60 images

5-3
Volume Analysis

3 PRE-PROCESSING
The protocol that you select after starting Volume Analysis 2 defines how the 3D
model in the computer memory will be constructed from the image set of the exam
that you have selected.
At its most basic, the pre-processing amounts to little more than loading and orga-
nizing the image set in the workstation memory.
However, a judicious choice of certain parameters (such as range of slices to be
loaded, range of voxel values, resolution) can significantly reduce the amount of
memory used, and hence improve the response rate during the 3D processing and
display of the exam.
Secondly, as noted in paragraph 1, the algorithms used to display 3D objects with
surface shading require the generation of additional data for each voxel during the
pre-processing.
Surface shading is not used for the display of reformatted images, or when 3D
objects are displayed using rendering modes such as MIP (see Chapter 9 “3D
Imaging”). In those cases the additional data are not required, and memory use
can be reduced as a consequence.
When loading an existing 3D model, no pre-processing takes place and protocol
selection is automatic.

5-4
 Volume Analysis

4 PROTOCOL PARAMETERS
Overview
When you use an existing protocol to build the 3D model, the pre-processing as
defined in the protocol is completely transparent.
You can also create your own custom protocol. The parameters that you can define
to set up a custom protocol are listed and described below.
Important:
During processing, you cannot restore any data to your 3D model that you
have excluded during the initial building of the 3D model.
For instance, if you have selected a given range of voxel values to be used
for the 3D model, by setting the low and/or high threshold in the protocol (see
further), none of the data from the exam with values outside this range will be
present in the 3D model.
The only way to include such data is to start from scratch and redefine the
settings in the protocol.

First slice / last slice

You can define whether to use the entire image set, or only part of it.
By not including those slices of the image set that do not contain any anatomical
features that you are interested in, you reduce memory use and hence speed up
processing.
You do this by modifying the First slice and Last slice numbers in the protocol.
Note: You can also do this by using the image list in the Image Works Browser to
select only the image range that you want to use before starting Volume
Analysis 2

5-5
Volume Analysis

Thresholds
You can use the protocol to include right from the start only a specific range of voxel
values in the 3D model.
To do this, set the Low Threshold and/or High Threshold as required.
Note: You will be able to further narrow down the range of voxel values later on
during the 3D processing, for instance to select a particular anatomical fea-
ture ("thresholding", see Chapter 9). You cannot, however, at a later stage
re-introduce a range of voxel values that you have excluded during the initial
building of the 3D model.

THE USE OF THRESHOLDING FOR THE BUILDING

WARNING! OF THE 3D MODEL EXCLUDES ALL VOXEL VALUES
OUTSIDE THE SELECTED RANGE FROM THE 3D
MODEL.
BEFORE APPLYING THE THRESHOLD(S), MAKE
SURE THAT THE SELECTED THRESHOLD SET-
TINGS WILL NOT RESULT IN REMOVING PATHOLO-
GIES OR OTHER ESSENTIAL ANATOMICAL
STRUCTURES FROM THE 3D MODEL.

5-6
 Volume Analysis

Model mode
The model mode further defines which data from the image set are used, and
whether additional data are added to the model to speed up the subsequent 3D
processing.
In the Volume mode, all of the selected data from the entire image set are used to
build a complete 3D model. This mode uses the largest amount of computer mem-
ory.
The Surface only mode is used to build a “hollow” 3D model: only the surface of
the object is generated. This mode uses much less computer memory than the Vol-
ume mode, hence display is faster, but since the model contains no data from the
inside of the object, little if any further processing is possible.
Surface-only mode can be a good way to rapidly visualize just the surface of a par-
ticular tissue, such as bone.
In both the above modes, for each voxel additional data are generated that are
needed when displaying 3D objects with surface shading (see paragraph 1 in this
chapter, and Chapter 9 “3D Imaging”).
The Reformat/MIP mode is used for the display of reformatted images, or when 3D
objects are displayed using rendering modes such as MIP (see Chapter 9 “3D
Imaging”). These types of display do not use surface shading, so in this mode the
additional surface shading data are not added and memory use is reduced, but sur-
face shading is not available on the 3D views.

High Res. on/off

If the image set was acquired with a 512x512 matrix, you can display and process it
either with a resolution of 512x512 (high resolution) or 256x256 (low resolution) by
setting High Res. on or off.
The advantage of using low resolution is a reduction in memory use up to eight-fold
with a corresponding improvement in response time.

It remains the responsibility of the physician to deter-

CAUTION mine whether the loss of fine detail and reduction in
measurement accuracy resulting from the use of low
resolution is acceptable for the exam.

Note: If the image set was acquired with a 256x256 matrix, setting High Res. on or
off will have no effect: the function cannot increase the resolution of an
image.

5-7
Volume Analysis

Filter floaters
“Floaters” are small residual objects in the 3D model, usually resulting from noise in
the original image set. You can choose to remove these during the process of
building the 3D model, or at a later stage if you prefer (see Chapter 9).
Floaters removed by this function while building the 3D model cannot later be
restored to the 3D volume
.

THE USE OF FLOATER FILTERING FOR THE BUILD-

WARNING! ING OF THE 3D MODEL REMOVES ALL OBJECTS
FROM THE 3D MODEL THAT HAVE A SIZE EQUAL
TO OR SMALLER THAN THE SELECTED FILTER
SIZE.
BEFORE APPLYING A FILTER, MAKE SURE THAT
THE SELECTED FILTER SIZE WILL NOT RESULT IN
REMOVING PATHOLOGIES OR OTHER ESSENTIAL
ANATOMICAL STRUCTURES FROM THE 3D MODEL.

Close holes
If the 3D model is built using only a selected range of voxel values (see paragraph
“Tresholds” above) this can result in “holes” appearing inside the 3D volume (i.e.
closed spaces inside the 3D volume where the voxel value is outside the selected
range).
If Close holes is off, the voxel value inside such inner holes will be set to the same
value as the outside of the 3D volume (“empty space”). If Close holes is on, the
inner holes will be filled with the original voxel values from the exam and thus pro-
cessed as part of the 3D volume.

View layout
The initial layout of the views on the screen is also defined in the build protocol. For
each of the four views you can choose the view type that will be used when the
images are first displayed. After that, you use the application controls to change
the views if and when necessary.

5-8
 Volume Analysis

5 LOADING A 3D MODEL
You can save the current 3D model onto the image disk so that you can reload it at
a later time for further processing. See Chapter 14 “Recording”, paragraph 5 “Sav-
ing the 3D Model”.
Volume Analysis 2 can also directly load and process 3D models that were gener-
ated by other applications such as AW.

5-9
Volume Analysis

Blank page.

5-10
2271012-100.book Page 1 Thursday, May 8, 2003 11:30 AM

CHAPTER 6 Volume Analysis

CHAPTER 6 - DISPLAY AND CON-

TROLS
The Volume Analysis 2 application uses the same display layout as other appli-
cations, with a control panel on the left of the screen and a viewing area on the
right.
The viewing area normally consists of four views. If required, you can use the
entire viewing area for a single view (see “Enlarge” function in paragraph 5).
The Review Controller is a square displayed on the selected view on which most
of the review controls are located and which can be closed if necessary.
You control the Volume Analysis 2 application through a graphic user interface.
Most of the image processing functions of the Volume Analysis 2 are accessed
via the control panel. Most of the review functions such as those used to move
through the image sets or to bookmark areas of interest are accessed from the
Review Controller displayed on the selected image. Many of the basic viewing
controls, such as those that you use to zoom in and out on the images, and
adjust window width and level (contrast and brightness), are accessed directly
on the views. All the basic adjustments of the viewing area can be stored in Lay-
out Presets to be reused in later sessions.
This chapter briefly summarizes the panel controls and the Review Controller
and describes the on-view controls and features that are common to most if not
all views and the procedure used to memorize in Layout Presets them from one
session to another.
The controls that are specific to particular functions (e.g., setting slice thickness
and rendering mode on reformatted views, or defining cut planes on 3D views)
are described in the chapters that deal with those functions.

6-1
2271012-100.book Page 2 Thursday, May 8, 2003 11:30 AM

Volume Analysis

1 INTRODUCTION
The main controls take the form of graphic buttons and menus, displayed in the
main control panel (at the left of the screen) or in secondary control panels that are
displayed after selecting a function from the main control panel or from a menu.
You will also use data entry fields, that allow you to enter text or numerical values
from the keyboard.
Some special controls such as sliders and thumbwheels allow you to make adjust-
ments interactively.
The use of graphic buttons, menus and the other controls available in the control
panels. They are summarized briefly in paragraph 2 “Panel controls”.
Most of the viewing controls such as those used to move through the image sets,
bookmark areas of interest or create cine sequences are accessed from the
Review Controller. Its different functions are described in paragraph 4 “Review
Controller”.
Certain basic viewing controls, such as those that you use to roam the images
within the viewport, zoom in and out on the images or adjust window width and
level (contrast and brightness), are accessed directly on the views.
These controls are described separately in Section 3 “On-View Controls”. They
include the active annotations, the use of the middle mouse button for window width
and level control, on-view menus, and the methods used to create and edit traces
on the views.
Some of these layout controls can be stored in Layout Presets to be reused from
one session to another. The procedure to create and manage Layout Presets is
described in paragraph 2 “Panel controls”.

6-2
2271012-100.book Page 3 Thursday, May 8, 2003 11:30 AM

Volume Analysis

2 PANEL CONTROLS
Overview
To access a function, open a menu, select a menu item, etc., you place the mouse
pointer on the button or menu item and click on the left mouse button.
Note: Whenever this manual refers to “clicking”, “selecting”, “pressing a button”,
etc. with the mouse, this always means using the left mouse button, unless
the middle or right button is specifically mentioned.

Control panels
Main control panel

The principal controls in the main control panel

New Protocol are:
Additional Guides • Protocol buttons,
• Layout Presets panel
Layout Presets
• Rotation and translation controls,
• Tool menu buttons,
• Save/Recall button,
Option/More...
• Close button.
These controls and the corresponding functions
Rotate/Translate are described where relevant in the different
chapters of this user guide, with the exception
S I A P L R of the (Close) buttons which are described
below.
3D Tools

Display Tools

Filming Tools

Save/Recall

Close

6-3
2271012-100.book Page 4 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Secondary control panels

Most of the controls in the main control panel give you access to secondary control
panels that contains the controls necessary for a particular function. Secondary
control panels are displayed on top of the main viewing area. At times, a second-
ary control panel may mask a view that you want to use. When this happens, you
can move the panel to a different location by clicking and dragging on its title bar.

Close
The (Close) button allows you to close the Volume Analysis 2 application and
return to the Image Works Browser.

When you close the application, any work in progress that

NOTICE has not yet been filmed or saved will be lost.

6-4
2271012-100.book Page 5 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Layout Presets
The user will be able to store a specific layout context, allowing him to recreate it
without having to independantly set each of the parameters again.
The following settings can be saved in the Layout Presets:
• Screen / Image layout (axial, coronal, sagittal...)
• Field of View (FOV) settings
• Width and Length (W/L)
• Slice thickness
• Display algorithm (MIP, MinIP...)
• Motion lock
• Orientation
• Rendering modes for Navigator views
• VR preset applied
• Lumen view settings

For a description of how to set the various parameters,

NOTICE
refer to paragraph 3 “On-view Controls”.

Note: This is a list of the currently existing parameters to be saved in Preset Lay-
out. However, new possible parameters can be added at any time.
The user can create, modify and delete a layout from the Layout Presets panel, as
well as activate a stored context for the currently displayed dataset.
Default Layout Preset
The Layout Preset panel always contains a Layout Preset which has the name of
the protocol initially used at software startup and is a copy of it. This Preset cannot
be modified or deleted.
This allows the user to reverse to the parameters of the original protocol after hav-
ing modified some of them or applied a different Layout Preset, without having to
leave the software or to restore them manually.
Applying a Layout Preset
Click on the name of the Layout Preset you want to apply to the currently displayed
exam. If the name of the layout you want to apply does not appear in the Layout
Presets window of the Control Panel, click on the (More) button to open up a pull-
down menu containing a list of all the created layouts. Apply the layout you want by
clicking on its name in the list.

6-5
2271012-100.book Page 6 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Creating a Layout Preset

To create a Layout Preset you must:
− Click on the (More) button of the Layout Presets panel. A pulldown menu opens
up on the right of the panel.
− Click on the [Save layout preset] option of the menu. The [Save a new layout
preset] panel is displayed on the bottom left side of the viewport.
− Enter the name of the Preset (the name of a type of exam should generally be
the best). Parameters can be set for each of the four views. Click on the corre to

sponding view number, 1 to 4 1 2 point on the view you want to modify.

3 4

− Click on the (Create shortcut) button to access to a secondary panel allowing

you to:
• have this Layout Preset appear as a favorite protocol in the Software Man-
ager of Image Works. You will then be able to paste it on one of the Applica-
tion buttons of the main Control Panel. Tick the box to validate the option.
• have this Preset appear as a new protocol on the Anatomy screen. Tick the
box to validate the option.
− Click on (Save) to save the layout and on (Cancel) to close the panel.
− The new Layout Preset is created on the window of the Layout Preset panel. To
remove it from the Layout Preset panel, click on it with the middle mouse button
and drag it out of the window. You can have up to 6 Presets on the panel (includ-
ing the default Layout Preset, which always appears on the first line). To bring
the new Layout Preset back on the window, click on it with the middle mouse but-
ton and drag it back on the panel. If you already have 6 Preset in the window,
the new Preset overwrites the one you drag it on.
Modifying/Deleting a Layout Preset
To modify/delete a Layout Preset you must:
− Click on the (More) button of the Layout Preset panel. A pulldown menu opens
up on the right of the panel.
− Click on the [Modify Layout] option of the menu. The [Preset Manager] panel is
displayed on the bottom left side of the viewport.

6-6
2271012-100.book Page 7 Thursday, May 8, 2003 11:30 AM

Volume Analysis

− Click on the [Click for Details] image to access to the [Current Settings] panel.
You can select or deselect the layout details you want to store in this particular
layout, by ticking the coresponding box. This can be done for each of the

four views. You click on the view 1 2 you want to modify.

3 4

− Click on the (Close) button to close the sub-panel when you are finished.
− Tick the (Save as Protocol) box if you want this layout to appear in the Anatomy
protocol list or not.
− Tick the (Save as Favorite) box if you want this layout to appear as a favorite
protocol in the Software Manager of AW or not.
− Click on the (Rename) button if you want to rename a layout. A dialog window
pops up allowing you to modify the name of the selected layout.
− Click on the (Delete) button if you want to suppress a layout. A message window
pops up asking you to confirm if you want to delete the selected layout.
− Click on the (Close) button to close the panel.

6-7
2271012-100.book Page 8 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Buttons
The functioning of most buttons is obvious, You use menu buttons open a menu of
related functions, then click on a menu item to use the corresponding function. You
use function buttons to turn a function on or off, or to select a particular mode for
that function (e.g. view type, or 3D cursor shape).

Data entry fields

You use data entry fields to enter text or numerical values from the keyboard.

10.0 Move the mouse pointer inside the field and click. If necessary
delete the existing data in the field by means of the <BackSpace> or
<Del> key, then enter the new data from the keyboard.

Sliders
You use sliders to adjust certain settings interactively: the result of moving the slider
is displayed immediately. Examples are threshold settings (see for instance Chap-
ter 9, paragraph “Tresholding”) or the frames-per-second setting for a cine loop dis-
play (see Chapter 14).

To adjust a slider, click and drag on the slider knob to set the
10 slider rapidly, or click to the right or left of the slider knob to
increase or decrease the value by one.

Thumbwheels
You use thumbwheels in a manner similar to sliders to adjust certain settings inter-
actively.

To adjust a thumbwheel, click and drag up or down on the central

“wheel” to set the thumbwheel rapidly, or click on the arrows at
top and bottom to increase or decrease the value by steps.

6-8
2271012-100.book Page 9 Thursday, May 8, 2003 11:30 AM

Volume Analysis

3 ON-VIEW CONTROLS AND OPERATIONS

View Selection
To work on a view it must be selected. View selection in the Volume Analysis 2
application functions in the same way.
• A single click on a view makes it the primary view, and changes any previously
selected views to secondary,
• A double click on a view makes it the currently selected view, and deselects all
other views,
• A triple click on a view makes it the primary view, and selects all other views as
secondary.
A red border designates the view currently selected as the primary view. Any views
selected as secondary are indicated by a green border. The absence of a colored
border indicates that a view is not selected.
Certain actions such as changing window width/level are replicated from the pri-
mary view to those views you have selected as secondary.
Note: View selection is usually implicit, i.e. as soon as you click on a view to make
an adjustment, or to perform an operation such as creating a trace, the view
is selected as the primary view at the same time.

6-9
2271012-100.book Page 10 Thursday, May 8, 2003 11:30 AM

Volume Analysis

3D Cursor
General

The 3D cursor identifies a three-dimensional point on all

selected views.
Normally the 3D cursor is displayed in red. If it is moved to
a position outside the 3D volume its color changes to yel-
low.
The 3D cursor appears on the views as crosshairs, as an
arrow or as a dot. You can select the 3D cursor style in
the (Display Tools) > [Preferences] panel.

To move the 3D cursor:

• Move the mouse pointer to the location where you want to place the 3D cursor,
and press the <Shift> key on the keyboard: the 3D cursor moves to the new posi-
tion. You can also hold down the <Shift> key and move the mouse pointer: the
3D cursor will follow the mouse movements for as long as you hold down the
<Shift> key.
Because the defined point is three-dimensional, as you move the 3D cursor around
on one view, the other selected views will be updated at the same time.
Note: If a view is not selected, the 3D cursor mark on that view remains frozen at
the last defined position, and is not updated until the view is selected again.

On 3D views
The 3D cursor is also displayed on three-dimensional views (this includes both
basic 3D views and views created with the Navigator and Volume Rendering
options).
Because the image on the screen is a two-dimensional projection of the 3D volume,
you have no indication on the view of how “deep” inside the 3D volume the 3D cur-
sor is located. To determine the 3D cursor position unambiguously, you need to
correlate the 3D view with the baseline views.
When you move the 3D cursor on a baseline or reformatted view (which is essen-
tially two-dimensional), the 3D cursor stays in the plane of the view.
On a 3D view, when you move the 3D cursor, the software automatically positions it
along the line of sight onto the feature that has generated the image at that point.
For instance, on a 3D MIP (Maximum Intensity Projection) view, the 3D cursor is
placed along the line of sight onto the point of maximum intensity. On a Navigator
view, the 3D cursor is placed on the “wall” of the vessel or other anatomical feature
being displayed.

6-10
2271012-100.book Page 11 Thursday, May 8, 2003 11:30 AM

Volume Analysis

As noted above, you have NO indication of this on the 3D view itself: you still need
to correlate the 3D view with the baseline views.
Note: The position along the line of sight of the feature that has generated the
image at a given point cannot always be determined unambiguously.
If the cursor position is not be what you expected, use a baseline or refor-
matted view to reposition the 3D cursor.
If the 3D volume is empty (“hollow”) along the line of sight, the 3D cursor will
move outside the volume. When this happens, place the 3D cursor on a
clearly identifiable feature on the 3D view.
Knowing how the 3D cursor behaves when you move it on a 3D view allows you to
deposit points and create traces directly on 3D views, e.g., to perform measure-
ments. However, you should use this feature with care. Correlation with baseline
and reformatted views is essential.

A 3D VIEW IS A TWO-DIMENSIONAL PROJECTION

ON THE SCREEN OF THE 3D VOLUME. THERE IS
NO INDICATION ON A 3D VIEW OF HOW “DEEP”
WARNING! INSIDE THE 3D VOLUME THE 3D CURSOR IS
LOCATED.
TO PREVENT MISINTERPRETATION OF THE CUR-
SOR LOCATION ON THE AXIS PERPENDICULAR TO
THE VIEWING POSITION, THE USER SHOULD
ALWAYS VERIFY THE CURSOR POSITION BY COR-
RELATION WITH OTHER VIEWS SUCH AS BASE-
LINE OR REFORMATTED VIEWS, OR ANOTHER 3D
VIEW.

6-11
2271012-100.book Page 12 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Active Annotations
Certain on-view annotations (image location, W/L, etc.) are shown in red, to indi-
cate they are active. When you move the mouse cursor onto any of these annota-
tions, a help message (in green) shows the function. The various functions are
described in the following paragraphs.

The value of some of these parameters can be stored in

Layout Presets to enable you to activate them without hav-
NOTICE ing to set again each of the parameters independently
(Refer to paragraph 2 section “Layout Presets”.

To use the numerical active annotations:

• Move the mouse pointer onto the annotation, and
− Click on the left or right mouse button to increase or decrease the value in steps,
or
− Click and hold the middle mouse button and drag to the left or to the right to
change the value in a continuous fashion, or
− Type the new value on the keyboard and validate with <Enter>.
Note: Unlike the other numerical active annotations, clicking left or right on the W/
L annotations displays a Preset menu. See paragraph “Modifying Window
Width and Level” below.
To use the text active annotations:
• Click with the left or right mouse button on the annotation and select the desired
function or setting from the pop-up menu.
The orientation annotations (L, R, A, P, S, I) on the sides of the images can be used
to roam (scroll) the image. See paragraph “Roam(Scroll)” bellow.

6-12
2271012-100.book Page 13 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Modifying Window Width and Level

To modify window width (“contrast”) and window level (“brightness”) of the images,
you can use the mouse or the active annotations on the views.

You use the mouse to modify window W/L in the

same way as on the Image Works Basic Display
Viewer. Click and hold the middle mouse button on a
view, then move the mouse left or right to modify win-
dow width, and up or down to modify window level.
The W/L annotations at the bottom of the views are
updated automatically whenever you use the mouse
to modify window W/L. These are active annota-
tions, so you can also modify them directly:

• Move the mouse pointer onto the annotation, and

− Click and hold the middle mouse button and drag to the left or to the right to
change the value in a continuous fashion, or
− Type the new value on the keyboard and validate with <Enter>, or
− Click on the left or right mouse button to display the W/L Preset menu. The W/L
preset values and names available in this menu are the same as those for the
Image Works Viewer.
Note: You can independently modify the window W/L of the reference images (nor-
mally at bottom right of the view, see paragraph 7). Click and hold the mid-
dle mouse button on the reference image, then move the mouse as before.

6-13
2271012-100.book Page 14 Thursday, May 8, 2003 11:30 AM

Volume Analysis

THE WINDOW WIDTH AND LEVEL (W/L) SETTINGS

DETERMINE HOW CLEARLY PATHOLOGIES AND
OTHER ANATOMICAL STRUCTURES PRESENT IN
WARNING! THE CURRENT 3D MODEL CAN BE DISCERNED ON
THE VIEWS.
INCORRECT W/L SETTINGS MAY RESULT IN
PATHOLOGIES AND OTHER ESSENTIAL ANATOMI-
CAL STRUCTURES NOT BEING DISPLAYED COR-
RECTLY, OR EVEN NOT BEING DISPLAYED AT ALL.
A SINGLE W/L SETTING CANNOT ALWAYS
CLEARLY DISPLAY ALL FEATURES PRESENT IN AN
EXAM. WHERE NECESSARY, USE SEVERAL DIF-
FERENT SETTINGS TO EXPLORE THE EXAM DATA.
ALSO NOTE THAT THRESHOLDING (SEE E.G.,
CHAPTER 5 AND CHAPTER 9) REMOVES ALL VOX-
ELS WITH VALUES OUTSIDE THE SELECTED
THRESHOLD RANGE FROM THE 3D VOLUME.
ANY PATHOLOGIES OR OTHER ANATOMICAL FEA-
TURES REMOVED IN THIS MANNER CAN NO
LONGER BE DISPLAYED, IRRESPECTIVE OF THE W/
L SETTINGS.

6-14
2271012-100.book Page 15 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Image Location
The location of the currently displayed images in terms of RAS coordinates is
shown by the image location annotations on the views. The main reference view
(acquisition view) also shows the index number of the current image.
These annotations are updated automatically whenever you move the 3D cursor on
the views.They are active annotations which means you can also modify them
directly to move to a specific location in the image set. See paragraph “Active
Annotations” above) on how to modify active annotations.
Note: If on an oblique view the image location annotation is marked (coi) this indi-
cates that it refers to the center of the image.
Alternatively, you can use the <left> and <right> cursor keys on the keyboard to
move through the image set.
Action of the <left>/<right> cursor keys depends on:
− View selection: the action of the key is determined by the primary view (red bor-
der). The view containing the mouse pointer becomes the primary view as soon
as you press a cursor key,
− View type: you move up/down on axial views, right/left on sagittal views, and
anterior/posterior on coronal views.
The position of the 3D cursor and the image location annotations are updated
whenever you use the <left>/<right> cursor keys.
Note: During batch filming (see Chapter 14) this function of the cursor keys is dis-
abled.

6-15
2271012-100.book Page 16 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Zoom
At startup, the DFOV (Display Field Of View) of the displayed images is the same
as those of the original acquisition (as shown in the Images list in the Browser).
You can “zoom in” on the images by reducing the DFOV, using the DFOV active
annotation on the view. See paragraph “Active Annotations” above on how to mod-
ify active annotations.
The maximum zoom factor (ratio between acquisition DFOV and actual DFOV) is
limited to 8.0.
You cannot “zoom out” (increase the DFOV beyond the original value), except if the
“height” of the stack of images (number of images x slice thickness) is larger than
the DFOV. In that case you can “zoom out” up to the height of the stack of images
(zoom factor < 1.0).

DFOV and filming

To simplify measurements on filmed images, it is obviously preferable that a filmed
image is either real size (scale 1:1) or scaled to a simple multiple or sub-multiple of
real size.
This means that the DFOV should preferably be set to a multiple or sub-multiple of
the size of the film slot used by the Film Composer.
For this reason, when you change the DFOV by using the left and right mouse but-
tons on the active annotation, the DFOV steps automatically to the nearest multiple
or sub-multiple of the user-defined FOV size of the film slot.
The default value is 13.0 cm. To change this value, select (Film/Save Tools) >
[Options] and modify the FOV numerical field.
Note: You will have to change this FOV value manually when you change the size
of the film slot used by the Film Composer. It is NOT updated automatically.

Roam (Scroll)
If you have “zoomed in” on an image (see above) and want to move to a part of the
image that you are interested in, you can use the orientation annotations (L, R, A,
P, S, I) on the sides of the images to roam (scroll) the image.
If not already active, these annotations become active (displayed in red) as soon as
you have zoomed in on an image.
Click and hold a mouse button on any of these annotations, and drag the image
around. The 3D cursor remains in the center of the image during this operation.

6-16
2271012-100.book Page 17 Thursday, May 8, 2003 11:30 AM

Volume Analysis

View Type
At startup, the type of view that is displayed in each of the four views of the viewing
area is determined by the build protocol (see Chapter 5). During image processing,
you can change the type of a view at any time:

• Use the (Display Tools) > [View Layout] option in the

control panel,
1 2
3 4 OR
Modify the view type active annotation in the top left cor-
ner on each view (click with the left or right mouse button
on the view type annotation and select the required view
type from the pop-up menu).

The following choices are available:

[3D]: 3D object (see Chapter 9),
[VR]: 3D object displayed using the Volume Rendering mode (see Chapter 11).
Only available if the Volume Rendering option is installed.
[Axial], [Sag.], [Coronal], [Oblique], [Curved]: acquisition and reformatted
images (see Chapter 8),
[Profile], [Histo.], [X Sect.]: measurement graphs (see Chapter 13),
[Navg.]: luminal views with viewpoint inside the 3D object (see Chapter 10). Only
available if the Navigator option is installed.
[Lumen]: Only used with the Advanced Vessel Analysis (AVA) option. Refer to AVA
User Guide, GE document ref. 2249101-100.

Patient Name
You can show or hide the patient name on the views. Click with the right mouse
button on the name and select [Show] or [Hide] from the pop-up menu.
You may want to hide the patient name on the views for increased confidentiality
while working on an exam, . If you have done so, make sure to show the patient
name again on the views BEFORE filming or saving images for diagnostic pur-
poses.

When filming or saving images for diagnostic purposes,

CAUTION always make sure the patient name is displayed on all
views.

6-17
2271012-100.book Page 18 Thursday, May 8, 2003 11:30 AM

Volume Analysis

On-View Traces
Several functions require the creation and editing of on-view traces. You can cre-
ate segment, curved and free-hand traces.

Creating and editing traces

• Select the type of trace: straight-line segments, free-hand or curved (splines) via
the (Display Tools) > [Preferences] panel. See paragraph “Type of Trace”
below.
• To create a segment or curved trace, you hold down the <Shift> key and click on
the view to deposit successive points of the trace.
To create a free-hand trace, you hold down the <Shift> key, click on the view to
start tracing and drag the mouse around to define the trace (while continuing to
hold down the <Shift> key). Also see paragraph “Creating a freehand trace”
below.
• To edit a trace, you hold down the <> key (“meta” key), then click and drag on the
red markers to move the original points, or on the green markers to split and
adjust segments between points.
• You can clear the last point of a trace, or the entire trace, via the on-view menu.
Click and hold right on the view, and select [Clear last point] or [Clear trace]
from the menu.
• If you want to create more than one trace on the views, select [Create trace] from
the menu to start defining a new separate trace.
(If no trace have been defined yet, this menu item has no effect.)
You can lock the 3D cursor to a trace.
• Click and hold right on the view, and select [Lock cursor to trace] from the
menu.
• To move the 3D cursor, hold down the <Shift> key and move the mouse pointer
as usual. The 3D cursor will follow the mouse pointer up and down, or left and
right, while remaining constrained to the trace.
• To unlock the cursor, again click and hold right on the view, and select [Unlock
cursor] from the menu.
Note the difference between moving the 3D cursor and creating a trace:
- To move the 3D cursor you hold down the <Shift> key and move the
mouse pointer, OR you click and hold the left mouse button on the 3D
cursor and drag it to a new position (as described in paragraph “3D
Cursor” above),
- To mark a point of a trace you hold down the <Shift> key and move the
mouse pointer, then click with the left mouse button at the position

6-18
2271012-100.book Page 19 Thursday, May 8, 2003 11:30 AM

Volume Analysis

where you want to deposit a point, while still holding down the <Shift>
key.
The easiest way to become accustomed to creating and editing traces is to practice
a few times on an exam with which you are familiar.
Type of trace
There are three types of trace that you can create:

• Segment trace: you define a series of points, and the

software connects these together with straight line seg-
ments,

• Free hand trace: the trace follows mouse movement dur-

ing creation. Creating a freehand trace differs slightly
from creating segment or curved traces. See below.

• Curved trace: you define a series of points, and the soft-

ware constructs a “best-fit” curve (spline) through these
points.

Use the buttons in the (Display Tools) > [Preferences] panel to select the type of
trace.
Note: If you switch the type of trace (e.g., from segment to curved) and a trace
already exists, the software recalculates the trace using the points you have
already defined.
Creating a freehand trace:
Creating a freehand trace differs slightly from creating segment or curved traces.
To create a free-hand trace:
• Holding the <Shift> key down, click (do not hold) the left mouse button to
activate tracing, then move the mouse to define your free hand trace while
still holding the <Shift> key down.
• To stop tracing, release the <Shift> key. To restart tracing, again hold the
<Shift> key down and click (do not hold) the left mouse button. A straight-
line segment is drawn from the end of the trace to the current mouse pointer
position. Again, move the mouse to define your trace.
Plane and 3D traces
A distinction should be made between plane traces and 3D traces (although the
procedure for creating and modifying them is identical).
A plane trace, as the name indicates, is created entirely in a single plane. This can
be an orthogonal (baseline) plane but also any oblique plane. A plane trace can be
used to define an area, e.g., for a measurement. If the trace is open (i.e., first and

6-19
2271012-100.book Page 20 Thursday, May 8, 2003 11:30 AM

Volume Analysis

last point are not the same), the software will normally close the trace for you by
adding a straight-line segment from the last point of the trace back to the starting
point.
A 3D trace is not constrained to a single plane. Successive points of the trace can
be located in different planes. Between placing successive points you can change
the position in the stack of images, switch to a different view, etc.
By its nature, a 3D trace cannot define an area; rather, it defines a trajectory in
three dimensions. As an example, you use a 3D trace to define a profile. A profile
graph shows the voxel values along a 3D trace (see Chapter 13, paragraph 3 for
details).
Notes
Bear in mind that traces (whether plane or 3D) are 3D objects, not just markings on
a view. Until erased, they are part of the 3D model, and will appear on all views (to
be precise: what is displayed is the projection of the trace onto the plane of each
view).
If one of the views is a 3D view, the 3D trace will also be displayed on this view.
While in theory you can also create or modify points on a 3D view, this is inadvis-
able in practice because on the 3D view you have no indication how “deep” into the
3D object the 3D cursor is located., unless you correlate continuously with the
baseline views. Also see paragraph “3D cursor on 3D views” on this subject.

6-20
2271012-100.book Page 21 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Other On-View Functions

Other on-view controls are only used for specific functions and are described in the
chapters that deal with that particular function.
This concerns, among others:
- Rotating the slice plane on an oblique view with the on-view “cube” (see
Chapter 8),
- Setting slice thickness and rendering mode on reformatted views (see
Chapter 8),
- Rotating a 3D object on a 3D view with the on-view “cube” (see Chapter
9),
- Setting rendering mode and defining cut planes on 3D views (see
Chapter 9),
- Setting up a batch (set of regularly spaced images to be filmed or
archived) by means of on-view controls (see Chapter 14).

6-21
2271012-100.book Page 22 Thursday, May 8, 2003 11:30 AM

Volume Analysis

4 THE REVIEW CONTROLLER

One of the new features of Volume Analysis 2 -which is going to exist from now on
in most software- is the existence of a Review Controller, which allows the user an
on-view operation of the main controls provided for the software. This section
describes the functions of each of these controls (shown on Illustration 1).

6-22
 2271012-100.book

Slice Thickness
Page 23 Thursday, May 8, 2003

Location Slider

Delete Current
11:30 AM

Bookmark
Measure Distances

Delete All Bookmarks Create Annotation

o to Previous Bookmark
Define Cuts
Create Bookmark Protocol±specific
Illustration 1 - THE REVIEW CONTROLLER

panel Enlarge

elete Bookmark Menu

Go to Next Bookmark
Select Number of Views
Jog Shuttle in the Loop
Auto Rock
Close Controller
Auto Loop Display Number of
Views in the Loop
Volume Analysis

6-23
2271012-100.book Page 24 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Reviewing the Slices

Different controls located on the Review Controller allow you to page through the
different slices, in order to locate a suspicious spot:
• {Image location slider}: Drag and drop the Location Slider with the left mouse
button to move within the different slices. Clicking once on the Location Slider, or
pressing the Control keyboard key, will link the Location Slider to the mouse cur-
sor (you can then move within the slices just by moving the mouse up and down).
Clicking again on the Location Slider, or releasing the Control keyboard key, then
unlinks the Location Slider from the mouse cursor.

When used on a VR model, with a Front Cut, the {Image

NOTICE
location slider} can be used to set the position of the Front
Cut.

• {Slice Thickness}: Click and drag the outer bars to modify the Slice Thickness
of the images within the active viewport.

Once you have linked the Location Slider to the mouse cur-
NOTICE
sor, you can still have access to both Window Level and
Slice Thickness functions:
To modify the Window Level without unlocking the Loca-
tion Slider, use the mouse middle button on the view. The
Window Level will be modified but the Location Slider will
remain locked on the current slide.
To modify the Slice Thickness without unlocking the Loca-
tion Slider, use the mouse right button to drag up and
down the Slice Thickness bars. The Thickness will be mod-
ified but the Location Slider will remain locked on the cur-
rent slide.

• {Jog shuttle}: Click on the horizontal Jog Shuttle to start paging at selected
speed in the selected direction. Clicking once will activate it. Another click then
deactivates it. Holding the left mouse button down will also produce similar
results.
• {Auto Loop}: The Loop mode allows you to have the views appear in a continu-
ous forward mode, in a loop movement : when the system reaches the last view,
it starts up again with the first one.

6-24
2271012-100.book Page 25 Thursday, May 8, 2003 11:30 AM

Volume Analysis

• {Auto Rock}: The Rock mode allows you to have the views appear in a continu-
ous forward to backward movement: when the system reaches the last view in a
forward movement, it starts up again backwards and vice versa.

Notice: When used on a VR model, the {Auto Rock} button

NOTICE
activates the Tumble mode for this model.

• {Select Number of Views in a Loop}: In both Loop and Rock modes you may
choose to view All the slices or to loop or rock through a certain number of views
around the current view (i.e. 5, 10, 25 or 50 views). Click on the up or down
arrow of this button to increase or decrease the number of views to be included
in a Loop or a Rock.
• {Display Number of Views in a Loop}: The number of views selected to be
included in the Loop or the Rock appears in the corresponding field (All = default
value of the field).

Managing Bookmarks
A bookmark is created when an area of interest has been located to allow the user
to more closely study this area .
• {Create Bookmark}: Creates a new Bookmark. Once you have clicked on the
button, bring the cursor on the view. It takes the shape of a pencil with which you
point on the interesting spot, before clicking with the left mouse button.
• {Go to Previous Bookmark}: Centers the focus on the previous bookmark.
• {Go to Next Bookmark}: Centers the focus on the next bookmark.
• {Delete Bookmark Menu}: The small black dot at the bottom of the Create-
Bookmark button allows to open, on the side, the Delete Bookmark Menu.
• {Delete Current Bookmark}: Deletes current bookmark. This function only
works when in Review Mode. This button is located on the Delete bookmark
Menu.
• {Delete all Bookmarks}: Deletes all bookmarks. This function only works when
in Review mode. This button is located on the Delete bookmark Menu.

Shortcuts
On the controller, a certain number of shortcuts to standard image management
tools, that can normally be accessed through tool menu buttons, are provided to
give the user an easier access to them:

6-25
2271012-100.book Page 26 Thursday, May 8, 2003 11:30 AM

Volume Analysis

• {Measure Distances}: Opens up the panel which allows you to execute distance
measures, for straight or curved lines. Refer to Chapter 13 “Measurements” or to
Chapter 10 “Navigator”).

CAUTION Measurements are more reliable when done on 2D

views; we recommend that you at least always check on
the 2D reformatted views where exactly the points have
been deposited.

• {Create Annotations}: Opens up the panel, which allows you to create pre-
defined annotations on the views. Refer to Chapter 12 “Annotations”.
• {Define Cuts}: Opens up the panel, which allows you to define any kind of cut
of the view you may need. Refer to Chapter 9 “3D Imaging”, Sections 3.3 and
3.4.
• {Enlarge}: enlarges the current view so that it takes up the entire viewing

area. The icon then becomes for returning the view to its normal size.

6-26
2271012-100.book Page 27 Thursday, May 8, 2003 11:30 AM

Volume Analysis

5 MAIN ON-VIEW MENU

Save image
The main on-view menu is displayed when
Shading parameters
you press and hold right anywhere on a view
Create trace
except on the red trackball control points, the
Clear last point
3D cursor, any of the red annotations or the ref-
Clear trace
erence image.
Lock cursor to trace The content of the menu depends on the view
Enlarge type and whether or not traces are present on
Lock orientation
the view. The illustration shows all the possible
options.
Center on FOV
Center on object
Center on cursor
Reset pointer

Restore Volume

[Save image] allows you to save the view for future display using the Image Works
Browser. The software assigns a name to the saved view that appears in the Image
Works Browser.
[Shading parameters] (3D views with surface shading only) allows you to modify
how a 3D view with surface shading is rendered. See Chapter 9 “3D Imaging”,
paragraph “3D Shading”.
[Create trace] starts creation of a new trace.
[Clear last point] clears the last point entered on a trace. [Clear trace] clears the
entire trace, enabling you to restart trace entry. These menu items exist only if a
trace is present.
[Lock cursor to trace] constrains the 3D cursor to move only along the trace. The
menu item is then replaced by [Unlock cursor] for restoring free movement to the
3D cursor. This menu item exists only if a trace is present.
[Enlarge] (not available on profile, histogram or cross section views) enlarges the
current view so that it takes up the entire viewing area. The menu item then
becomes [Reset size] for returning the view to normal size.
[Lock orientation] (3D and oblique views only) locks view orientation (i.e., pre-
vents rotation). The menu item then becomes [Unlock orientation] for restoring
rotation ability.
[Center on FOV] (3D views only) centers the 3D model in the field of view.

6-27
2271012-100.book Page 28 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Note: This operation centers the entire model regardless of whether or not parts of
it are invisible on the screen due to operations such as cutting, thresholding,
removing, etc.
[Center on object] (3D views only) centers the object defined by the current 3D
cursor position in the field of view. This is the default selection.
[Center on cursor] centers a zoomed image on the current 3D cursor position.
[Reset pointer] (not on profile, histogram, cross section or curved views) centers
the 3D cursor in the 3D field of view.
[Restore Volume] Reverses the view to the originally loaded image.

6-28
2271012-100.book Page 29 Thursday, May 8, 2003 11:30 AM

Volume Analysis

6 SMART CURSOR
When the 3D cursor is located in airways or in the colon, it will be possible for the
user to move along the structure by pressing one key of the keyboard as follows:
• When pressing the F key (Forward), the cursor will move forward within the
structure.
• When pressing the B key (Backward), the cursor will move backward within the
structure.
When doing this, the cursor is locked on the center of the structure.
Note: This feature also works with other tubular structures, but because the con-
trast in those structures is not well defined, the tracking is, in this case, much
less reliable.
Note: This feature may only be used with CT datasets.

6-29
2271012-100.book Page 30 Thursday, May 8, 2003 11:30 AM

Volume Analysis

7 REFERENCE IMAGE
Reference images (orientation indicators) have two main functions: .
• Keep track of the orientation of a slice plane or view direction with respect to the
patient’s body
• Enable the user to simplify the paging operation of all 2D views and the adjust-
ment of orientation of Oblique reformatted views.
At times, you may want to momentarily hide the reference images on all the views
(e.g. to better examine a detail on a view, or to film without reference images). To
do so, select (Display Tools) > [Preferences] and switch off the (Reference
Image) control.
This setting is memorized by the software and remains active from one session to
another. To return the reference images to the views, again select (Display Tools)
> [Preferences] and switch on the (Reference image) control.
Note: You can modify the window W/L of a reference image independently. Click
and hold the middle mouse button on the reference image, then move the
mouse in the same manner as on the view itself.
Note: This function can also be executed using the [Reference Image] function of
the main on-view menu, which allows you to hide the reference image, but
only that of the current view (refer to Chapter 5 “Display and Controls”, Sec-
tion “Main On-View Menu”).

6-30
2271012-100.book Page 31 Thursday, May 8, 2003 11:30 AM

Volume Analysis

View Orientation of a Slice Plane

On baseline and oblique reformatted

views, the reference image shows, in
perspective, the intersection between
one of the baseline images (axial, coro-
nal or sagittal) and the slice plane dis-
played in the view (see illustration).
On a 3D view, an arrow on the reference
image shows the orientation of the cur-
rent viewing direction with respect to the
reference image (not illustrated).

A pop-up menu (click and hold left or right on the reference image) allows you to:
• Select which baseline view is used as a reference (axial, sagittal or coronal),
• Move the reference image to another corner of the view,
• Reset the focal point, i.e., switch the baseline image used for the reference
image to the one that matches the current 3D cursor location.

6-31
2271012-100.book Page 32 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Paging and adjustment of orientation

On all reference images of 2D views, the intersection line is active and allows you
to quickly page through the images, if a great accuracy is not needed.
On the reference images of the Oblique reformatted view, pointing the mouse cur-
sor at the end of the Intersection lines allows you to adjust the orientation of the
concerned view.

Place the cursor on the

Intersection between the view plane
Intersection line (I).
and the reference icon plane(E) Green double arrows
appear showing you
Front line of the schematic the possible directions
representation of the view plan (F)
of the movement (up or
down).
Place the cursor on the
Front line (F). Green
double arrows appear
End of Intersection line (E) showing you the possi-
ble directions of the
movement (front or
back rotation around
the intersection).
Place the cursor at the
end of the Intersection
line (E). Green double
arrows appear showing
you the possible direc-
tions of the mouvement
(left or right rotation of
the intersection line).

6-32
2271012-100.book Page 1 Thursday, May 8, 2003 11:30 AM

CHAPTER 7 Volume Analysis

CHAPTER 7 - PROCESSING

We can consider a CT exam as the representation in digital form of a three-

dimensional volume that can be loaded into computer memory. Various 3D pro-
cessing techniques can now be used to “look” at this three-dimensional volume.
Reformatting allows you to define and display cross-sections of the 3D volume
that are oriented differently from the original acquisition images.
3D imaging allows you to extract a region or anatomical feature of interest from
the 3D volume (volume segmentation), then view, tilt and rotate the resulting “3D
object” (or objects) on the screen.
With the Navigator option you can also move the viewpoint to locations inside
“hollow” structures, providing real-time luminal views.
The Volume Rendering option uses the notion of ”opacity” to display several lay-
ers of tissue in a 3D object simultaneously.
The choice of the processing technique to be used for a given exam is made by
the physician. The criteria for this choice are obviously outside the scope of this
User Guide.
Processing can be performed either by using the ”tools” of the application
directly, or (in particular for routine exams) by using the pre-defined protocols
and guides supplied with the application.

7-1
2271012-100.book Page 2 Thursday, May 8, 2003 11:30 AM

Volume Analysis

1 OVERVIEW
We can simply consider a CT exam as a stack of pictures representing successive
cross-sections through the anatomy of the patient.
In CT exams the successive cross-sections are parallel (slices).
Simple diagnostics can be performed by examining those images that contain
cross-sections of an anatomical feature of interest.
We can also think of a CT exam as the representation of a three-dimensional vol-
ume. Such an exam in digital form (image set) can be loaded into computer mem-
ory.
Volume Analysis 2 can load an CT image set (parallel slices) and construct a “3D
model” from the data. It can load a 3D model generated by the separate AW 4.1
application.
Various 3D processing techniques can now be used to “look” at this three-dimen-
sional volume, not only by examining the original cross-sections, but also in other
ways.
• Reformatting allows you to define and display cross-sections of the 3D volume
that are oriented differently from the original acquisition images. See Chapter 8,
• 3D imaging allows you to extract a region or anatomical feature of interest from
the 3D volume (volume segmentation), then view, tilt and rotate the resulting “3D
object” (or objects) on the screen. See Chapter 9,
• With the Navigator option you can even move the viewpoint to locations inside
the 3D volume. You can display the inside of “hollow” structures, providing real-
time luminal views, but you can also view the outside of “solid” structures in
close-up and in perspective. See Chapter 10.
• The Volume rendering option allows you to display several anatomical features
in a 3D volume simultaneously by assigning them varying degrees of opacity.
See Chapter 11.
The Volume Analysis 2 application provides you with two options for your 3D
image processing requirements:
• On the one hand, Volume Analysis 2 presents you with a full “tool kit” that con-
tains the controls and functions that you need to analyze an individual exam in
detail. See paragraph 2.
• On the other hand, for standard exam processing, you can use the set of pre-
defined protocols and processing guides provided with Volume Analysis 2. See
paragraph 3.

7-2
2271012-100.book Page 3 Thursday, May 8, 2003 11:30 AM

Volume Analysis

2 USING THE VOLUME ANALYSIS 2. TOOLS

When dealing with routine exams you will mostly use the standard pre-defined pro-
tocols and processing guides provided with Volume Analysis 2 (see paragraph 3).
At times, however, you may be confronted with special situations where none of the
standard protocols are applicable to analyze a particular exam in full detail. At such
times, you can access the full set of “tools” of the Volume Analysis 2 application
directly via the main control panel, the menus, and the on-view controls as
described in this User Guide.
The primary purpose of this User Guide is to familiarize you with all functions and
options available in the Volume Analysis 2 application so as to use the application
with optimum efficiency.

7-3
2271012-100.book Page 4 Thursday, May 8, 2003 11:30 AM

Volume Analysis

3 USING THE VOLUME ANALYSIS 2 PROTOCOLS

Overview
If you use 3D processing routinely to analyze a particular type of exam, setting up
and performing the successive processing steps each time anew for each exam
can become time-consuming. Also, any minor mistake, such as omitting a step in
the processing sequence or setting a threshold to the wrong value, may mean that
you have to restart the entire procedure from the beginning.
This is where you can use the set of pre-defined protocols provided with Volume
Analysis 2. Protocols combine detailed processing instructions with the necessary
controls and functions in the same control panel, thereby eliminating much of the
data entry, command selection, etc. required for processing an individual exam.
Note: In Volume Rendering views, the name of the protocol is displayed as a Sys-
tem Annotation on the bottom left-hand side of the view.
• A build protocol defines how the 3D model is built and how it will be displayed ini-
tially. Build protocols are described in Chapter 5.
• A processing protocol consists of a build protocol combined with a sequence of
processing steps to perform a standard task. Processing protocols are
described below.
• A guide is a small additional protocols used for specific tasks such as switching
the view layout or performing certain measurements,
• A batch protocol defines a new image set that can be filmed or saved on the
image disk for later viewing, or displayed as a "movie" (animated sequence).
Such an image set can consist of parallel or radially reformatted images, or a
sequence obtained by rotating a 3D object in steps. Batch protocols are
described in Chapter 14.
Note: The “Volume Rendering” option uses its own specific set-up protocols.
Refer to Chapter 11 for details.

7-4
2271012-100.book Page 5 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Guide protocols
A Guide protocol is usually invoked at the start of a Volume Analysis 2 session,
but during a session you can call up a new protocol from the main control panel,
and call up additional guides.
A Guide protocol consists of one or more control panels, that are displayed when
you invoke it.
For each step in the protocol sequence all necessary tools, such as command but-
tons, sliders, data entry windows, etc. are available in the protocol window together
with the instructions on how to perform the sequence.
You step through the protocol sequence by means of the (Next) button in each
panel. You can move back in the sequence, e.g., to check a setting, by means of
the (Back) button
Note: While you can move back in a protocol sequence and make corrections, this
may lead to unexpected results. Use the (Undo) button to undo the pro-
cessing steps as required, then start the protocol again from the point where
you require a correction.
You will note that a Guide protocol is not an “automated” sequence.
As an example, the protocol may request you to set a threshold to select an ana-
tomical feature. The protocol window will contain the threshold slider, initially set to
an appropriate default value, and a description of the current step, but it is up to you
to set the threshold correctly as a function of the image content.
At the same time, all on-view tools and tools from the “toolbox” (that you access via
the control panel) remain fully active. You can zoom in on a feature, change win-
dow width or level, add an annotation, etc. while working through the steps of a pro-
tocol.
You can also invoke the additional guides. These are small additional protocols
used for specific tasks such as switching the view layout or performing certain mea-
surements, that you can call up without closing the main protocol.
By their nature, the Guide protocols are largely self-documenting, hence the con-
tents of the individual protocols supplied with the Volume Analysis 2 application
are not described in any detail in this User Guide. The controls and functions avail-
able in the protocol control panels are the same as those in the Volume Analysis 2
“tool kit”. For their description refer to the relevant chapters and paragraphs in this
User Guide.

7-5
2271012-100.book Page 6 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Blank page.

7-6
Chapter8.fm Page 1 Friday, June 27, 2003 10:07 AM

CHAPTER 8 Volume Analysis

CHAPTER 8 - REFORMATTING

A CT image set represents a 3D volume, defined by successive cross-sections

through the anatomy of the patient.
Reformatting allows you to define and display cross-sections of the 3D volume
that are oriented differently from the original acquisition images.
A baseline view is a basic axial, coronal or sagittal view. Of these, the acquisi-
tion view displays the images in the acquisition plane of the original image set,
the other two are the corresponding orthogonally reformatted views. They can
be moved to show any location in the 3D volume, but remain aligned parallel to
the three main axes of the RAS coordinate system.
An oblique view is a plane reformatted view that can be both moved and rotated
to any location and orientation within the 3D volume.
If a feature of interest extends significantly in three dimensions, a single stan-
dard baseline or oblique view cannot show the full extent of the feature. To over-
come this, you can use curved reformatting or MPVR; alternatively you can
create a batch of regularly spaced reformatted images.
With a curved view, instead of using a plane cross-section, you “unfold” a curved
cross-section of the 3D volume, that conforms as much as possible to the shape
of the feature of interest.
With MPVR (Multi-Planar Volume Reconstruction), instead of using baseline and
oblique views representing slices that are only one voxel thick, you define a
“thick” slice that encompasses the feature of interest.

8-1
Chapter8.fm Page 2 Friday, June 27, 2003 10:07 AM

Volume Analysis

1 BASELINE REFORMATTING
The baseline views are the basic axial, coronal and sagittal views. One of these is
the acquisition view (images in the acquisition plane of the original image set). The
other two are the corresponding orthogonally reformatted views.
Usually, these three baseline views are already displayed at startup, after you have
selected a protocol and the 3D model has been built (see Chapter 5).

1 2 Otherwise, to display the baseline views, select [Axial], [Coro-

3 4 nal] or [Sag.] in the View type active annotation menu, or use
the Display Options > View Layout panel.

To move to any given location in the image set, move the 3D cursor on any of the
views. The other two views are updated in real time: the point of intersection of the
three baseline views always corresponds to the location of the 3D cursor.
Note: This assumes the views are selected. Views that are not selected are not
updated.
The current location in the image set is indicated by the on-view annotations (RAS
coordinates on all three views, and additionally image index number on the acquisi-
tion view). These are active annotations, and can be used to move through the
image set in the same manner as the 3D cursor (see Chapter 6, paragraph “Active
Annotations”).
The reference images (normally in the bottom right corner of the views) provide you
with a further means of keeping track of the currently displayed location relative to
the image set and the patient's body (see Chapter 6, paragraph 7 “Reference
Images”).

8-2
 Chapter8.fm

View type
Patient name
annotation
annotation
Slice position (Annotation control)
and control
Image number
and control
(original series
Page 3 Friday, June 27, 2003 10:07 AM

slices only)

Field of view Orientation

(zooming control) annotations
(Roam control)

Reference
Thickness Image
(orientation and
paging control
Rendering mode

Window Axialviewwith
width/level annotations
annotations andcontrols

8-3
Volume Analysis
Chapter8.fm Page 4 Friday, June 27, 2003 10:07 AM

Volume Analysis

2 OBLIQUE REFORMATTING
An oblique view is a reformatted view that can be both moved and rotated to any
position within the 3D volume (the baseline views can be moved to show any loca-
tion, but remain aligned parallel with the three main axes of the RAS coordinate
system).
To display an oblique view together with the three baseline views at startup, select
the Reformat protocol.
Otherwise, to display an oblique view, select [Oblique] in the View type active
annotation menu, or use the Display Options > View Layout panel.
Note: Initially the oblique view is aligned with one of the reformatted baseline
views (usually the coronal view).

Controls
To move the oblique view to any given location in the image set, move the 3D cur-
sor on any of the views in the same manner as for the baseline views (see para-
graph 1).
To rotate the oblique view (i.e. rotate the plane of intersection with the 3D volume
around the 3D cursor position), use the on-view oblique and tilt/rotate modes, or
the (Rotate/Translate) controls.
The controls are complementary:
- -Use the 3D cursor to move to the location of interest,
- Use oblique mode to obtain an initial alignment with the feature of
interest,
- Use tilt/rotate mode with the on-view “cube” for final alignment,
- Use the (Rotate/Translate) controls to move the 3D cursor or the
orientation of the oblique plane in precise steps, or the (S) (I) (A) (P) (L)
(R) buttons on the main control panel to return the oblique view to a
specific initial orientation.
The reference image (normally in the bottom right corner of the view) indicates the
currently displayed location and orientation of the oblique view relative to the image
set and the patient’s body (see Chapter 6, paragraph 7 “Reference Images”).

8-4
Chapter8.fm Page 5 Friday, June 27, 2003 10:07 AM

Volume Analysis

On-view oblique mode

• Select (Oblique) mode on the control panel.

• Select one of the baseline views as primary.
- The oblique view is aligned so as to be
perpendicular to the primary view,
- A solid line on the primary view indicates the
intersection with the oblique view,
- Thin dotted lines indicate the intersection with the
baseline views.
• You can now rotate the oblique plane around the 3D cur-
sor (i.e. around an axis perpendicular to the baseline
view) by clicking and dragging on the solid line on the
primary view. The oblique view is updated at the same
time.

Note: The software rotates the oblique view so as to be as close as possible to one
of the baseline orientations: this means that the view may jump when it is
halfway between two orientations.
In oblique mode, you cannot perform two or more successive rotations in different
planes by selecting first one baseline view, then another. As soon as you select
another baseline view in oblique mode, the oblique plane is realigned to be perpen-
dicular to that baseline view, and you lose your previous settings.
If you want to perform rotations in different planes, you can either use oblique mode
for the first alignment and then switch to tilt/rotate mode, or create a second oblique
view as described in the paragraphs below.

8-5
Chapter8.fm Page 6 Friday, June 27, 2003 10:07 AM

Volume Analysis

On-view tilt/rotate mode

• Select (Tilt/Rotate) mode on the control panel.

In this mode, as soon as you move the mouse pointer
onto the oblique view, a red “cube” symbol with “han-
dles” in the corners and on the sides appears on the
view.

You can now use two different methods to rotate the oblique plane.
To rotate and tilt the oblique plane in any direction:
• Click and hold with the left mouse button on one of the corners of the cube. If the
view wasn’t selected before, it is now selected.
• Still holding the mouse button down, drag the mouse around. The oblique plane
will start to rotate with mouse movement. Keep holding down the mouse button
until the oblique plane has rotated to the new position. When the desired orien-
tation is reached, release the mouse button.
To constrain the rotation of the oblique plane to an axis parallel to one of the edges
of the cube:
• Press and hold with the left mouse button on the rotation marker on the middle of
an edge of the cube parallel to the axis around which you want to rotate the
oblique plane. If the view wasn’t selected before, it is now selected.
• Still holding the mouse button down, drag the mouse around. The oblique plane
will start to rotate with mouse movement, but it will rotate only around the axis
parallel to the selected edge. Keep holding down the mouse button until the
oblique plane has rotated to the new position. When the desired orientation is
reached, release the mouse button.
Note: You use the same cube to rotate a 3D model on a 3D view (see Chapter 9).

8-6
Chapter8.fm Page 7 Friday, June 27, 2003 10:07 AM

Volume Analysis

Rotation and translation

You can use the (Rotate/Translate) controls to change the orientation of the
oblique plane or move the 3D cursor in precise steps. You can also return the
oblique plane to one of the baseline orientations.
• Select the oblique view by clicking on it (red border).
To rotate the oblique plane in steps:
• Select (Rotate/Translate) > (By deg),
• In the numerical entry field, enter the required angle step in degrees,
• Click on the arrow button corresponding to the desired direction of rotation.
To move the 3D cursor in steps:
• Select (Rotate/Translate) > (By mm),
• In the numerical entry field, enter the required distance step in millimeters,
• Click on the arrow button corresponding to the desired direction of translation.
If after repeated rotations you want to rapidly return the oblique view to a known
baseline orientation:
• Use the (S) (I) (A) (P) (L) (R) buttons on the main control panel.
Note: To return the 3D cursor to the initial position at the center of the 3D volume,
select [Reset pointer] in the on-view menu.

8-7
Chapter8.fm Page 8 Friday, June 27, 2003 10:07 AM

Volume Analysis

Using two oblique views

As noted here above, in oblique mode you cannot perform two or more successive
rotations in different planes by selecting first one baseline view, then another. As
soon as you select another baseline view in oblique mode, the oblique plane is
realigned to be perpendicular to that baseline view, and you lose your previous set-
tings.
What you can do, however, is first to set up and rotate an oblique view perpendicu-
lar to a baseline view as described, then create a second oblique view that you now
can rotate around an axis perpendicular to the first oblique view.
• Set up and rotate an oblique view perpendicular to a baseline view as described
in paragraph “On-view oblique mode” above,
• Create a second oblique view, either by selecting [Oblique] in the View type
active annotation menu, or using the Display Options > View Layout panel,
• Select the first oblique view as primary,
- The second oblique view is now aligned so as to be perpendicular to the
primary view,
- The solid line on the primary view indicates the intersection with the
second oblique view,
• As before, you can now rotate the plane of the second oblique view around the
3D cursor (i.e. around an axis perpendicular to the first oblique view) by clicking
and dragging on the solid line on the primary view. The second oblique view is
updated at the same time.

8-8
Chapter8.fm Page 9 Friday, June 27, 2003 10:07 AM

Volume Analysis

3 CURVED REFORMATTING
The standard baseline and oblique views (see paragraphs 1 and 2) represent plane
cross-sections of the 3D volume.
If a feature of interest is not restricted to a single plane, plane reformatting cannot
show the entire feature: whichever way you position the oblique plane, only part of
the feature is shown at any one time.
To create a single view that includes the entire feature, you can use different tech-
niques:
- Curved reformatting: instead of a plane cross-section you can create a
curved cross-section, as described below in this paragraph.
- MPVR (Multi-Planar Volume Reconstruction): instead of using the
standard baseline and oblique views that represent slices of the 3D
volume that are only one voxel thick, you define a “thick” slice that
encompasses the feature of interest (see paragraph 4 “MPVR”),
- 3D imaging: you “extract” the feature of interest and display it as a 3D
object (see Chapter 9, “3D Imaging”).
Alternatively, you may at times prefer to create a set of evenly spaced reformatted
views, positioned and oriented so as to show the feature as clearly as possible. For
this you can use the “batch” function (see paragraph 5 “Batch Reformatting”).
To create a curved reformatted view:
• First select the view where you want to display the curved reformatted view. On
this view, select [Curved] in the View type active annotation menu, or use the
Display Options > View Layout panel. Initially the view will be displayed in
blue, with a message indicating “curve not defined”.
• Select the “definition” view on which you will define the curve (this can be one of
the baseline views, or an oblique view), and create a trace on this view.
As soon as you have defined the first segment of the trace, the curved view
shows a reformatted view that corresponds to a plane that passes through the
trace and is perpendicular to the definition view.
• Continue defining the trace: the curved view is updated each time you add a seg-
ment to the trace. You can make corrections by editing the trace: again the
curved view is updated each time.
For full details on how to create and edit a trace, refer to Chapter 6. In brief:
- To place successive points of the trace, hold down the <Shift> key and
click,
- To edit the trace, hold down the <> key (“meta” key), then click and drag
on a red marker to move an original point, or on a green marker to split
a segment.

8-9
Chapter8.fm Page 10 Friday, June 27, 2003 10:07 AM

Volume Analysis

Note: When filming or saving the result of a curved reformatting operation, always
include the view on which you have defined the trace in the record. Without
this information it is impossible to interpret a curved reformatted view cor-
rectly.

8-10
Chapter8.fm Page 11 Friday, June 27, 2003 10:07 AM

Volume Analysis

4 MPVR
Overview
As noted already in paragraph 3 (“Curved Reformatting”), most anatomical features
are not “flat” but extend into the third dimension.
Curved reformatting is one available technique to overcome this problem. How-
ever, baseline, oblique and curved reformatted views still represent cross-sections
of the 3D volume, or, in other words, they represent slices through the 3D volume
that are only one voxel thick. As a consequence, you may still need to move the
slice plane to different positions to examine the full extent of a feature.
An alternative is to use MPVR (Multi-Planar Volume Reconstruction). This is a
technique that allows you to generate a “thick” slice that encompasses the feature
of interest while still excluding unwanted tissue. Mostly, you will be using MPVR on
oblique views, but it can also be used on the baseline views.

MPVR mode
If an oblique view and/or the baseline views are already displayed, you switch to
MPVR by increasing the slice thickness active annotation on the view to a value
greater than the base value (one voxel). As with the other active annotations, click
left on the annotation to increase it in steps, click and drag on it to change it in a
continuous manner, or place the mouse pointer on it and enter the new value from
the keyboard.
Alternatively, select the MPVR protocol at startup to display an MPVR oblique view
together with the three baseline views.
Note: To return a view to a “thin” slice (disable MPVR), return the slice thickness
active annotation to its minimum value, either by clicking and/or dragging on
the annotation, or by pointing on the annotation and entering 0 (zero) from
the keyboard.
When you use oblique mode, parallel dotted lines on either side of the reference
lines that show the intersection between the view planes indicate the MPVR slice
thickness.
In particular, these dotted lines are displayed on either side of the solid line on the
primary view that shows the intersection between the oblique plane and the primary
baseline view.
They indicate the thickness of the oblique slice, and by clicking and dragging on
them you can directly match the slice thickness of the oblique view to the dimen-
sions of a feature of interest visible on the primary view.
You can also modify the slice thickness by dragging the outer bars of the Location
Slider of the Review Controller (refer to Chapter 6, Section 4 “The Review Control-
ler”).

8-11
Chapter8.fm Page 12 Friday, June 27, 2003 10:07 AM

Volume Analysis

Rendering
While the image of a “thin” slice simply shows the voxel values, the image of a
“thick” slice is a projection of the voxel values across the thickness of the slice. Dif-
ferent options are available of how the voxel values across each projection will be
rendered (displayed), i.e., how the pixels making up the image will be constructed
from the voxel values:
MIP (Maximum Intensity Projection): the pixel value is the maximum voxel value
along the projection,
MinIP (Minimum Intensity Projection) : the pixel value is the minimum voxel value
along the projection,
Average: the pixel value is the average of the voxel values along the projection
(sum of the voxel values divided by the number of voxels),
VR (available only if the Volume Rendering option is installed): the pixel value is
obtained by considering the voxel values along the projection as representing a
varying degree of opacity. See Chapter 11 “Volume Rendering”.
Select the required option by using the rendering mode active annotation which
appears next to the slice thickness annotation as soon as you have set the slice
thickness to a value greater than one voxel.
Because each pixel (picture element) of an MPVR view has to be computed from
the voxel values across the thickness of the slice, MPVR views are updated much
more slowly than “thin slice” oblique and baseline views. Beyond a certain slice
thickness setting, the software at first uses a low resolution display while you are
setting up and adjusting an MPVR view, then shows a message: “Hit spacebar to
display high definition view”.
This allows you to set up the MPVR view much more rapidly, then view the final
result recomputed in high resolution by pressing the spacebar on the keyboard.
Note: MPVR can be used on a baseline view at any time, regardless of what other
views (oblique, 3D, Navigator, etc.) are displayed.
MPVR cannot be used on curved reformatted views.
MPVR does not supply the same degree of depth cues as full 3D imaging. How-
ever, it can often provide the required information on an anatomical feature rapidly
by using the right oblique plane orientation, slice thickness and rendering mode,
without having to resort to full 3D image processing (thresholding, volume segmen-
tation and 3D object manipulation).

8-12
Chapter8.fm Page 13 Friday, June 27, 2003 10:07 AM

Volume Analysis

5 BATCH REFORMATTING
Curved reformatting and MPVR (see above) allow you to create a single view
showing a maximum of information concerning a feature of interest that extends
significantly in three dimensions.
At times, you may prefer to create a set of reformatted images that intersect the
feature of interest, positioned and oriented so as to show the feature as clearly as
possible by means of several regularly spaced cross-sections.
The batch function allows you to rapidly define such a set, either of parallel oblique
images or radial oblique images, with full control over position, orientation, spacing
and number of images in the set.
A batch image set can be viewed step by step or as a “movie loop”, it can be filmed,
and it can be saved on the image disk as a new series in the exam.
Refer to Chapter 14 “Recording” for information on how to set up, view and record a
batch image set.

8-13
Chapter8.fm Page 14 Friday, June 27, 2003 10:07 AM

Volume Analysis

Blank page.

8-14
2271012-100.book Page 1 Thursday, May 8, 2003 11:30 AM

CHAPTER 9 Volume Analysis

CHAPTER 9 - 3D IMAGING

A CT image set represents a three-dimensional (3D) volume, defined by succes-

sive cross-sections through the anatomy of the patient.
3D imaging allows you to extract a region or anatomical feature of interest from
the 3D volume (volume segmentation) and display the result as one or more 3D
objects.
You use thresholding to extract a 3D object by selecting a range of voxel values
that represents a specific tissue or anatomical feature. The scalpel tool allows
you to perform “cuts” in the 3D volume to define a regio of interest. The paint
tools allow you to mark a region of interest with colored “paint”, then display only
this region.
Different rendering modes define how the resulting 3D object will be displayed:
− Surface shading treats the 3D object as if it were fully opaque.
− Projection shading treats the 3D object as if it were translucent. You can
select how the voxel values across the depth of the 3D object will be dis-
played (maximum intensity, minimum intensity, average, etc.).
− Volume rendering (option) treats the voxel values along the the depth of the
3D object as representing a varying degree of opacity.
You can rotate a 3D object to examine it from all angles, define cut planes, use a
spherical “shutter” to show use color to distinguish these separate 3D models.

9-1
2271012-100.book Page 2 Thursday, May 8, 2003 11:30 AM

Volume Analysis

1 3D OBJECT
3D processing techniques allow you to isolate an anatomical feature in a 3D vol-
ume, then view, tilt and rotate the resulting “3D object” (or objects) on the
screen.
Depending on what you want to see, you can apply different types of treatment
to the image, so as to highlight such or such structure within the 3D object:
If we display an entire exam (a stack of slices) as a solid 3D object all we see is
the outside of the body part in question (head, chest, abdomen, etc.), and extra-
neous items such as headboards.
To display a specific feature “buried” somewhere inside, we first have to define
what part of the exam data should be visible, and what part shouldn't. The main
tools you will use for this are:
• Thresholding: you extract a region of interest by selecting a range of voxel
values that represents a specific tissue or anatomical feature,
• Scalpel: you perform “cuts” in the 3D volume to define the region of interest,
• Paint: you mark the region of interest with colored “paint”, then display only
this region.
If we display the entire exam as a translucent 3D object (using for instance the
MIP rendering mode, see paragraph 2) we can see through the body part, but to
clearly display a specific feature without interference from other anatomy and
extraneous items such as headboards, we still have to define what part of the
exam data should be visible, and remove the remainder, using the same tools as
before.
The process is sometimes referred to as “volume segmentation” because the 3D
volume is segmented, or split, in two parts: the volume of interest that is cur-
rently displayed, and the remainder that is removed from view.
In the context of 3D processing, 3D model refers to the representation of the
exam data set in the workstation computer memory (data may have been added
or removed, see Chapter 5 “Building the 3D Model”). 3D volume usually refers
to the currently displayed part of the 3D model. At times the two terms may be
used interchangeably.
After volume segmentation, the displayed part of the 3D model consists of one
or more 3D objects. A 3D object is a part of the 3D model that is separate from
other parts, or, more precisely, two 3D objects are separate if there is every-
where at least one voxel width of empty space between them.
Each 3D object can be selected separately, by placing the 3D cursor on it; you
can then either keep the object (and remove everything else), or remove the
object (and keep everything else).
Occasionally this may lead to surprises, as when two apparently separate
objects still act as one, because they are still connected somewhere by a

9-2
2271012-100.book Page 3 Thursday, May 8, 2003 11:30 AM

Volume Analysis

“bridge” of voxels, or when an apparently single object turns out to consist of two
or more parts, separated by narrow gaps. Tools such as the “Open Bridges” and
“Close Gaps” functions can help you to deal with these effects.
An important point to note here: when part of the 3D model is removed from
view, it is not removed from the 3D model in the workstation memory. A “Show
Removed” function allows you at any time to display everything you have
removed from view earlier. This then becomes the current 3D view, and it is the
earlier feature of interest that is removed from view.
This means that you can use the segmentation tools (thresholding, scalpel, paint
tools) not only to select an anatomical feature such as a particular organ that you
want to view and remove everything else, but also to select a particular feature
that you want to remove, while keeping the rest of the 3D volume.
Note: You can do this by first selecting the feature to be removed using the seg-
mentation tools (thresholding, scalpel, paint tools) and, once you are sat-
isfied with your selection, use the “Show Removed” function, which will
now show you everything in the 3D volume except the feature you
wanted to remove.
A typical example is bone in a CT angiography exam that tends to
obstruct certain regions of interest (in particular if you are using the MIP
rendering mode, see paragraph 2). Rather than trying to isolate the vas-
culature (and possible calcifications) within the 3D volume, which can be
laborious, you may find it faster to select the obstructing bone, and only
the bone, using the segmentation tools (thresholding, scalpel, paint
tools), and then use the (Show Removed) function, which will show you
everything except the obstructing bone.

9-3
2271012-100.book Page 4 Thursday, May 8, 2003 11:30 AM

Volume Analysis

2 RENDERING
Overview
In general terms, “rendering” refers to the techniques used to represent a 3D
(three-dimensional) object on a two-dimensional surface.

As a example, if a sphere is projected on a flat surface, the

result is a circle. Any indication of it being a 3D object is lost.
The perception of depth can be restored to the projection with
rendering techniques.

In the simple illustration at the left, a few levels of shading are

sufficient to provide depth cues and restore the impression of a
sphere. Here, the shading represents the amount of light that
would be reflected from a light source above and to the left of
the sphere.

In medical imaging, it will be obvious that the complexities of shape, spatial ori-
entation and tissue properties require more sophisticated display and rendering
techniques than in the simple illustration above.
The main rendering techniques used by the Volume Analysis 2 software are
surface shading, projection shading and volume rendering (option).
Additional depth cues are supplied by rotating the displayed 3D object. See
paragraph “Rotating the 3D Model” bellow.

9-4
2271012-100.book Page 5 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Rendering Techniques
The main rendering techniques used in Volume Analysis 2 are:
• Surface shading,
• Projection shading,
• Volume rendering (option).
They are described in more detail in the paragraphs below.
You select the required option by using the rendering mode active annotation.
Note: The rendering mode determines how the 3D volume is displayed on the
screen. The different rendering modes provide widely different ways of
displaying the same 3D volume.
The choice of a given rendering mode is determined by the type of exam,
type of diagnostics, anatomical feature to be studied, etc. and is outside
the scope of this user guide.

A 3D VIEW IS A TWO-DIMENSIONAL PROJECTION

WARNING! ON THE SCREEN OF THE 3D VOLUME. THERE IS
NO INDICATION ON A 3D VIEW OF HOW “DEEP”
INSIDE THE 3D VOLUME THE 3D CURSOR IS
LOCATED.
TO PREVENT MISINTERPRETATION OF THE CUR-
SOR LOCATION ON THE AXIS PERPENDICULAR TO
THE VIEWING POSITION, THE USER SHOULD
ALWAYS VERIFY THE CURSOR POSITION BY COR-
RELATION WITH THE BASELINE AND REFORMAT-
TED VIEWS.

9-5
2271012-100.book Page 6 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Surface shading
Surface shading treats the 3D object as if it were fully opaque. The shading
value for a voxel is defined by the local orientation of the surface at the location
of the voxel.
The result is much as if you were taking a photograph of the 3D object, with a
light source placed at the viewpoint, and the shading value defined by the angle
of the reflected light.
Note: At times, you may want to modify the location of the light source or the
amount of “ambient” light. See paragraph “3D Shading” for details.
To use surface shading, the 3D model must have been built using a protocol with
the model mode set to “Volume” or “Surface only”. The [Surface] mode is not
available in the rendering mode on-view menu otherwise. See Chapter 5
“Building the 3D Model” for more information.
Note: Initially, a 3D view with surface shading will show only the outside of the
3D volume, including extraneous items such as headboards. To display
a specific feature “buried” somewhere inside, you will have to use the var-
ious volume segmentation tools (such as thresholding, scalpel and paint
tools) to isolate the feature of interest.
During this stage, the 3D view with surface shading tends to be of little
help. For the initial setup, you can switch the rendering mode to projec-
tion shading, which shows the 3D volume as translucent, using for
instance the MIP rendering mode (see paragraph bellow), then switch
back to surface shading after volume segmentation.

9-6
2271012-100.book Page 7 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Projection shading
When using surface shading, all of the structure inside the 3D object is hidden.
Projection shading treats the 3D object as if it were translucent. More precisely,
each element (pixel) of the final picture on the workstation screen is computed
by taking into account the voxel values in the 3D volume on a line (“ray path”)
perpendicular to the screen (view plane) or, in other words, each pixel is a pro-
jection of the voxel values across the depth of the 3D object.
Different options are available of how the voxel values across each projection
will be rendered (displayed), i.e., how the pixels making up the image will be
constructed from the voxel values.
MIP (Maximum Intensity Projection): the pixel value is the maximum voxel value
along the line perpendicular to the screen.
HD MIP (High Definition Maximum Intensity Projection): this mode is identical to
the “MIP” mode described above, except that image definition is greater and
system processing speed is slower.
MinIP (Minimum Intensity Projection) : the pixel value is the minimum voxel
value along the line perpendicular to the screen.
RaySum: the pixel value is the sum of the voxel values along the line perpendic-
ular to the screen. The result is similar in aspect to conventional radiography
images.
Integral: the pixel value is the sum of voxel values along a shallow depth below
the displayed surface point. This mode differs from surface shading in two
respects: it can show features located just below the surface of the 3D model,
but it does not use shading based on the angle of the “reflected” light (see para-
graph “Surface Shading” above) so the result may look “flatter”.

Volume Rendering
With the Volume Rendering mode (available only if the Volume Rendering
option is installed), the pixel value is obtained by considering the voxel values
along the projection (line perpendicular to the screen) as representing a varying
degree of opacity.
Refer to Chapter 11 “Volume Rendering”.

9-7
2271012-100.book Page 8 Thursday, May 8, 2003 11:30 AM

Volume Analysis

3 VOLUME SEGMENTATION
Overview
To extract a 3D object or volume of interest from the original 3D model, three
main techniques are available:
• Thresholding,
• Scalpel,
• Paint.
If the voxel value clearly defines a tissue property or anatomical feature, you can
define a 3D object by selecting the range of voxel values that represents the tis-
sue or anatomical feature. This is referred to as thresholding.
The scalpel tool allows you to perform “cuts” in the 3D volume so as to define a
volume of interest or to split an object into separate objects.
The paint tools allow you to mark a volume of interest in the 3D volume with col-
ored “paint”. Once the volume of interest has been “painted”, an “Apply” function
isolates this volume by removing the remainder of the 3D volume.
Note: You can use these same techniques to remove part of the original 3D
model-obstructing bone in an angiographic exam is a typical case-by first
defining the structure you want to remove, then using the (Show
Removed) function, as mentioned in paragraph 1.

Thresholding
If the voxel value represents a clearly defined property of the tissue, or of a par-
ticular anatomical feature, we can use the voxel value to select a particular type
of tissue, or a particular anatomical feature, by selecting a particular range of
voxel values. This is referred to as thresholding.
Thresholding works well for CT, where there are usually clear distinctions
between the density values, hence voxel values, for different tissue types such
as bone, contrast-enhanced vasculature, or fat.
Thresholding is nearly always used in combination with the other tools.

9-8
2271012-100.book Page 9 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Thresholding controls
You adjust the range of voxel values by means of the “Minimum value” and
“Maximum value” {sliders} on the Threshold control panel. Click and drag on
the slider knob to set the sliders rapidly, or click to the right or left of the slider
knob to increase or decrease the value by one.
For CT exams, you can also use the preset buttons on the panel:
- (Bone) sets the minimum value to 160, without modifying the
maximum value,
- (Air) sets the maximum value to -800, without modifying the
minimum value.
To identify the part of the 3D volume that is being selected, the voxels with val-
ues within the threshold range defined by the sliders are displayed in green on
all views, as soon as you activate the Threshold control panel.
Once the threshold values are set, use (Apply Threshold) to select and display
only the part of the 3D volume with voxel values inside the set range.

WARNING: THRESHOLDING REMOVES ALL VOX-

WARNING! ELS WITH VALUES OUTSIDE THE SELECTED
RANGE FROM THE DISPLAYED 3D VOLUME.
BEFORE APPLYING THE THRESHOLD(S), MAKE
SURE THAT THE SELECTED THRESHOLD SET-
TINGS WILL NOT RESULT IN REMOVING PATHOLO-
GIES OR OTHER ESSENTIAL ANATOMICAL
STRUCTURES.

You can now use the (Keep Object) and/or the (Remove Object) controls to
select and/or remove the resulting 3D objects, and use the “Scalpel” and “Paint”
tools (see following paragraphs) to further narrow down the definition of the
region of interest.
With (Keep Object), the object designated by the current 3D cursor position is
selected and all other objects not physically connected to it via one or more vox-
els, are deleted from the view.
When you activate the (Remove Object) button of the panel, the 3D cursor
takes the shape of a pencil. You move the pencil on the object you want to
remove and click on it with the left mouse button. The object, and everything
physically connected to it via one or more voxels, is dynamically deleted from
the view. You can then move the pencil on the next object to be deleted. When
you have removed all the objects you needed, you click again on the (Remove
Object) button to desactivate it.

9-9
2271012-100.book Page 10 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Note: You can only reduce the voxel value range with respect to the range used
by the build protocol to construct the 3D model (see Chapter 5).
Also, after having used (Apply Threshold) a first time, you can only fur-
ther reduce the voxel value range;
you cannot expand the range to return to an earlier setting, except by
means of (Undo) directly after using (Apply Threshold).
If inadvertently you have reduced the voxel value range too much, reload
the exam.

9-10
2271012-100.book Page 11 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Scalpel
Overview
You use the “scalpel” tool to perform “cuts” in the 3D volume, to split an object
into separate objects, or to define a volume of interest, or on the contrary to
remove part of the 3D volume.
You first select the type of cut (outside, inside or along the trace) and set the cut
depth, then create a trace on a view and apply the cut.
The software now “cuts” along the trace, extending it in the direction perpendicu-
lar to the view on which you created the trace, and according to the type of cut
removes everything outside, inside or along the trace from the 3D volume.
Before applying the cut, if you are not satisfied with the trace you have created,
you can edit the trace, or clear the trace and restart trace entry.

Type of cut
You have a choice of three types of cut: (Cut outside), (Cut inside) or (Cut
along).
(Cut along): the result is a one voxel wide cut in the 3D volume, defined by the
trace and by the cut depth. You will use this function, for instance, to separate
two anatomical features that are touching, or that are connected by a “bridge” of
a few voxels, into two separate 3D objects.
(Cut inside): after defining the trace and applying the cut, everything inside the
cut (over the extent of the cut depth) is removed from the 3D volume.
(Cut outside): as with (Cut inside), but now everything outside the cut is
removed.

You can also execute a Cut by clicking on the

NOTICE

icon of the Review Controller. This will display the Out-

side Cut Definition menu, which gives access to other
type of cuts through the (Other Modes) button.

9-11
2271012-100.book Page 12 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Cut depth
Select (Options) in the Cut Definition panel to set the cut depth.
You can select an “infinite cut depth, in which case the cut will extend right
through the 3D volume.
Alternatively, you can select a “restricted” cut depth and set the depth value. If,
for example, you set a depth value of 10 mm, the cut will extend only to those
slices that are less than 10 mm “above” and “below” the current slice.
The cut depth value should be at least half the slice thickness. Also, obviously,
the cut depth value cannot be greater than the original DFOV of the image set.

Creating and editing a trace

To create a trace:
• Click left on the view where you want to start the cut, then mark the trace by
moving the mouse pointer while continuing to hold the left mouse button
down.
This produces a free-hand trace. You can also click left to deposit points that
are connected by straight-line segments.
If you have selected either (Cut inside) or (Cut outside), the trace is displayed as
a green dotted contour. If you have selected (Cut along), the trace is displayed
as a red line (not as a contour).
To edit a trace:
• Select (Start Edit) in the Cut Definition panel.
• Click and hold the left mouse button on the location of the trace you want to
modify. A section of the trace changes to a thick green line that you can mod-
ify by dragging with the mouse while continuing to hold the left mouse button
down.
• To quit the trace editing mode, select (Stop edit).
To clear the trace and restart trace entry:
• Select (Clear) in the Cut Definition panel.

Applying the cut

Once you are satisfied with the trace, select (Apply) in the Cut Definition panel
to apply the cut.
If for any reason the resulting cut is not what you wanted, select (Undo Scalpel)
in the main control panel, and redefine the cut.

9-12
2271012-100.book Page 13 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Views
You can use baseline and oblique reformatted views to define and perform a cut.
The cut will always be made at right angles to the view on which you create the
trace.
You can also use a 3D view to define and perform a cut. However, you should
use this feature with care, because on the 3D view you have no indication how
“deep” into the 3D object the 3D cursor is located. Always correlate the 3D view
with the baseline views. Also see Chapter 6 “Display and Controls”, paragraph
“3D cursor on 3D views”.

Paint Tools
Overview
You use the “paint” tools to mark, in three dimensions, a region of interest in the
3D volume with colored “paint”. Once the region of interest has been “painted”,
an “Apply” function isolates this region by removing the remainder of the 3D vol-
ume.
This technique is equally effective with CT and MR modality exams.
With the “QuickPaint” tool, you paint with a sphere-shaped cursor on reformatted
slices (baseline or oblique views) to define the volume of interest.
With the “Paintbrush” tool, you trace contours on the baseline (axial, sagittal and
coronal) views, to outline and mark the region of interest on the slices that inter-
sect the region. You do not need to mark every slice: the software automatically
interpolates between marked slices.
Note: When referring to a “region of interest” this does not necessarily mean
only a single 3D object: a region of interest can encompass several dis-
connected features “painted” one at a time.

QuickPaint
Overview
You paint with a sphere-shaped cursor on reformatted slices (these can be
baseline or oblique views) to define the volume of interest. You can then select
to use only this volume for 3D display.
Note: The sphere-shaped cursor is also referred to as the "QuickPaint tool" or
the "SVOI tool", where SVOI stands for Spherical Volume of Interest.

9-13
2271012-100.book Page 14 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Using the QuickPaint tool

You define an object by painting on slices with a “brush” that extends in 3D to
form a sphere.
• Select the view you want to use (oblique or baseline view) by clicking on it
(red border).
• To activate the QuickPaint tool, select (3D Tools) > (Quick brush). The
QuickPaint toolremains active for as long as the QuickPaint panel is dis-
played.
• Set the diameter of the sphere with the {Brush Diameter} slider to adjust for
the size of the object to be painted.
• To paint, move the mouse pointer on the region of interest, then hold down the
left mouse button. A sphere is deposited for each point along the path where
the left mouse button is held down.
The selected areas are highlighted in color on all selected baseline and oblique
views.

Corrections
If paint spills outside the feature of interest, use (Undo Last Paint). Each press
on (Undo Last Paint) removes the paint five spheres at a time, starting with the
most recent ones.
To remove all paint from the 3D volume and start again, press (Clear All Paint).
Before using (Apply) (see below), check that the feature of interest is correctly
painted on all slices, by moving up and down in the "stack" of slices that contains
the feature. If small parts of a feature have remained unpainted, paint them
before using (Apply).

Apply
Once you are fully satisfied with the definition of the region of interest, (Apply)
segments the 3D volume by removing all parts that are not painted.
As noted before, the term “region of interest” does not necessarily refer to only a
single 3D object: the region of interest can encompass several disconnected
features painted one at a time. However, all should have been painted before
using (Apply).
Immediately after using (Apply), you can still (Undo) the segmentation and
make corrections. The colored “paint” is not removed until you (Close) the
QuickPaint control panel.

9-14
2271012-100.book Page 15 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Paintbrush
Overview
You use the “Paintbrush” tool to mark, in three dimensions, a region of interest in
the 3D volume with colored “paint”. Once the region of interest has been
“painted”, an “Apply” function isolates this region by removing the remainder of
the 3D volume.
To “paint”, you trace contours on the baseline (axial, sagittal and coronal) views,
to outline and mark the region of interest on the slices that intersect the region.
Note: You cannot use the “Paintbrush” tool on oblique views.
You do not need to mark every slice: whenever the painted contours on two or
more slices “overlap” (i.e., lie at least partly on a common perpendicular axis)
the software automatically performs a smooth interpolation between them by
applying paint to the three-dimensional region between them.
While tracing contours, you can use an edge attraction feature, that keeps the
contour “on track” along the border between different pixel densities even if you
don’t follow it precisely with the cursor. To use the feature, you must adjust the
window W/L settings so that the edge (border) can be distinguished clearly.
To make corrections, you can erase the contour on one of the slices and trace it
again, erase all and start over, or use the “paint” tool in “clear” mode, to mark a
region where you want to remove the “paint” by means of the same technique
you used when painting.

9-15
2271012-100.book Page 16 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Using the Paintbrush tool

• Select the desired baseline view (axial, sagittal or coronal) as primary (red
border)
• Move the 3D cursor location to the feature of interest and, if necessary, adjust
window width and level on the view so that the feature of interest is clearly dis-
tinguishable.
• Select the slice (image location) on which you want to start “painting”.
• Select (3D Tools) > (Paint on slices) to open the control panel. Make sure
the (Painting mode) is set to “Paint”.
• Move the mouse pointer onto the edge of the feature of interest, then press
and hold the <Shift> key on the keyboard and start moving the mouse pointer
along the contour of the feature of interest.
The contour is marked by a green trace. When you release the <Shift> key
the software marks the inside of the contour with red “paint”.
Note: If the contour was not closed when you released the <Shift> key, the
software connects start and finish of the contour (green trace) with a
straight line before painting the inside of the contour.
If you are using the edge attraction function, first adjust window W/L so that
the edge of the feature of interest is clearly visible on the view.
You can now move the mouse cursor slightly ahead of the trace, “dragging” it
round the edge of the feature of interest: the software keeps the contour “on
track” along the border between different pixel densities even if you don’t fol-
low it precisely with the cursor.
Without edge attraction, the trace simply follows the mouse cursor.
From the Paint control panel, use (Edge attraction) to switch edge attraction
on or off.
• Move to the next slice (image location) on which you want to “paint”.
Note: to move the 3D cursor to different slices (image locations) use the
active annotations or the <left> and <right> cursor keys on the key-
board, or click and drag on the 3D cursor itself. To avoid creating
unwanted “paint” traces, do not use the <Shift> key to to move the
3D cursor at this stage, unless you first set the painting mode to
“None” (see below).
• Again press and hold the <Shift> key and trace the contour, then release the
<Shift> key to paint.

9-16
2271012-100.book Page 17 Thursday, May 8, 2003 11:30 AM

Volume Analysis

You do not need to paint every slice one by one.

Whenever painted contours on two slices “overlap” (i.e., lie at least partly on a
common perpendicular axis) the software automatically performs an interpola-
tion and applies paint to the intervening slices.
• Continue in the same manner until the entire feature of interest is painted.
Exactly how you paint the slices (cross-sections) that define the feature of inter-
est, in what order, and how many, obviously depends on the shape of the fea-
ture. You can work from “top” to “bottom”, or start from the center of the feature,
and work “up” and “down”.

9-17
2271012-100.book Page 18 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Painting mode
To mark the feature of interest with colored “paint” as described above, you set
the (Painting mode) in the Paint control panel to “Paint”.
The opposite function, to clear selected areas of “paint”, is also available.
• Switch the (Painting mode) to “Clear”.
• Press and hold the <Shift> key and trace the contour of the area you want to
clear (blue trace), then release the <Shift> key to to remove the paint.
Note: If you use “Clear” mode on different slices, the software will interpo-
late and clear the intervening slices in the same manner as it inter-
polates and applies paint in “Paint” mode.
For this reason, use the function with care: you may at times find
yourself unintentionally clearing more of the “paint” than you
intended.
If you want to use the <Shift> key and mouse cursor to move the 3D cursor with-
out creating “paint” traces, set the (Painting mode) to “None”.

Corrections
Before using (Apply) (see below), check that the feature of interest is correctly
painted on all slices, by moving up and down in the “stack” of slices that contains
the feature. In particular, check the interpolated slices.
• If small parts of a feature have remained unpainted, you can simply contour
and paint the unpainted area. You do not need to repaint the part that has
already been painted.
• If paint has spilled outside the feature of interest, you have two options:
- Use (Erase slice) to remove all paint from the current slice, and
retrace and paint again,
- Use the “Clear” mode described above.
Be careful when using this mode: because of the interpolation fea-
ture you may find yourself unintentionally removing paint on slices
other than the current slice.
Additionally:
• You can use (Erase last) to undo the last “paint” or “clear” action.
• You can use (Erase all) to erase all paint from the 3D volume. A confirmation
window allows you to confirm or cancel the action.

9-18
2271012-100.book Page 19 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Apply
Once you are fully satisfied with the definition of the region of interest, (Apply)
segments the 3D volume by removing all parts that are not painted.
As noted before, the term “region of interest” does not necessarily refer to only a
single 3D object: the region of interest can encompass several disconnected
features painted one at a time. However, all should have been painted before
using (Apply).
Immediately after using (Apply), you can still (Undo) the segmentation and
make corrections. The “paint” is not removed until you (Close) the Paintbrush
control panel.

Special Tools
Overview
A number of special tools are available in the Volume Analysis 2 application
and described further in the following paragraphs.
Several of these tools can help you to remove or attenuate the effects of “partial
volume averaging”. This “partial volume effect” is discussed in some detail in the
next paragraph.

9-19
2271012-100.book Page 20 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Partial volume effect

The “partial volume effect” refers to the appearance of voxels with intermediate
values at the interface of two dissimilar tissue types.
The effect is discussed and illustrated below for a CT exam, but occurs with
other modalities as well.
With CT, each voxel represents the average density in a small volume. If the
voxel is located entirely within bone, the voxel value will be representative of
bone. If the voxel is located entirely within the fatty tissue surrounding the bone,
the voxel value will be representative of fat. But if the voxel is located at the
abrupt transition between bone and fat, the voxel value will be somewhere inter-
mediate between that for bone and for fat.
Now this intermediate value may well be the same as that of an anatomical fea-
ture that you are interested in.

A common example occurs in

Fat/bone interface CT Angiography, where the
typical value for contrast-
Voxel size enhanced vasculature (in
theillustration 300 HU is
Bone assumed) is intermediate
Fat (0 HU) between fat (0 HU assumed)
Fat Fat
and bone (600 HU assumed).
Vessel or fat/bone As a result, artificial "vessels"
interface (300 HU) may appear at the interface
between fat and bone, where
Bone (600 HU) the average of these two tis-
sues is in the range of vascu-
"V essel" artifact lature.
In many cases this makes it
difficult to select a single class
of tissue, because partial vol-
ume artifacts exist in the
desired range of values.
Note: The HU values
assumed for the differ-
ent tissue types in the
illustration are to be
regarded as examples
only. Real values may
differ significantly.

At first sight, you are now faced with a dilemma when trying to remove the bone
from a 3D model by using thresholding.

9-20
2271012-100.book Page 21 Thursday, May 8, 2003 11:30 AM

Volume Analysis

If you set the threshold high enough to be sure to select bone only, a thin shell of
partial volume voxels will remain in place after you remove the bone, and this
shell of partial volume voxels will appear as pseudo-vasculature and continue to
obscure the features you were trying to display.
If instead you set a low threshold that includes the partial volume voxels, other
features such as vasculature will disappear from the 3D model together with the
bone.
In other words, the partial volume effect can make it difficult to cleanly differenti-
ate between two tissue types by means of thresholding alone.
This is where you can use the “dilation” and “erosion” 3D processing tools.\
In the above example, you would set the threshold to, say, 500 HU, selecting
only those voxels clearly representing bone. By using the "dilation" tool, and a
setting of, say, 2 voxels you can expand the "bone" volume slightly, to include
the partial volume voxels. You can now select the entire volume and remove it:
both bone and associated partial volume voxels are gone, but the other features
with the same voxel value as the partial volume voxels-in this example the vas-
culature - have remained in place.
The “erosion” tool performs the opposite function. If you set thresholds to make
sure to include all of an anatomical feature, you are certain to include partial vol-
ume voxels at the surface of the feature. The “erosion” tool allows you to “peel
off” these partial volume voxels.
Note: The “Volume Rendering” software option contains additional functions to
reduce the partial volume effect (variable opacity settings, partial volume
filter). Refer to Chapter 11 “Volume Rendering”.
The same or similar partial volume effects can also be at the origin of other
anomalies you may encounter, such as floaters, bridges and gaps. See para-
graphs “Filter” and “Advanced Processing” bellow.

9-21
2271012-100.book Page 22 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Dilate/Erode
Dilate
(Dilate) allows you to add one or more layers of voxels to the surface of the cur-
rent 3D objects. You will be asked to enter the number of “layers” you want to
add. The default value is 1, the maximum possible value is 20.
Note: Although this function is in some ways the reverse operation of erosion
(see below), it cannot restore voxels which do not exist in the original 3D
model. However, it can restore voxels which were removed by some
operations like thresholding, erosion or opening bridges.
Erode
(Erode) allows you to “peel” one or more layers of voxels from the surface of the
current 3D objects. You will be asked to enter the number of “layers” you want to
remove. The default value is 1, the maximum possible value is 20.

9-22
2271012-100.book Page 23 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Filter
Filter floaters
“Floaters” are small residual objects in the 3D model that can appear after
thresholding, usually resulting from noise in the original image set.
You can select “small”, “medium” or “large” (the corresponding sizes in mm3 are
shown on the control panel) or allows you to remove these residual objects. You
will be asked to enter the floater size limit in mm3.
enter a “custom” value.

FLOATER FILTERING REMOVES ALL 3D OBJECTS

WARNING! FROM THE DISPLAYED 3D VOLUME, THAT HAVE A
SIZE EQUAL TO OR SMALLER THAN THE
SELECTED FILTER SIZE.
BEFORE APPLYING A FILTER, MAKE SURE THAT
THE SELECTED FILTER SIZE WILL NOT RESULT IN
REMOVING PATHOLOGIES OR OTHER ESSENTIAL
ANATOMICAL STRUCTURES.

Extract surface
(Extract surface) removes all data from the “inside” of the current 3D objects,
leaving only the surface. This results in a faster display, because much less
computer memory is used. However, since the inside of the objects no longer
contain any data, little if any further processing is possible.
You can use this function to speed up the display after you have fully defined
your region or object(s) of interest, e.g., during rotation or batch filming, or to
modify 3D shading.
When using (Extract surface) you will be asked to enter the surface thickness in
voxels. It is advisable to enter a value of at least 2 to guarantee connectivity
information for any further operations such as selecting or removing objects.
With a value of 1 the result appears visually correct, but the surface is too thin to
be considered as one or more coherent objects.

9-23
2271012-100.book Page 24 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Close holes
When using thresholding, i.e., reducing the selected range of voxel values in the
3D model, this can result in “holes” appearing inside the 3D volume (i.e., closed
spaces inside the 3D volume where the voxel value is outside the selected
range).
By default, the voxel value inside such inner holes will be set to the same value
as the outside of the 3D volume (“empty space”). (Close holes) resets any such
inner holes to the original voxel values.
Note: If you have used the “Scalpel” or “Paint” tools to select and then remove
part of the 3D model, any inner holes (i.e., enclosed within the 3D vol-
ume) resulting from these operations will also be re-filled when you use
the (Close holes) function.

Advanced Processing
Open bridges
Adjacent features in the 3D volume that are apparently separate objects may
turn out to be a single 3D object when you try to use (Keep Object) or (Remove
Object), in particular after thresholding. This can be caused by the presence of
small residual “bridges”, usually only a few voxels in size, that still connect the
features.
To separate the features you can use the “scalpel” tool, on condition that you can
easily identify the location where the features are still connected.
Alternatively, use (Open Bridges) to remove the residual bridges and separate
the features. You will be asked to set the size in voxels of the bridges you want
removed.
Note: This function resembles the (Erode) function, but only the “bridges” up to
the specified size are eroded, not the rest of the objects.
More specifically, the function consists of performing an erosion followed
by a dilation. Since dilation does not totally restore the voxels removed by
erosion, some fine structures remain eroded. An opening size of N is
obtained by N erosions followed by N dilations. This means that fine
structures having one of their X, Y or Z dimensions less than 2N are
removed. For example, an opening size of 3 is obtained by 3 erosions
followed by 3 dilations. This means that fine structures having one of their
three dimensions less than 6 are removed.
Note: Also see the next paragraph “Close gaps” which describes the opposite
function.

9-24
2271012-100.book Page 25 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Close gaps
Adjacent features in the 3D volume that apparently constitute a single object
may turn out to be composed of more than one 3D object when you try to use
(Keep Object) or (Remove Object). This can be caused by the presence of
narrow gaps, often only a few voxels in size, that separate the features.
This mostly occurs when using thresholding to define a feature of interest, when
the threshold setting is marginal. You can try to modify the threshold setting, or
alternatively use (Close Gaps) to fill in the gaps and connect the adjacent fea-
tures. You will be asked to set the size in voxels of the gaps you want filled in.
Note: This function resembles the (Dilate) function, but only gaps up to the
specified size are filled in; the rest of the objects are not dilated.
More specifically, the function consists of performing a dilation followed
by an erosion. Since erosion does not totally remove the voxels added by
dilation, some small gaps or holes are filled in. A closing size of N is
obtained by N dilations followed by N erosions. This means that small
gaps or holes having one of their X, Y or Z dimensions less than 2N are
filled in. For example, a closing size of 3 is obtained by 3 dilations fol-
lowed by 3 erosions. This means that small gaps having one of their three
dimensions less than 6 are filled in.
Note: Also see the previous paragraph "Open bridges" which describes the
opposite function.

9-25
2271012-100.book Page 26 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Set operations
You can perform a certain number of set operations on the 3D object data dis-
played in two views (this assumes that the 3D object data in these two views are
not the same).
To use these operations, one view must be selected as the primary view (red
border) and the other must be a secondary view (green border). All other views
must be deselected (no colored border).
In each case the result of the operation replaces what was displayed in the pri-
mary view.
Intersection

Intersection The set intersection operation keeps only the voxels that
exist in the same location in both objects. The values of
the resulting voxels are those of the original object in the
primary view.

Addition

Addition The set addition operation keeps all the voxels that exist
in either of the objects. If a voxel belongs to both objects,
its value in the primary view is kept.

Subtraction

Subtraction The set subtraction operation removes all the voxels in

the primary view that also exist in the secondary view. In
other words the secondary view is subtracted from the pri-
mary view.

The result of using a set operation on the data from two 3D models is a single
object (3D model).
This is different from the “merge” operations described further in this chapter
which allow you to display more than one 3D model at the same time, and show
their spatial relation by means of cut planes and different levels of transparency.

9-26
2271012-100.book Page 27 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Merge operations are strictly a display feature, they do not combine the separate
3D models.
The “set addition” operation can be used to combine structures that have been
obtained using different processing tools.
As an example, you may well need different tools and settings to process a ves-
sel from the left side of the patient and another vessel from the right side. By
treating the two vessels separately and storing the results in the Save/Recall
panel, you can optimize the processing for each side.
After recalling both from the Save/Recall panel in two separate views, the “set
union” operation allows you to join and display them as a single object.
The “set subtraction” operation is used to subtract structures.
As an example, you can start by using thresholding to select the entire hip bone
structure and store the result in the Save/Recall panel. You can now use differ-
ent tools (such as scalpel and paint tools) to isolate and select only the femur
and store the result separately.
Subtracting the femur from the complete bone structure may now allow you to
see the extent of a hip bone fracture that was obscured by the femur.

3D Object Management
As already noted earlier, each 3D object can be selected separately, by placing
the 3D cursor on it; you can then either keep the object (and remove everything
else), or remove the object (and keep everything else).
When you click on (Keep Object), the object defined by the current cursor posi-
tion is kept and all other objects not physically connected to it via one or more
voxels, are deleted from the view.
When you click on (Remove Object), the object defined by the current cursor
position is removed from the view.
If you have performed volume segmentation, and removed parts from view,
(Show Removed) allows you to view the removed parts.

9-27
2271012-100.book Page 28 Thursday, May 8, 2003 11:30 AM

Volume Analysis

4 3D MODEL MANAGEMENT
Overview
The Volume Analysis 2 application allows you to manage several 3D models
simultaneously using the (Save/Recall) function. This allows you to temporarily
store a 3D model that you have modified in a certain way and call it back when-
ever needed.
In particular, you can use this function to store two or more dissimilar 3D models
(an example would be a first 3D model defining the skull structure, and a second
one defining a specific feature inside the brain), during a session and then recall
and merge the stored 3D models to display and analyze the spatial relations
between the structures.
On the other hand, an option of the on-view menu allows you to reverse back to
the volume you originally loaded (refer to Chapter 6, Section 5 “Main On-View
Menu”).

Icon Area
Overview

The icon area in the Save/Recall control panel contains

between one and eight small 3D images (the icons) rep-
resenting the currently built 3D models. Icons cannot
be modified, but allow you to see the available volumes
and manage them.
Icons represent objects stored for use during the current
session only. They do NOT represent files stored on
disk and cannot be recovered once you quit Volume
Analysis 2. To manage objects as files on disk, see
“Saving the 3D Model” in Chapter 14 “Recording”.
Volume Analysis 2 can manage a maximum of eight
3D models during a given session.

Main icon
The main icon is the one that appears in the icon area of the Save/Recall control
panel when a model is initially built. It remains there until you either build another
model or quit Volume Analysis 2.
At any time during the Volume Analysis 2 session, regardless of any modifica-
tions you have done to the 3D model, you can recall the ORIGINAL built model
by pressing and holding left on the main icon, dragging it to a view and dropping
it. The prior view content is deleted and replaced with the original model.

9-28
2271012-100.book Page 29 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Storing a 3D Model as an Icon View

To temporarily “set aside” a 3D model after performing certain modifications to it,
you store it in an icon view as follows:
• Open the Save/Recall panel,
• Press and hold left on a view displaying the 3D model (but not on the track-
ball, 3D cursor or red annotations), and drag from the view to the icon area,
• Drop the view into the icon area by releasing the mouse button,
Note: The maximum number of icons allowed in the icon area (including the
main icon) is EIGHT. If you try to store another 3D model when there are
already eight icons, a message window will pop up and inform you of this.

Recalling a 3D Model
To recall a 3D model on a view:
• Open the Save/Recall panel,
• Press and hold left on the icon to be restored, drag it to a view and drop it by
releasing the mouse button. The prior view content is deleted and replaced
with the 3D model corresponding to the icon.
Note that, when you recall a 3D model on a view, it will take on the properties of
that view. For example, if a 3D model is dragged onto a 3D view it will be dis-
played in 3D, but if it is dragged onto a coronal view it will be displayed as a
coronal slice. Also the display parameters (window level/width, cut planes, col-
ors, rendering mode, etc.) will be those of the view.

Deleting an Icon
To delete an icon from the Save/Recall panel icon area, press right on the corre-
sponding icon and select [Delete] in the pop-up menu (that is the only choice
that the menu contains).

9-29
2271012-100.book Page 30 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Applying Colors
Overview

Colors can be applied to the views, and in particular the 3D views

using the color buttons in the (Display Tools) > Colors panel.

There are two cases for applying colors to 3D views:

• Non-merged 3D views,
• Merged 3D views.
Note: To custom-set the on-view colors, see paragraph “Using Colors”.

Using colors in non-merged views

When a non-merged view is colored, everything in it takes on the same color. To
do this, select the view to be colored, then click on the desired color button in the
(Display Tools) > View Colors panel.
Note: Coloring of non-merged views is typically done in preparation for a 3D
merging operation. This makes distinguishing the merged objects easier.

Using colors in merged 3D views

When coloring is applied to a merged 3D view (see “Merging 3D Models” below),
each merged 3D model can be colored independently of the others.
This means that you must first decide which merged 3D model you want to color,
and then select it.
The selected merged 3D model by default following the merge operation is the
3D model that was in the target view prior to merging. If that is the merged 3D
model that you want to color, DO NOT move the 3D cursor before clicking on the
desired color button in the (Display Tools) > View Colors panel.
If you want to select a different merged 3D model for coloring, do one of the fol-
lowing:
• Move the 3D cursor to the merged 3D model that you want to select for color-
ing, OR
• Press right on the transparency status annotation (see “Transparency Status
Annotation” below) and select [Change Focus Object] from the menu that
pops up.
Note: If you use the second method, DO NOT move the 3D cursor before click-
ing on the desired color button in the (Display Tools) > View Colors
panel.

9-30
2271012-100.book Page 31 Thursday, May 8, 2003 11:30 AM

Volume Analysis

The most deeply imbedded VISIBLE merged 3D model pointed to by the

3D cursor is the one that is selected.
The newly-selected 3D model flashes briefly in green to indicate that it
was selected.
Once you have selected the desired merged 3D model for coloring, click on the
desired color button in the (Display Tools) > View Colors panel.

Transferring 3D Models Between Views

Overview
You can transfer a 3D model from one view to another using the same “drag and
drop” principle as for icons described above.
Again,, when you transfer a 3D model from one view to another, it will take on
the properties of the destination view. For example, if a 3D model is transferred
to a 3D view it will be displayed in 3D, but if it is transferred to a coronal view it
will be displayed as a coronal slice.
The display parameters (window level/width, cut planes, colors, rendering mode,
etc.) will be those of the view into which it is being transferred.
Objects can be transferred between views in one of two ways:
• Copying of an object into a view with deletion of the original object in that
view,
• Merging together of one 3D object from one 3D view with another 3D object in
another 3D view.

Copying a 3D model from one view to another

To copy a 3D model from one view to another, with deletion of the original 3D
model in the target view, proceed as follows:
• Press and hold left on the view containing the 3D model to be moved (but not
on the trackball, 3D cursor or red annotations),
• Drag the 3D model to the desired target view,
• Drop the 3D model into the flashing box marked “Drop here to reassign view”
by releasing the left button.
The dropped 3D model replaces the prior content of the target view.

9-31
2271012-100.book Page 32 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Merging 3D models
To merge a 3D model in one view with a 3D model in another view, proceed as
follows:
• Press and hold left on one of the views containing a 3D model to be merged
(but not on the trackball, 3D cursor or red annotations),
• Drag the model to the desired target view containing the other 3D model to be
merged,
• Drop the model into the flashing box marked “Drop here to merge views” by
releasing the mouse button.
The dropped 3D model is now merged with the prior content of the target view.
Also, the transparency status annotation appears in the upper left corner of the
view. See paragraph “Transparency Status Annotation (Merged 3D Views
Only)” below for use of merged 3D model control functions.
Note: A maximum of eight 3D models can be merged.

9-32
2271012-100.book Page 33 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Transparency Status Annotation (Merged 3D Views Only)

If you have merged two or more 3D models together (see “Transferring Objects
Between Views” above), the transparency status annotation appears in the
upper left corner of the 3D view containing the merged 3D models. See illustra-
tion.
The default setting of the transparency status annotation for merging 3D models
is Transparent.

Transparency and merged 3D model selec-

tion functions can be accessed by pressing
right on this annotation and selecting one of
the choices from the menu that pops up:

Set Transparency
Change Focus Object
Keep Focus Object Only
Remove Focus Object Only

[Set Transparency] exists only if all the merged 3D models in the view are dis-
played in surface rendering mode. It is used to make the largest merged 3D
model transparent. The transparency status annotation changes to Transparent
when this choice is selected.
This choice then becomes [Cancel Transparency] for restoring the largest
merged 3D model to its solid status.
[Change Focus Object] is used to select another merged 3D model for display
mode manipulation (window level/width, cut planes, colors, rendering mode,
etc.). The newly-selected 3D model flashes briefly in green to indicate that it is
selected.
[Keep Focus Object Only] is used to delete all merged 3D models EXCEPT
the selected one.
[Remove Focus Object Only] is used to delete ONLY the selected merged 3D
model.
Note: The selected merged 3D model by default following the merge operation
is the 3D model that was in the TARGET view prior to merging. This
remains true until you either move the 3D cursor or use [Change Focus
Object] in the menu described above.

9-33
2271012-100.book Page 34 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Note: Another way to select a merged 3D model for display mode manipulation
is to move the 3D cursor to it. (The most deeply imbedded VISIBLE
merged 3D model pointed to by the 3D cursor is the one that is selected.)

MODIFYING TRANSPARENCY WILL AFFECT HOW

PATHOLOGIES AND OTHER FEATURES IN THE 3D
MODELS ARE DISPLAYED.
WARNING!
MAKE SURE THAT PATHOLOGIES AND OTHER
ESSENTIAL ANATOMICAL STRUCTURES ARE
CLEARLY SHOWN WHEN USING MERGED 3D
MODEL VIEWS FOR DIAGNOSTIC PURPOSES.

9-34
2271012-100.book Page 35 Thursday, May 8, 2003 11:30 AM

Volume Analysis

5 DISPLAY
Overview
In most respects, a 3D view is like any other view. You can adjust window width
and level as required, and zoom and scroll the image. You can add text annota-
tions, and change the color of the displayed object.
In theory, you can also perform measurements, or define and perform a “scalpel”
cut, on a 3D view, since you can mark points and create traces on it. In practice,
this is highly inadvisable because on the 3D view you have no indication how
“deep” into the 3D object the 3D cursor is located, and the result may not be at
all what you wanted or expected.
A 3D view is different in that it displays an image of the 3D model (which may
consist of one or more 3D objects), and that you can manipulate this 3D model.
On a 3D view, you can:
• Rotate and tilt the 3D model in all directions,
• Define one or more cut planes and display the 3D model with part of it
removed, showing a cross section of the 3D model at the location of the cut
plane,
• Place a spherical “shutter” (mask) on the view to show only an essential part
of the 3D model, with the remainder masked out.
• Use colors, in particular to distinguish merged 3D models in merged 3D
views.
• Adjust the way the 3D model will be rendered, in the same way as you would
adjust the lighting when taking a photograph.

9-35
9-36
2271012-100.book

Volume Analysis

Hospital
View type Name
Rendering mode
Patient name
annotation
and control

Cut plane
Page 36 Thursday, May 8, 2003

Field of view
(zooming control)
11:30 AM

3D cursor

Orientation
Protocol
annotations
Name
(roaming control)
VOI
annotations
Reference
Image
Window
width/level 3D view with
annotations annotations
and controls
2271012-100.book Page 37 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Rotating the 3D Model

Rotating the displayed 3D object provides you with important additional depth
cues.
On surface-shaded views, it allows you to look at the object from all sides, both
to look at hidden features at the “back” of the object, and to obtain a better 3D
perception of the object.
On projection-shaded views, it allows you to look “through” the 3D object at dif-
ferent angles, which provides the depth cues that are mostly lacking on a single
view.
To rotate the 3D model on the view, you can:
• Use the 3D “cube” control on the view (see Figure 1 bellow),
• Use the Rotate/Translate controls to rotate the 3D model in steps, or return it
to one of the baseline orientations.
• se the (Tumble) function in the Rotate/Translate panel to “rock” the 3D model
from side to side over a small angle.
Note: On 3D views, the Tumble function can also be executed using the Rock
mode button of the Review Controller.
• Create a batch movie loop that allows you to rotate the 3D model in a continu-
ous fashion and to stop it in any position: see Chapter 14.

9-37
2271012-100.book Page 38 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Using the on-view cube

To use the on-view cube, first select (Tilt / Rotate) mode.

In this mode, as soon as you move the mouse pointer onto
the 3D view, a red “cube” symbol with “handles” in the cor-
ners and on the sides appears on the view.

You can now use two different methods to rotate the 3D model.
To rotate and tilt the 3D model in any direction:
• Click and hold with the left mouse button on one of the corners of the cube. If
the view wasn”t selected before, it is now selected.
• Still holding the mouse button down, drag the mouse around. The model will
start to rotate with mouse movement. Keep holding down the mouse button
until the model has rotated to the new position. When the desired orientation
is reached, release the mouse button.
To constrain the rotation of the 3D model to an axis parallel to one of the edges
of the cube:
• Press and hold with the left mouse button on the rotation marker on the mid-
dle of an edge of the cube parallel to the axis around which you want to rotate
the model. If the view wasn't selected before, it is now selected.
• Still holding the mouse button down, drag the mouse around. The model will
start to rotate with mouse movement, but it will rotate only around the axis par-
allel to the selected edge. Keep holding down the mouse button until the
model has rotated to the new position. When the desired orientation is
reached, release the mouse button.
Note: You use the same cube to rotate the oblique plane on an oblique view
(see Chapter 8).

9-38
 2271012-100.book
Page 39 Thursday, May 8, 2003
11:30 AM

Fig.1: Horizontal Handling of the on-view cube

Fig.2: Vertical Handling of the on-view cube Fig.3: Diagonal Handling of the on-view cube
Volume Analysis

9-39
2271012-100.book Page 40 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Rotation and translation

You can use the (Rotate/Translate) controls to rotate the 3D model or move the
3D cursor in precise steps. You can also return the 3D model to one of the base-
line orientations.
• Select the 3D view by clicking on it (red border).
To rotate the 3D model in steps:
• Select (Rotate/Translate) > (By deg),
• In the numerical entry field, enter the required angle step in degrees,
• Click on the arrow button corresponding to the desired direction of rotation.
To move the 3D cursor in steps:
• Select (Rotate/Translate) > (By mm),
• In the numerical entry field, enter the required distance step in millimeters,
• Click on the arrow button corresponding to the desired direction of translation.
If after repeated rotations you want to rapidly return the 3D model to a known
baseline orientation:
• Use the (S) (I) (A) (P) (L) (R) buttons on the main control panel.
Note: To return the 3D cursor to the initial position at the center of the 3D vol-
ume, select [Reset pointer] in the on-view menu.

Using the Tumble mode

You can use the “Tumble” mode to “tumble” or “rock” the 3D model from side to
side over a small angle. This mode can be highly effective in providing addi-
tional depth cues.
• Select (Rotate/Translate) > (Tumble),
Note: Adjust window width and level on the 3D view before selecting “Tumble”
mode.
There may be a few moments delay before the “tumble” motion starts,
because, in “Tumble” mode, the software first has to compute a small
sequence of images, that is then displayed as a movie loop. To stop
“Tumble” mode, click again on the (Tumble) button, or (Close) the
(Rotate/Translate) panel.

9-40
2271012-100.book Page 41 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Using Batch Film mode

A batch movie loop allows you to rotate the 3D model in a continuous fashion, to
adjust the display speed and to stop it in any position. When creating a batch
movie loop you can select the number of views making up the movie loop, the
FOV to use, etc.
See Chapter 14 “Recording” for details on how to set up a batch movie loop and
display it using the (Preview) function.

Cut Planes
You can use the No Cut active annotation to define cut planes in 3D views.
These Cut Planes can be defined whatever the Rendering Mode (Surface, VR,
HD MIP, MIP, MinMIP, RaySum and Integral).
As an example, selecting a “left cut” defines a “vertical” anterior-to-posterior cut
plane at the location of the 3D cursor, and everything to the left of this plane is
removed from the view. You can move the position of the cut plane by moving
the 3D cursor to the left or right on the axial or coronal view.
You can also use a “triple” cut: this defines three orthogonal planes that intersect
at the location of the 3D cursor, dividing the 3D object in eight. You select which
part you want removed from the view. Again, the position of the cut planes can
be changed by moving the 3D cursor.
The cut planes are defined relative to the RAS coordinate system, except for the
[Front cut] and the [Back cut] which are relative to the user’s point of view.
When you tilt or rotate the 3D object on the 3D view, the cut planes move with it.
In VR Rendering Mode, the Cuts can picture either the density value (voxel) at
the level of the cut, or a standard 3D rendering view (you can toggle from one to
the other using the first function of the Cut mode on-view menu, [Show voxel
value on Cuts]
Note: Cut planes are strictly a display feature. Images with cut planes can be
filmed or saved (screen save) but the cut planes do not change the 3D
model itself in any way.

9-41
2271012-100.book Page 42 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Volume of Interest
You can use the VOI (Volume of Interest) active annotation to define a spherical
mask on 3D views, to display a particular feature of interest while masking out
the remainder of the 3D volume. There are two types of masks:
The Curved VOI option:
• Select the [Curved VOI] option in the No VOI annotation drop-down menu.
• Define a trace on the volume, using the standard procedure for that.
Only the part of the 3D model included within a given distance from the curved
surface will be displayed. By default, the initial distance is 0.5 mm. This value is
an active annotation and can therefore be modified.
This option is available on all 3D views in VR, MIP and MinMIP rendering
modes.

A CURVED VOI VIEW CAN INTRODUCE DISTOR-

WARNING! SIONS IN THE SHAPE OF OBJECTS.
TO PREVENT MISINTERPRETATION OF THE SHAPE
OF AN OBJECT, THE USER SHOULD ALWAYS VER-
IFY THE CURSOR POSITION BY CORRELATION
WITH THE BASELINE AND REFORMATTED VIEWS.

The Shutter on Cursor option:

• Select the [Shutter on Cursor] option in the No VOI annotation drop-down
menu.
Only the part of the 3D model included inside the shutter sphere is displayed,
and the remainder is masked out. The projection of the shutter sphere on the
baseline views is shown by circles on these views.
Use the shutter size active annotation (in cm) to modify the size of the shutter
sphere as required.
The feature of interest in the shutter sphere is centered on the 3D cursor and fol-
lows the 3D cursor when you move it.
To remove the shutter or the CVOI from the view:
• Select [No VOI] in the drop-down menu.
Note: Shutter and CVOI are strictly display features. Images with a shutter or a
CVOI in place can be filmed or saved (screen save) but neither of these
options change in any way the 3D model itself.

9-42
2271012-100.book Page 43 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Using Colors
By default, the images in the views are displayed in black-and-white.

You can apply colors to the 3D views using the color buttons in
the (Display Tools) > View Colors panel.

The main application is in merged 3D views where each merged 3D model can
be colored independently of the others, to make it easier to distinguish the
merged 3D models. See paragraph 4 “3D Model management”, and in particu-
lar paragraph “Applying Colors”.
You have a choice of eight different colors, which you can adjust to your own
requirements:
• Select the view that you want to color,
• Open the (Display Tools) > View Colors panel,
• Select the color you want to use,
• If required, adjust the color using the red, green and blue color sliders.
Note: This function does not apply to Volume Rendering.

3D Shading
When you use surface rendering to display objects on a 3D view, you can adjust
the way the objects will be rendered in the same way as you can adjust the light-
ing when taking a photograph of a real object, and you use the controls much as
if you were adjusting light sources.

Depth cueing
If you place a light source close to an object, the parts at the front of the object
will be more brightly illuminated than the parts at the back, giving an impression
of depth (image a long corridor with objects along it and a light at one end only).
If you move the light source further away from the object (and increase the inten-
sity so that the overall level of illumination stays the same) the parts at the front
and at the back of the object will receive more nearly the same illumination, and
the impression of depth due to this effect will be less.

9-43
2271012-100.book Page 44 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Ambient light
If you use only a single light source (“spotlight”), any shadows and surfaces at
right angles to the incident light will be completely black. By adding a degree of
ambient light, coming from all directions, you can lighten the shadows (think of it
as turning on the “room lighting”).
Low ambient light values can accentuate contours, but details in shadow areas
tend to get lost. The higher the ambient light value, the more shadows are sup-
pressed (“washed-out”) on the object; at the highest value only a silhouette of
the object remains.

Blending ratio (merged 3D views with transparency only)

In a merged 3D view with transparency activated, the blending ratio determines
the relative balance of displayed intensity of the objects .
You can the blending ratio to 25%, 50% or 75%. This corresponds to the per-
centage of displayed intensity that you wish to attribute to the innermost object in
the merged view. The intensity is updated in real time.
See “Transferring Objects Between Views” and “Transparency Status Annota-
tion” for more information about merged 3D views and transparency.
Note: The (Blending ratio) setting has no effect on views that are not merged
3D views with transparency activated.

Light X and Y
By default the main light source is assumed to be located at the same position
(X=0, Y=0) as the viewpoint (the “camera”), but you can move the light source
left and right, and up and down. For instance, setting X= -90 and Y=45 would
place the light source at the left of the object and at 45 above it. The highlights
and shadows on the object change position correspondingly.

Smooth 3D image
At times (depending on the anatomical feature being viewed and the type of 3D
processing used), small unrealistic details, jagged edges or even staircase-like
artifacts in the case of thick slices can appear on the 3D views. The “Smooth 3D
image” function allows you to attenuate these effects.
Note: The same effects can occur if the Ambient light setting (see above) is too
low.

9-44
2271012-100.book Page 1 Thursday, May 8, 2003 11:30 AM

CHAPTER 10 Volume Analysis

CHAPTER 10 - NAVIGATOR
The Navigator software package is an optional extension of the CT application
software.
The Navigator application allows you to move the viewpoint to locations inside
the 3D volume. You can display the inside of “hollow” structures, providing real-
time luminal views. You can also view the outside of “solid” structures inside the
3D volume, in close-up and in perspective. (With 3D imaging, you construct a
3D object that you can examine from all directions, but the viewpoint is always
located outside the object.)
To define what part of the 3D model will be displayed as “solid”, and what part
will be displayed as “hollow”, you select a Navigator threshold mode (“Black in
white”, “White in black” or “Within borders”) and set the corresponding Navi-
gator threshold(s).
By means of the record-and-playback feature, you can define a path through the
anatomy, and view and record the resulting “fly-through” sequences. You can
vary the view aperture (from narrow-angle to wide-angle views).
Note: With the right settings, you can use Navigator to explore a 3D image set
in a manner very similar to endoscopy. However, Navigator is NOT a
replacement for any endoscopic or angiographic procedures, and is NOT
approved as a screening device.

10-1
2271012-100.book Page 2 Thursday, May 8, 2003 11:30 AM

Volume Analysis

WHEN USING NAVIGATOR SOFTWARE, AN INCOR-

RECT THRESHOLD SETTING OR INCORRECT THRESH-
WARNING! OLD MODE CAN RESULT IN PATHOLOGIES OR OTHER
ESSENTIAL ANATOMY NOT BEING VISIBLE ON THE
NAVIGATOR VIEWS.
ALWAYS CORRELATE THE NAVIGATOR VIEW WITH
THE ORIGINAL DATA (ACQUISITION VIEW, BASELINE
VIEWS).

THE NAVIGATOR SOFTWARE USES A CONICAL PRO-

JECTION (PERSPECTIVE). FOR THIS REASON, THE
WARNING! ASSESSMENT OF DIMENSIONS AND DISTANCES ON
NAVIGATOR VIEWS TENDS TO BE SUBJECTIVE.
ALWAYS CORRELATE APPARENT DIMENSIONS AND
DISTANCES ON THE NAVIGATOR VIEW WITH THE
ORIGINAL DATA (ACQUISITION VIEW, BASELINE
VIEWS).

The Navigator software is not approved as a screening

CAUTION device.

The Navigator software procedures have not been demon-

CAUTION strated to be a replacement for any endoscopic or angio-
graphic procedures.

10-2
2271012-100.book Page 3 Thursday, May 8, 2003 11:30 AM

Volume Analysis

1 INTRODUCTION
The Navigator application generates 3D images of anatomical structures from a 3D
image set. It uses a conical view, as if a camera was placed at a viewpoint inside
the structure.
View layout and view management is the same as for other applications, with a
“Navigator” view at top left and baseline and oblique images in the other three
views.
To move and point the “camera” you use the “navigator” symbol on the Navigator
view. Various controls allow you to “steer” the navigator up/down and left/right, and
move it forward or back.
To initially move the viewpoint to any position in the image set and roughly set the
view direction, you use small simplified navigator symbols (the navigator markers)
on the baseline and oblique views.
When using the Navigator application, you will normally go through the following
steps:
• Select the exam/series in the Image Works Browser, click the button of “VA”,start
the Volume Analysis 2 application, and select Navg Guide in the protocol list
(see paragraph 2),
• Set the threshold mode and threshold(s) to obtain a satisfactory image (see
paragraph 4),
• Explore the anatomical structure of interest with the navigator (see paragraph 5),
• Adjust the image settings as necessary (see paragraph 6),
• Define a FlyThrough path (see paragraph 8),
• Set the movie parameters (see paragraph 9),
• View the sequence (see paragraph 10),
• Save the sequence (see paragraph 11).
Note: In this chapter, “Navigator” (with a capital) refers to the application, while
“navigator” refers to the navigator symbol on the Navigator view.

10-3
2271012-100.book Page 4 Thursday, May 8, 2003 11:30 AM

Volume Analysis

2 START NAVIGATOR
To start using the Navigator application:
• Select the exam/series in the Image Works Browser.
The image set should meet the same requirements as for the other Volume Anal-
ysis 2 applications. See Chapter 4.
• Start the Volume Analysis 2 application, then select [Navg Guide] in the proto-
col list.
− The Navg Guide protocol uses low resolution and is the preferred choice when
exploring larger anatomical structures and when setting up “fly-through”
sequences: processing is faster and small irregularities and “noise” tends to be
smoothed out.
Alternatively, you can create a Navigator view using the (Display Tools) > (View
Layout) controls.

3 CONTROLS
Apart from the “navigator” symbols mentioned in paragraph 1 and further described
in paragraph 5, the Navigator application uses the same types of controls as the
other Volume Analysis 2 applications: active annotations, on-view menus, proto-
col panels and control panels.

Active annotations
The active annotations on the Navg. view that are specific to the Navigator applica-
tion are illustrated below. They are displayed in red as in the other Volume Analy-
sis 2 applications.

10-4
 2271012-100.book
Page 5 Thursday, May 8, 2003
11:30 AM

Navg. P 155 set aperture

Ex: 1234 (30.0 to 120.0 in standard mode)
Se: 3 (120.0 to 360.0 in fisheye mode)
set rendering mode Smooth -524 ^ 90.0
(Fast, Smooth, Integral or VR)

adjust threshold
Cut off
Illustration 1 - Active Annotations on Nave. View

Cut mode set threshold mode

^
(Off or On) ( ^ or or 0 )
Volume Analysis

10-5
2271012-100.book Page 6 Thursday, May 8, 2003 11:30 AM

Volume Analysis

A Navigator view contains the following active annotations:

• Smoothing mode: select Fast, Smooth, Integral or VR (see Section 6, paragraph
Rendering mode below),
• Threshold(s): increment or decrement current value (see paragraph 4 “Set-up
the Images”),
• Threshold mode: select [Black in White], [White in Black] or [Within Borders]
from the drop-down menu (see paragraph 4 “Set-up the Images”),
• Aperture size: Increment or decrement current value,
• Cut mode: select Off or On. If the Cut mode is On an additional active annotation
is displayed at the end of the line, representing the Distance between the Navi-
gator point of view and the Cut.
• Patient Name (top right on the view, not shown in Illustration 1): show or hide the
patient name field.
You use the Navigator active annotations in the same way as those for other view
types. See Chapter 6 “Display and Controls”.
Briefly:
• To change a numerical active annotation:
- Point on it, hold the middle mouse button down and drag left or right, or
- Point on it and click with the left or right mouse button to change it in
steps, or
- Point on it, type a new value on the keyboard and validate with <Enter>
(remember to enter a - for negative values).
• To change a text or symbol active annotation, hold the left or right mouse button
down, and select the desired option from the drop-down menu.
The various controls are described further in this chapter as applicable.

10-6
2271012-100.book Page 7 Thursday, May 8, 2003 11:30 AM

Volume Analysis

On-view menu
Additional Navigator functions can be accessed through the Navigator on-view
menu. Click and hold right on the Navigator view (away from the navigator symbol
or any active annotation) to display the menu, then select the desired function.
[Save image] saves the current view in a separate series in the current exam.
[Light/contrast] opens the Navigator Shading panel, that allows you to change
the brightness and contrast of the Navigator view (Smooth mode only).
[Turnaround] turns the viewing orientation 180 degrees around the vertical axis of
the view (reverses direction of the Navigator).
[Create trace] and the associated [Clear last point], [Clear trace] and [Lock cur-
sor to trace] menu items can be used to create and edit traces on the Navigator
view, as described in Chapter 6 “Display and Controls” paragraph 5.
If you take into account that on a Navigator view the 3D cursor is placed on the
“wall” of the vessel or other anatomical feature being displayed, you can, in the-
ory, create and edit traces, e.g., to perform measurements. In practice, you
should use this feature with care, because on the Navigator view you have no
indication how “deep” into the 3D object the 3D cursor is located, unless you con-
tinuously correlate the Navigator view with the baseline views. The result may
not be at all what you wanted or expected.
Also refer to Chapter 6 “Display and Controls” paragraph “3D cursor on 3D
views”.
[Enlarge] increases the Navg. view to full screen (4x) size. The menu option
changes to [Reset size] which decreases the Navg. view to the original size.
[Lock orientation] prevents rotation and limits Navigator control to forward/back-
ward, up/down/left/right motion. The menu option changes to [Unlock Orientation]
which restores rotation control to the Navigator.
[Center on cursor] centers the view on the current position of the 3D cursor.
[Reset pointer] returns the 3D cursor, and the navigator with it, to its initial display
position in the center of the image set.

10-7
2271012-100.book Page 8 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Protocol and control panels

If you start to use Navigator via the Navg. Guide protocol, the protocol panels for
procedures such as setting up and viewing a FlyThrough sequence are displayed
automatically.
You can remove them from the screen by means of the (Close) button at the bot-
tom of each panel, then return them to the screen by clicking on the (Navigator)
protocol button in the main control panel on the left of the screen.
Alternatively, select the function you require from the (Navg. Tools) menu.
The controls in the Navigator protocol and control panels are described in the fol-
lowing paragraphs.

10-8
2271012-100.book Page 9 Thursday, May 8, 2003 11:30 AM

Volume Analysis

4 SET UP THE IMAGE

At startup the Navigator screen consists of the Navigator view (top left), and the
three baseline views, axial, sagittal and coronal. You can change any of the base-
line views to an oblique view if required.
On the baseline and oblique views the navigator viewpoint (which coincides with
the 3D cursor) is indicated by a yellow dot, and the view direction by a short yellow
line.
The Navigator view shows a three-dimensional image. However this image may
need to be set up in threshold and/or threshold mode.
Go through the following steps to set up the Navigator view:
• On the baseline and oblique views move the 3D cursor inside the structure of
interest (the navigator viewpoint follows the 3D cursor). Point the view direction
marker approximately the right way (click and drag on the end of the short yellow
line to rotate it around the viewpoint).

Black In White • Open the threshold mode drop-down menu using the
active annotation on the Navigator view (see paragraph
White In Black
3), and select the threshold mode:

Within Borders

- [Black in White] “Black in white” mode ( annotation) if on the baseline

and oblique views the lumen (the structure that is to be displayed as
hollow) is darker than the surrounding tissue. An example would be an
airway.
- [White in Black] mode ( annotation) if the lumen (the structure that is to
be displayed as hollow) is lighter than the surrounding tissue (e.g. a
contrast-enhanced vessel).
- [Within Borders] mode (0 annotation) if the lumen (the structure that is
to be displayed as hollow) is defined by a specific range of voxel values,
or, in other words, is surrounded by both lighter and darker structures
that should be displayed as solid.
An example is a contrast-enhanced vessel with calcifications where
both the surrounding soft tissue (darker) and the calcifications
(brighter) should be displayed as solid.
• Now adjust the threshold (or thresholds in the case of the “Within borders” mode)
using the active annotation(s) on the Navigator view, until you obtain a satisfac-
tory visualization of the feature of interest.

10-9
2271012-100.book Page 10 Thursday, May 8, 2003 11:30 AM

Volume Analysis

The easiest way to adjust a threshold annotation on the Navigator view is to click
and drag to the left and right on it (also see paragraph 6).
If you use this method, the range of voxels that will be displayed as hollow is indi-
cated on the baseline and oblique views by green hatching for as long as you
keep the mouse button down.
Note: The software attempts to choose a reasonable threshold when the threshold
mode is changed.
Once you have obtained an initial image in the Navigator view, you still must adjust
the threshold(s) to more accurately match the voxel values of the structures you
want displayed as hollow and solid, respectively (also see paragraph 12).
You can now start “navigating” in the structure, as described in the following para-
graphs.
Note: In terms of voxel values:
- The [Black in White] mode ( annotation) displays all voxels with values
above the threshold as solid,
- The [White in Black] mode ( annotation) displays all voxels with values
below the threshold as solid,
- The [Within Borders] mode (0 annotation) displays all voxels with
values outside the threshold range as solid.
On the view, voxels below the lower threshold are differentiated from
those above the upper threshold by the use of two different colors for
the surface shading. You can use the (Display Tools) > (View Col-
ors) controls to set the color to be used for all objects above the upper
threshold.
Note: Selection of threshold mode and settings depends on various factors, such
as type of exam, type of anatomical structure, acquisition parameters, etc.
Some suggested settings for CT exams are shown below.

Type of Structure Threshold Mode Threshold Value

Lungs, ENT Black in White -500
Colon Black in White -800
Skin Black in White -500
CTA Vessels White Borders Variable
Bone Black in White 160

10-10
2271012-100.book Page 11 Thursday, May 8, 2003 11:30 AM

Volume Analysis

WHEN NAVIGATOR SOFTWARE, AN INCORRECT

THRESHOLD SETTING OR INCORRECT THRESHOLD
WARNING! MODE CAN RESULT IN PATHOLOGIES OR OTHER
ESSENTIAL ANATOMY NOT BEING VISIBLE ON THE
NAVIGATOR VIEWS.
ALWAYS CORRELATE THE NAVIGATOR VIEW WITH
THE ORIGINAL DATA (ACQUISITION VIEW, BASELINE
VIEWS).

10-11
2271012-100.book Page 12 Thursday, May 8, 2003 11:30 AM

Volume Analysis

5 EXPLORE WITH THE “NAVIGATOR”

Once you have set up the Navigator, you can explore the anatomical structures.
To move around inside the anatomical structures, you use the “navigator” markers
on the baseline and oblique views, and the “3D cube” on the Navigator view.
The navigator markers are the yellow dot-and-line markers on the baseline and
oblique views, that indicate the position of the current view point, and the projection
of the view direction on the view. See Figure 4 below.
The 3D cube is the red symbol in the shape of large cross-hairs with attached han-
dles on the Navigator view that you can use to “fly” through the structures, just in
the same way as in any 3D view. The 3D cube is displayed whenever you move the
mouse pointer onto the Navg. view.
The functions of the navigator markers and the navigator are complementary:
- You can move the navigator markers on the baseline and oblique views
rapidly to any location in the image set, but it is not easy to move the
viewpoint marker accurately over a small distance, or point the view
direction marker precisely in a given direction.
- On the other hand, you use the navigator to point and advance
accurately in the anatomical structure of interest, but you cannot use it
to move rapidly to a completely different location in the image set.

10-12
2271012-100.book Page 13 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Navigator markers
You use the navigator markers to move rapidly to any location in the image set, and
point the view direction approximately the right way.

Fig. 4: Navigator markers on axial, sagittal and coronal views

• Move the navigator viewpoint marker in the same manner as you move the 3D
cursor:
- Click and drag on the 3D cursor and yellow dot of the navigation marker,
or
- Point with the mouse to the new position and press the <Shift> key.
• To point the view direction marker:
- Click and drag on the end of the short yellow line to rotate it around the
viewpoint.

10-13
2271012-100.book Page 14 Thursday, May 8, 2003 11:30 AM

Volume Analysis

The navigator symbol

You use the “navigator symbol” on the Navigator view to move and point the “cam-
era” inside the anatomical structure of interest.
The red navigator symbol is displayed whenever you move the mouse pointer onto
the Navg. view.
With the various navigator controls you can move the view direction (line of sight)
up, down, left or right, and move the viewpoint (the position of the camera) forward
or back along the line of sight.

Navg. P 155 Hospital Name

Ex: 1234 PatientName
Se: 3 Sex Age
Smooth-524 ^ 45.0 Date
Navigator symbol

L L
1 1
0 2
1 3D cursor 1

Handles

P 135

View direction controls

To point the navigator in a new direction:
• Click and drag on any one of the handles of the navigator symbol (see Figure 5
on previous page).
If you want to rotate the view:
• Use the (Rotate/Translate) > (By deg) controls.
When you are changing the view direction, releasing the mouse button for a
moment re-aligns the navigator symbol with the new view direction.
To align the view with the axis of the object:
• Press the <L> key on the keyboard.
This points the current view toward the most remote portion of the image, which
aligns the view with the main axis of the object.
You can use this function to realign the view direction after moving the navigator
manually, and to negotiate sharp turns in the sequence.

10-14
2271012-100.book Page 15 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Movement controls
To move the viewpoint forward in steps:
• Click and drag (with the left mouse button) on any one of the handles of the nav-
igator symbol as described above to set the view direction, then click on the mid-
dle mouse button while still holding down the left mouse button.
Alternatively, to move the viewpoint:
• Press the <F> key on the keyboard to move forward in steps.
• Press the <B> key on the keyboard to move backwards in steps.
In particular, if you inadvertently move the navigator “inside object” (hit the “wall”),
the <B> key on the keyboard allows you to step back along the current view direc-
tion, until you are back in the "hollow" part of the structure.
To align the view with the axis of the object, then move the viewpoint forward:
• Press the <P> key on the keyboard.
Note: The step size is determined by the setting (in mm) in the (Rotate/Translate)
> (By mm) control panel.
You may want to make sure that the step size is not set to zero: the naviga-
tor forward and backward movement will appear to be blocked in that case.
Note: For more information on this function, refer to Chapter 6 “Display and Con-
trols”, Section 6 “Smart Cursor”.

Keyboard controls (summary)

As already noted in the previous paragraphs, you can use keyboard shortcuts to
move the navigator, and align it with the main axis of the object. When using any of
the keyboard shortcuts, make sure that the mouse pointer is on the Navigator view.
• <F> (“Forward”) moves the navigator forward by one step,
• <B> (“Back”) moves the navigator backwards by one step,
• <L> (“aLign”) aligns the view direction with the axis of the object,
• <P> (“Path”) aligns the view direction with the axis of the object, then moves the
navigator forward by one step.
Note: The <S> key is used to save the current view to the image disk. See para-
graph 11 “Save”.

10-15
2271012-100.book Page 16 Thursday, May 8, 2003 11:30 AM

Volume Analysis

“Flying” the navigator

You can combine the use of the controls as described above to “fly” the navigator
through the structure of interest.
You “steer” the navigator by holding and dragging with the left mouse button on one
of the handles while at the same time each click on the middle mouse button steps
you forward.
Alternatively you can “steer” the navigator with the mouse and use the <F> and
<B> keys on the keyboard to move forward and backward.
With some practice you can “fly through” and explore a structure easily and accu-
rately.
Note: Rather than clicking each time on the middle mouse button to step forward,
you can even “fly” the navigator through the anatomy in a near-continuous
manner by holding down both left and the middle mouse button.
Releasing the mouse buttons stops the navigator motion.

Rotate/Translate controls
As already described in Chapter 8 “Reformatting” and Chapter 9 “3D Imaging”, you
can use the controls in the Rotate/Translate control panels to change view angle
and move the 3D cursor.
Since the navigator is closely linked to the 3D cursor, you can also use these con-
trols to point and move the navigator.
You can use the (By deg) controls to change the view direction in steps, and the
(By mm) controls to move the view point in steps.
To return the view direction to one of the six standard orientations (Superior, Infe-
rior, Anterior, Posterior, Left or Right).
• Use the (S) (I) (A) (P) (L) (R) buttons on the main control panel.

Navigator markers
As already noted, moving the navigator markers on a baseline or oblique view
moves the viewpoint and view direction on the Navg view. When you move the
mouse pointer back on the Navg view the position and orientation of the navigator
symbol will reflect the new viewpoint and view direction.

10-16
2271012-100.book Page 17 Thursday, May 8, 2003 11:30 AM

Volume Analysis

6 ADJUSTING THE NAVIGATOR VIEW

Rendering mode
Three-dimensional image processing tends to produce complex images, and Navi-
gator is no exception. Because of this, Navigator uses four different types of image
rendering modes:
• two surfacic rendering modes:
- a Fast mode to explore the image set and quickly move to an area of
interest but without displaying a high level of detail. It is mainly useful to
explore very small objects, such as small vessels and in particular on
3DXA images.
- a Smooth mode to produce high-quality rendering of the structures in
the area of interest; it is mainly useful to explore bigger objects, such as
airways or the colon.
• two volumic rendering modes:
- an Integral mode, which produces a gross and quick volumic rendering
of the studied area. It displays the pixels with a certain penetration
thickness. The thickness appears on the view at the end of the line of
the Integral annotation. The value is an active annotation and can
therefore be modified.
- a standard VR mode.
You use the rendering mode active annotation (see illustration in paragraph 3) to
switch between Fast, Smooth Integral or VR modes.
Note: The Smooth rendering mode should not be confused with the Navg. Smooth
protocol at startup (see paragraph 2). Both have a smoothing effect on the
display, but for entirely different reasons. The Navg. Smooth protocol uses a
lower resolution to display the view, while the Smooth rendering mode acts
by filtering out jagged edges and contour effects from the image.
Note: If the amount of image data in the field of view exceeds certain limits the
software automatically switches to a coarse low-resolution display mode to
accelerate the updating of the Navg view while you are modifying the view
point or view direction. Once you have made the changes, a “Hit spacebar
to display high definition view” message is displayed. Press <Space>on the
keyboard to return to the higher resolution view.

10-17
2271012-100.book Page 18 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Cut mode
By default the Cut mode is off.
The Cut on visualization mode allows you to cut a cylinder within the 3D view with
a given depth and a given diameter. It allows you to see what lies behind a superflu-
ous structure, for example to see the lung from inside the bronchis, or to study what
is at the bottom of the cylinder, which is displayed as a 2D view. Window level is
available on this 2D view.

Field of view
The Navigator view is a conical view. By default the aperture of the view cone (view
angle) is 90.
You can modify the field of view by adjusting the aperture active annotation (see
illustration in paragraph 3) from 30.0 (maximum “zoom” effect) to 120.0 (wide-angle
view) and from 120.0 to 360.0 (Fish-eye mode).

THE NAVIGATOR SOFTWARE USES A CONICAL PRO-

JECTION (PERSPECTIVE) THAT CAN INTRODUCE DIS-
WARNING! TORSIONS, ESPECIALLY FOR FISH EYE VIEWS. SUCH
VIEWS SHOULD ALWAYS BE USED IN CORRELATION
WITH 2D BASELINE VIEWS.

As with the DFOV (Display Field of View) function, when you decrease the aper-
ture value, you narrow the field of view and enlarge anatomy.
As soon as you set the aperture value above 120.0, you automatically toggle into
Fish-eye mode. A red circle appears on the view,visualizing the value of the aper-
ture at this spot on the image. At first it represents the 120.0aperture, then another
circle appears when you reach the 180.0aperture, next the aperture shifts while the
image grows smaller. The fish-eye effect allows you to see what is behind your
point of view, as if you were progressively turning a glove inside out.

Reference image
The small square reference image on the Navigator view (normally in the bottom
right corner) shows the projection of the pyramid-shaped view cone on one of the
baseline views (by default the axial view). The base of the pyramid represents the
field of view, the tip of the cone corresponds to the current viewing location.
A pop-up menu (click and hold right on the reference image) allows you to:
• Select which baseline view is used as a reference (axial, sagittal or coronal),
• Move the reference image to another corner of the Navigator view,
• Adjust how much of the baseline view is shown in the reference image (magnify/
unmagnify).

10-18
2271012-100.book Page 19 Thursday, May 8, 2003 11:30 AM

Volume Analysis

7 SYNCHRONIZED NAVIGATION MODE

At startup the Navigator screen consists of the Navigator view (top left), and the
three baseline views, axial, sagittal and coronal. These baseline views are calcu-
lated according to the acquisition plane of the original images.
Activating the Synchronized Navigation mode means that the baseline views repre-
sent instead views calculated in respect with the direction of the Navigator.
Note: When in Synchronized Navigation mode, the navigator markers can no
longer be used to change the viewpoint, as the views are centered on them.
To do that, you must act on the navigator symbol itself.

Starting/stopping Synchronized Navigation

You start the Synchronized Navigation mode by activating the [Start Synchronized
Navigation] function of the Navigator on-view menu. The menu option changes to
[Stop Synchronized Navigation] which allows the Navigator to return to standard
baseline views.

Synchro-Axial View
When the Synchronized Navigation mode is selected, the first 2D view toggles to
the oblique view transversal to the direction of the Navigator.

Synchro-Sagittal View
When the Synchronized Navigation mode is selected, the second 2D view toggles
to the oblique view parallel to the direction of the Navigator, in th evertical plan.

Synchro-Coronal View
When the Synchronized Navigation mode is selected, the third 2D view toggles to
the oblique view parallel to the direction of the Navigator, in the horizontal plan

10-19
2271012-100.book Page 20 Thursday, May 8, 2003 11:30 AM

Volume Analysis

8 DEFINE THE FLYTHROUGH PATH

Paragraph 5 described how to view and move through the anatomical features of
interest step by step using the navigator.
You can also define a path through the structure, and have the software calculate a
“FlyThrough” movie sequence. You can then view this sequence repeatedly, pause
and restart it, modify the display speed, and/or save the sequence as a separate
series.
You do not need to define each view in the sequence, but only the points (steps) on
the trajectory of the “movie camera” where the view direction changes. The Navi-
gator software interpolates frames between the designated locations to smooth out
the FlyThrough sequence. The software accepts up to 256 defined steps per Fly-
Through sequence.
To prescribe a path through the structure, first activate the FlyThrough sequence
controls (via (Navg Tools) > [Navg Path] or from the Navigator protocol). You can
now insert steps in the FlyThrough sequence prescription, placing the steps either
manually with the (Insert Step) command or semi-automatically with the (Insert &
Seek Step) command.
Note: See paragraph 14 for a list of possible error messages.

THE NAVIGATOR SOFTWARE USES A CONICAL PRO-

JECTION (PERSPECTIVE). FOR THIS REASON, THE
WARNING! ASSESSMENT OF DIMENSIONS AND DISTANCES ON
NAVIGATOR VIEWS TENDS TO BE SUBJECTIVE.
ALWAYS CORRELATE APPARENT DIMENSIONS AND
DISTANCES ON THE NAVIGATOR VIEW WITH THE
ORIGINAL DATA (ACQUISITION VIEW, BASELINE
VIEWS).

Insert Step
(Insert Step) adds the current location of the navigator (view point and view direc-
tion) to the path prescription. You use this command to insert the starting point for
the sequence, then to define steps in the path prescription manually.
• You can add locations to a prescription in progress. The software automatically
inserts the step into the correct place in the sequence.
• The software will warn you if you try to insert a new step too close to an existing
step in the current sequence.

10-20
2271012-100.book Page 21 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Insert & Seek Step (recommended)

(Insert & Seek Step) adds the current location of the navigator to the path prescrip-
tion, moves 1 cm to the next location in the current viewing direction, and then
adjusts the view to align it to the axis of the structure. You use this command to
quickly prescribe a path through the anatomy of interest.
Note: If you insert a step that flies the Navigator too close to an object, the soft-
ware automatically selects a smaller step, to avoid collision.

Other controls
(Delete Step) removes the current step from the prescription.
(Backtrack) allows you to move back to the last prescribed step of the path.
(Align to Axis) rotates the current view toward the most remote portion of the
image, which aligns the view with the main axis of the object. You can use this
function to realign the view direction after moving the navigator manually, and to
negotiate sharp turns in the sequence.
Note: You can also use the <L> key on the keyboard.
(Show/Hide Path) controls whether the prescribed path is displayed on the non-
Navg. views.
The {Step slider} indicates the current location in the FlyThrough sequence. As
soon as you have defined more than one point, you can use this slider to rapidly
move to a particular step in the sequence, e.g. to insert an intermediate step or to
delete a step. The display updates accordingly.

10-21
2271012-100.book Page 22 Thursday, May 8, 2003 11:30 AM

Volume Analysis

9 SET MOVIE PARAMETERS

Before actually viewing a “FlyThrough” movie sequence, you must set up the movie
parameters:
• Sampling distance,
• Display speed,
• Multiview mode on/off,
• Round trip on/off.

Sampling
Use the {Sampling} slider to select the default distance, in millimeters, between
each view in the FlyThrough sequence, which helps determine the number of
images that the software interpolates between each step.
You can adjust the default distance between 0.5mm and 20.0mm. The software will
automatically adjust the actual distance between views where necessary to opti-
mize the FlyThrough path.
You must set this control before using (Navg Options) > [Navg Cine] > [Fly
through] > (View Sequence) to compute and view the FlyThrough sequence.
Note: The setting of the {Sampling} slider determines approximately how many
images there will be in the sequence (number of images = length of path
divided by sampling distance).
You should make sure the {Sampling} slider setting corresponds to your
requirements. Too small a sampling distance increases the time required to
compute and display the sequence, without necessarily contributing addi-
tional information.
As an example, using the minimum sampling distance of 0.5mm over a
10cm path will produce 200 images. Increasing the sampling distance to
1mm reduces this to 100 images, and halves both the computing and the
display time of the sequence.
Note: If the sequence is too large for the available workstation memory this will be
indicated by a warning message. When this happens you can:
• Increase the sampling distance, to reduce the number of images in the
sequence, and/or
• If you have used the Navg Guide protocol to build the 3D model (see para-
graph 2), or
• Accept the sequence “as is”: the sequence will be displayed, but much more
slowly, because the system will need to swap images between memory and
image disk.

10-22
2271012-100.book Page 23 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Display speed
Use the {Display speed} slider to set the speed in images/sec at which the view
sequence will be displayed. The default setting is 10 images/sec, but you can
adjust it between 1 to 30 images/sec.
You can adjust this control while the FlyThrough sequence is being displayed.

Multiview mode
Multiview mode simultaneously updates the images in up to four views during the
FlyThrough sequence, screen saving or filming. This feature changes the Navg.
view to a four-on-one Multiview format that displays reduced-size versions of the
Navg., Sagittal, Coronal, Axial, or selected 2-dimensional views. During the FlyTh-
rough sequence,
the reduced-size 2-dimensional views update in sync with the 3-dimensional Navi-
gator image. The normal-sized Axial, Sagittal and Coronal planes do not update
until you pause or stop the FlyThrough sequence.
• When you pause the FlyThrough sequence for more that a few seconds, the nor-
mal-sized Navg. image replaces the four-on-one Multiview image.
• When you restart the sequence, the four-on-one Multiview image replaces the
normal-sized Navg. image.
• The Multiview image returns for a moment when you step forward and pause, or
step backward and pause. The view changes back to the Navigator image, and
the other three views also update, each time you step and pause.
You must select whether to use Multiview mode BEFORE using (Navg Tools) >
[Navg Cine] > [Fly through] > (View Sequence) to compute and view the FlyTh-
rough sequence (the system ignores the state of the (Multiview mode) control
afterwards).
You select or cancel Multiview mode from the [Navg Tools] > [Navg Cine] panel.
Note: The system requires significantly more time to compute a Multiview mode
sequence. The software displays and updates messages (“xx images of xx
to compute” and “xx min xx sec remaining”) to keep you informed of its
progress.

Round trip
• To display the view sequence from first to last image, then from last to first (for-
ward/backward), set (Round Trip) to off.
• To display the view sequence from first to last image, then start again with the
first image (i.e., as a loop), set (Round Trip) to on.
You must set this control BEFORE using (View Sequence) to compute and view
the FlyThrough sequence.

10-23
2271012-100.book Page 24 Thursday, May 8, 2003 11:30 AM

Volume Analysis

10 VIEW SEQUENCE
Once you have defined the path (see paragraph 8) and set the movie parameters
(see paragraph 9), you can compute, view and save the FlyThrough sequence.
You should note that you cannot modify a FlyThrough sequence. However, you
can at all times stop the FlyThrough sequence display, and modify the path defini-
tion, or change the movie parameters, then recompute the sequence.

View Sequence
Select (Navg Tools) > [Navg Cine] > [Fly through] > (View Sequence) to start
the computation of the images that compose the sequence.
During the computation a small pop-up window indicates progress. The (Stop) but-
ton in this window allows you to interrupt the computation if necessary.
As soon as all images have been computed, the FlyThrough movie sequence starts
automatically.
Note: If you have selected Multiview mode (see paragraph above) the system
requires significantly more time to compute the sequence. In this case, the
software displays and updates messages indicating “xx images of xx to
compute” and “(xx min xx sec remaining)” to keep you informed of its
progress.
Note: During the computation of the sequence a warning message may indicate
that the sequence will "hit the object" (run into the 3D surface) if it follows
your current FlyThrough sequence prescription.
You can accept this: in that case the software will fly into the 3D surface, per
your request, and "Inside object" will be displayed in the view sequence for
the section of the path inside the 3D object.
Otherwise, return to the Navg path > Define Path controls (see paragraph
8), delete one or more steps as necessary and insert one or more steps
close to the current position to prevent the collision, then select (Navg
Tools) > [Navg Cine] > [Fly through] > (View Sequence) again to restart
the computation of the sequence.

10-24
2271012-100.book Page 25 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Viewing controls
Once the FlyThrough view sequence is computed, you can use the viewing con-
trols:
• (Pause/Restart) button
• {Image Index} slider,
• (Stop) button.

Pause/Restart
• (Pause) simply pauses the FlyThrough movie display. The button label changes
to (Restart), and a second click on the button resumes the FlyThrough movie
display,

Image Index Slider

Grab the {Image Index} slider to pause the FlyThrough movie display (or use the
(Pause) button).
Click and hold on the slider, then drag it to the left or right to to display the corre-
sponding image in the sequence.
Note: When using the (Pause/Restart) and step buttons, or the {Image Index}
slider:
- After you step forward or step backward and pause, the software
updates all the non-FlyThrough views about 2 seconds later.
- In Multiview mode, the Multiview image returns for a moment when you
step forward and pause, or step backward and pause. The view
changes back to the Navigator image, and the other three views also
update, each time you step and pause.
- You can pause the FlyThrough movie display, and manipulate the
navigator on the Navg. view. When you restart the sequence, the
software automatically returns to the original pause point, and resumes
the FlyThrough movie display.

10-25
2271012-100.book Page 26 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Stop
Once the view sequence is computed, it can no longer be modified. The label on
the (View Sequence) button changes to (Stop).
You use this control to stop the FlyThrough movie display.
This allows you to return to the Define Path controls (paragraph 8) and modify the
path, or to modify the movie parameter settings (paragraph 9), then recompute the
sequence.
You also must stop the FlyThrough movie display if you want to save the FlyTh-
rough sequence on the image disk (see paragraph 11).
• Click on (Stop) to stop the FlyThrough movie display. Once you stop the display,
you can no longer use (Restart) to resume the display.
The label of the (Stop) button changes back to (View Sequence). When you
reselect this button, the view sequence is recomputed entirely.

Distance annotation
Once the view sequence is computed, a distance annotation is added on the view
(near top left) which indicates the distance from the starting point along the fly-
through path. This is particularly useful while using (Pause) and (Step) to accu-
rately locate a particular view along the path.

10-26
2271012-100.book Page 27 Thursday, May 8, 2003 11:30 AM

Volume Analysis

11 SAVE
After you have prescribed a FlyThrough sequence and used (View Sequence) to
compute the views in the sequence, you can save individual images from the
sequence, or save the entire current FlyThrough sequence as a new series of
screen save images.
• To save individual images from the sequence, (Pause) the display, and if nec-
essary use the (<) and (>) controls to display the exact image you want, then use
[Save image] in the on-view menu, or the <S> key on the keyboard to save the
image in screen save format on the image disk.
• To save the entire FlyThrough sequence, (Stop) the display. You can now use
(Save Sequence) to save the current FlyThrough sequence as a new series of
screen save images.
Note: To save Navigator movie sequences on the image disk, make sure before-
hand that you have adequate free disk space available. A FlyThrough
sequence may contain several hundred images, hence it can take up a sig-
nificant amount of disk space.
Note: Once you (Stop) the display, you can no longer (Restart) it.

10-27
2271012-100.book Page 28 Thursday, May 8, 2003 11:30 AM

Volume Analysis

12 USING NAVIGATOR
With the appropriate settings, you can use Navigator to explore a 3D image set in a
manner very similar to endoscopy. However, as noted, Navigator is NOT a replace-
ment for any endoscopic or angiographic procedure, and is NOT approved as a
screening device.
When we use the terms “hollow” and “solid” you should keep in mind that these are
artificial. What appears in the Navigator view as “solid” and what appears as “hol-
low” is totally determined by the threshold mode and threshold setting.
As an example, take a contrast-enhanced vessel that appears bright among the
soft tissue that surrounds it.
With “white in black” mode and the right threshold setting, the “inside” of the vessel
will be displayed as hollow and the Navigator allows you to explore this “inside”.
The wall of the vessel, and the vessel contents, are not of uniform density, so when
you increase the threshold the vessel will appear narrower. Increase the threshold
even more and the vessel will appear to start filling with floating lumps. Physically
these lumps are real: they indicate the statistical variations (noise) of the density of
the fluid filling the vessel, but anatomically they have no significance.
On the other hand, lower the threshold too far, and “holes” will start appearing in the
“walls” of the vessel because the software starts “seeing” part of the vessel walls
and the surrounding soft tissue as “hollow”. Again, the holes you see are not an
anatomical reality.
In the same data set, switch to “black in white” mode, move the Navigator viewpoint
outside the vessel, and adjust the threshold as necessary. Now the soft tissue is
treated as “hollow”, the vessel as “solid”, and you’re looking at the “outside” of the
vessel! Again, adjusting the threshold determines what exactly you will see.
In other words, the threshold mode and threshold setting give you complete control
over what is displayed as “solid” and what is displayed as “hollow” in the Navigator
view. You could even display solid bone as a hollow space, if that was what you
wanted!
Because it gives you such complete control over the viewpoint and over what is dis-
played (through the threshold mode and threshold setting), Navigator can be a
powerful tool for exploring details in a 3D data set, in a manner not available from
other applications.
BUT, you should use it with common sense. You should remain aware that what is
being displayed is totally determined by the display parameters you have chosen.
Navigator allows you to “look” at a 3D data set in a manner unlike other applica-
tions. This can give you valuable added insights during a diagnosis. Nevertheless
you should always correlate what you see on the Navigator view with the original
data (acquisition, baseline and oblique views).
Because of the use of perspective and varying view angles, the assessment of
dimensions and distances on Navigator views tends to be subjective. To perform

10-28
2271012-100.book Page 29 Thursday, May 8, 2003 11:30 AM

Volume Analysis

measurements, use baseline or oblique views. Navigator views are 3D views and
should NOT be used for measurements without continuous correlation with the
baseline views.

WHEN USING THE NAVIGATOR SOFTWARE, AN

INCORRECT THRESHOLD SETTING OR INCORRECT
WARNING! THRESHOLD MODE CAN RESULT IN ESSENTIAL
ANATOMY, OR PATHOLOGIES, NOT BEING VISIBLE
ON THE NAVIGATOR VIEWS.
ALWAYS CORRELATE THE NAVIGATOR VIEW WITH
THE ORIGINAL DATA (ACQUISITION VIEW, BASELINE
VIEWS).

THE NAVIGATOR SOFTWARE USES A CONICAL PRO-

JECTION (PERSPECTIVE). FOR THIS REASON, THE
WARNING! ASSESSMENT OF DIMENSIONS AND DISTANCES ON
NAVIGATOR VIEWS TENDS TO BE SUBJECTIVE.
ALWAYS CORRELATE APPARENT DIMENSIONS AND
DISTANCES ON THE NAVIGATOR VIEW WITH THE
ORIGINAL DATA (ACQUISITION VIEW, BASELINE
VIEWS).

10-29
2271012-100.book Page 30 Thursday, May 8, 2003 11:30 AM

Volume Analysis

13 ERROR MESSAGES
This paragraph lists the error and warning messages that may be displayed while
you are defining or displaying a FlyThrough sequence, with suggestions for correc-
tive action.
Create a Navigator view: You tried to create a FlyThrough sequence while no
Navg. view is displayed. Create a Navg. view.
Select one Navigator view or Select only one Navigator view: You tried to cre-
ate a FlyThrough sequence with two or more Navg. views on display. Display a sin-
gle Navg. view.
Only 256 steps can be defined: A FlyThrough sequence can contain at most 256
steps; you tried to define a 257th step. Use (Delete Step) or (Backtrack) to
remove steps.
The new point is too close to an existing point: You tried to prescribe a step
less than one voxel away from a previously defined step. Use (Delete Step) or
(Backtrack) to remove the step and try again.
Use the (Insert & Seek Step) function to prevent this problem.
Try to define points more evenly: While you prescribe steps in a FlyThrough
sequence, the system monitors the average distance between steps. It warns you
when you try to prescribe a step less than one third the average distance, or larger
than three times the average distance.
Use the (Insert & Seek Step) function to prevent this problem.
Path will hit the object: The navigator will run into the 3D surface if it follows your
current FlyThrough sequence prescription. The software repositions the cursor and
prompts you to accept the current position, or to insert a step close to the current
position, to prevent a collision. However, if you insist, the software will fly into the
3D surface, per your request, and “Inside object” will be displayed in the view
sequence for the section of the path inside the 3D object.
Could not insert this point: avoid sharp turns: You asked the impossible. The
navigator should fly smoothly through the object. If, for example, you prescribe a
new step between two previously prescribed steps, you are asking the navigator to
rapidly change directions, and the software cannot find a satisfactory insertion point
in the sequence. Use (Delete Step) or (Backtrack) to remove the step and try
again. Use the (Insert & Seek Step) function to prevent this problem.
Cine display of movie is going to be slow (%d images)
use low-resolution mode to build the object or use larger sampling step
( )=%f mm ):
The memory requirements of the FlyThrough Sequence prescription exceeds cur-
rently available memory space.

10-30
2271012-100.book Page 31 Thursday, May 8, 2003 11:30 AM

Volume Analysis

The software prompts you to accept the slow version of the movie, or cancel the
current movie and change one or more parameters.
Accept the slow movie mode, or cancel the movie and change the 3D model build
protocol to Navg Guide, or set the {Sampling} slider to a larger sampling distance.
Cine display of movie is going to be slow (%d images)
use larger sampling step ( ) =%f mm ):
The memory requirements of the FlyThrough Sequence prescription exceeds cur-
rently available memory space, and you already set the 3D model build protocol to
Navg Smooth. Accept the slow movie mode, or cancel the current movie and
change the {Sampling} slider to a larger sampling distance.

10-31
2271012-100.book Page 32 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Blank page.

10-32
2271012-100.book Page 1 Thursday, May 8, 2003 11:30 AM

CHAPTER 11 Volume Analysis

CHAPTER 11 - VOLUME RENDER-

ING
Volume Rendering is an optional software package to be used with Volume
Analysis 2 .
In general terms, “rendering” refers to the techniques used to represent a 3D
(three-dimensional) object on a two-dimensional surface.
The basic Volume Analysis 2 application uses two main rendering techniques:
surface shading, which treats the 3D object as a solid and fully opaque object,
and projection shading, which treats the 3D object as if it were translucent.
With the Volume Rendering option, volume rendering is added to these render-
ing techniques.
Volume rendering uses the notion of opacity, which can be defined as the
degree to which light cannot penetrate an object. Different “opacity” values are
assigned to different voxel values which can represent different tissue properties
such as density.
The effect is to make highly opaque objects more clearly visible among less
opaque objects which appear transparent to a greater or lesser degree. The
result is the ability to see many layers of tissue, instead of only the first layer as
with surface shading.
Volume rendering can be used in three different modes: black-and-white shad-
ing, color shading, and multiple object color coding.

WHEN USING THE VOLUME RENDERING SOFT-

WARNING! WARE OPTION, AN INCORRECT THRESHOLD SET-
TING, INCORRECT CURVE TYPE (THRESHOLD
MODE), OR INCORRECT OPACITY SETTING, CAN
RESULT IN ESSENTIAL ANATOMY, OR PATHOLO-
GIES, NOT BEING VISIBLE ON THE VIEWS DIS-
PLAYED IN VOLUME RENDERING MODE.
ALWAYS CORRELATE VIEWS THAT USE VOLUME
RENDERING WITH THE ORIGINAL DATA (ACQUISI-
TION VIEW, BASELINE VIEWS).

11-1
2271012-100.book Page 2 Thursday, May 8, 2003 11:30 AM

Volume Analysis

1 HOW TO USE THIS CHAPTER

For an introduction to the underlying principles of volume rendering, read para-
graph 2 “Introduction”.
To start using volume rendering on 3D or MPVR oblique views, refer to paragraph 3
“Using Volume Rendering”.
For the controls and functions used by the Volume Rendering software, refer to
paragraph 4 “Controls”.
For a list of the terms that are specific to the application and to volume rendering
techniques, refer to the Glossary at the back of the manual.
Note: In this chapter, “Volume Rendering” (with capitals) refers to the software
option, while “volume rendering” refers to the volume rendering techniques
as used on the 3D and MPVR oblique views.

11-2
2271012-100.book Page 3 Thursday, May 8, 2003 11:30 AM

Volume Analysis

2 INTRODUCTION
Rendering
In general terms, “rendering” refers to the techniques used to represent a 3D
(three-dimensional) object on a two-dimensional surface.

As a example, if a sphere is projected on a flat surface, the

result is a circle. Any indication of it being a 3D object is lost.

In the simple illustration at the left, a few levels of shading are

sufficient to provide depth cues and restore the impression of a
sphere. Here, the shading represents the amount of light that
would be reflected from a light source above and to the left of
the sphere.

11-3
2271012-100.book Page 4 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Rendering Techniques
In medical imaging, it will be obvious that the complexities of shape, spatial orienta-
tion and tissue properties require more sophisticated display and rendering tech-
niques than in the simple illustration above.
The aim is to extract as much information as possible from a three-dimensional CT,
MR or 3DXR data set displayed on a two-dimensional workstation screen.
To this purpose, Volume Analysis 2 combines various types of rendering with
other techniques such as volume segmentation (thresholding, scalpel) to define a
3D object, rotation of the 3D object on the screen to view it from different angles, or
perspective with the Navigator option.
The basic Volume Analysis 2 application uses two main rendering techniques:
• Surface shading, which treats the 3D object as a solid, fully opaque object.
• Projection shading, which treats the 3D object as if it were translucent, using
different projection algorithms, such as Maximum Intensity Projection (MIP), Min-
imum Intensity Projection (MinIP) and RaySum.
With the Volume Rendering option, volume rendering is added to these rendering
techniques.
Volume rendering uses the notion of opacity, which can be defined as the degree
to which light cannot penetrate an object. In the context of medical imaging, differ-
ent opacity values are assigned to different voxel values which can represent differ-
ent tissue properties such as density.
The effect is to make highly opaque objects more clearly visible among less opaque
objects which appear transparent to a greater or lesser degree. The result is the
ability to see many layers of tissue, instead of only the first layer as with surface
shading.
Volume rendering can be used in three different modes:
• Black-and-white shading Color shading
• Color shading
• Multiple object color coding

11-4
2271012-100.book Page 5 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Volume Rendering, Overview

A brief overview of the rendering process may help in understanding the volume
rendering technique.
Digital medical image data are acquired and manipulated in a matrix of volume ele-
ments called “voxels”. An image is constructed by analyzing each voxel and pro-
jecting the result on a two-dimensional surface subdivided in picture elements
called “pixels”. In 3D imaging, given the tissue properties and information about the
perspective of the viewer, the resulting image shows the result of how light behaves
as it passes through the volume of voxels.
With a 3D model defined in terms of opacity, rays of light will penetrate the tissue in
varying degrees. At each point, a certain amount of light is reflected and the rest is
transmitted to the next layer where the same process occurs. In the end, the
viewer sees an image which is the sum of reflections from each layer. This differs
from the image of a surface-shaded 3D model where the light rays are entirely
reflected by the first surface they encounter.
Because the transition between what can and cannot be seen is now a gradual
one, and not abrupt as in surface rendering, the user should find it faster to define
suitable views.
Volume rendering allows the simultaneous display of objects with different proper-
ties.
The simplified example below should help illustrate this.
In this example, we consider a CT contrast study where we seek to display a con-
trast-enhanced blood vessel among the surrounding soft tissue.

A Illustration (A) depicts a vessel embedded in soft tissue of a

lower density.
Using thresholding techniques and surface shading, the vessel
and surrounding soft tissue cannot be displayed clearly at the
same time.

B If the threshold is set low enough to show the soft tissue, then
the vessel is hidden except at the cut plane (B).

11-5
2271012-100.book Page 6 Thursday, May 8, 2003 11:30 AM

Volume Analysis

C If the threshold is set high enough to clearly show the vessel,

then the soft tissue is no longer visible (C).

D Rendering modes such as Maximum Intensity Projection (MIP)

can be used to display structures of different density. However,
the structures are displayed as projections on the surface of the
object and all depth cues are lost (D). In addition, since MIP dis-
plays the brightest, or most dense, object in the ray path,
“bright” structures such as bone tend to obscure any other
structure present.

E Volume rendering can be used to show both objects at the

same time. By assigning a low opacity to the soft tissue and a
high opacity to the dense vessel, the vessel becomes clearly
visible among its semi-transparent surroundings (E).

The example is simplified because it treats the two objects as though they have uni-
form and distinctly different densities. In reality, any single object is likely to have a
non-uniform density, and the difference in density between multiple objects is often
not enough to make them clearly distinguishable.
Therefore, the process of segmenting tissue based on voxel value threshold can be
difficult. If the threshold is set too low, some of the unwanted surrounding tissue will
appear on the image and mask the vessel behind it. If the threshold is set too high,
part of the vessel will not be displayed.
The volume rendering technique is more tolerant to variation in tissue properties.
Where thresholding either entirely shows or hides tissue of a given density, volume
rendering allows the tissue to "fade" from visible to invisible.
The following paragraphs provide a more detailed explanation of how this works.

11-6
2271012-100.book Page 7 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Volume Rendering, Principles

Volume rendering is a 3D visualization technique which uses the data from the
entire volume to create an image. Information from every point along a ray path is
used to determine the resulting reflection along that path. As light travels through
the volume, the opacity at each point determines the amount of light which is
reflected from that point. The remaining light travels to the next layer where the
process is repeated.
In medical imaging, each element of volume is represented by a voxel. The reflec-
tion which creates the final image is the result of the light reaching a voxel times the
opacity of the voxel, or, expressed as an equation:
reflection = (amount of light) x (opacity).
The figure below illustrates this.
The row of shaded squares represents voxels within tissue of increasing density.
The graphs show the opacity value, the light that reaches each voxel, and the
resulting reflection.

The opacity of the voxel

defines how visible it is.
Light
− An opacity of 0 (zero) is
light reaching each voxel assigned to completely
transparent voxels, that will
not be seen on the image.
opacity − An opacity of 1 (one) is
assigned to completely
opaque voxels that do not
transmit light but completely
reflect it, and hence appear
reflection solid.
− Voxels with intermediate
1 2 3 4 5 6 7 8 9
depth
opacity values will appear
semi-transparent.
The amount of light reaching
each voxel decreases as it
passes deeper into the tis-
sue. Each voxel reflects some
of the light, and only the
remaining light reaches the
next layer.
The reflected light is used to
generate the final image
which the user sees.

11-7
2271012-100.book Page 8 Thursday, May 8, 2003 11:30 AM

Volume Analysis

The final result now depends on the shading technique:

• Black-and-white shading
(voxel value averaging)
The shading value for a
voxel is simply defined by
its value, i.e., its opacity.
The final result is the
weighted average of voxel
values along each ray path.
Also see paragraph 5 of this
chapter.
In practice, the most signifi-
cant contribution will come
from voxels located close to
a surface where the tissue
property is the same. Edges
will be visible as dark lines
because a ray will strike a
larger number of partial vol-
ume voxels.

This type of shading is useful for vascular or bone display from CT acquisitions.

• Color shading
The shading value for a
voxel is defined by its value,
i.e., its opacity, and the local
orientation of the surface at
the location of the voxel.
The color is based on the
voxel value.
Also see paragraph 6 of this
chapter.

11-8
2271012-100.book Page 9 Thursday, May 8, 2003 11:30 AM

Volume Analysis

• 3D object color coding

As with color shading (see
above), the shading value
for a voxel is defined by its
value, i.e., its opacity, and
the local orientation of the
surface at the location of
the voxel.
The color is based on the
color assigned to the
attached object, or objects
in the case where multiple
objects are defined.
Also see paragraph 7 of this
chapter

11-9
2271012-100.book Page 10 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Comparison with other Rendering Techniques

Surface Shading
Surface shading is a 3D visualization technique where all objects are fully opaque.
As a consequence, surface shading is more sensitive to threshold or classification
choices.

The choice of a low threshold may

hide objects by including more tis-
sue than desired.
If the threshold is too high (under-
estimation of objects), part of the
object of interest may be lost.

By comparison, volume rendering will only show faint traces of tissue if there is a
distinguishable difference in tissue property.

MIP
The MIP (Maximum Intensity Projection) rendering mode displays only the maxi-
mum pixel intensity along a ray path.

This mode may be used to render

vascular structures. No depth
cues are provided, making it diffi-
cult to display complex structures.
In CT Angiography (CTA), bright
values from bone will hide vessels
in front or behind them. In MR
Angiography (MRA), larger ves-
sels may hide thinner vessels.

By comparison, volume rendering is often better suited to the display of vascular

anatomy, because it is less sensitive to threshold choices than either surface shad-
ing or MIP. For angiography, volume rendering can display thinner vessels in front
of larger ones and will provide the correct depth cues without surface shading.

11-10
2271012-100.book Page 11 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Partial Volume Effect

Overview

The “partial volume effect” refers

to the appearance of voxels with
Bone intermediate values at the inter-
Fat (0 HU)
Fat Fat face of two dissimilar tissue
types.
Vessel or fat/bone
interface (300 HU) A common example occurs dur-
ing CT Angiography exams,
Bone (600 HU) where the typical value for vas-
culature (in the illustration 300
"Vessel" artifact HU is assumed) is intermediate
between fat (0 HU assumed)
and bone (600 HU assumed).
As a result, artificial “vessels”
may appear at the interface
between fat and bone, where the
average of these two tissues is
in the range of vasculature.
In many cases this makes it diffi-
cult to select a single class of tis-
sue, because partial volume
artifacts exist in the desired
range of values.
Note: The HU values assumed
for the different tissue
types in the illustration
are to be regarded as
examples only. Real val-
ues may differ signifi-
cantly.

11-11
2271012-100.book Page 12 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Partial Volume Filter

The Partial Volume Filter attenuates the partial volume by associating partial vol-
ume voxels with adjacent voxels which represent the more opaque object. In the
illustration above, the value of the nearby bone voxels would be used for the fat/
bone interface.
The filter has two effects:
• It improves the visibility of structures that would otherwise be hidden by partial
volume voxels.
• It improves the color coding of voxels in color mode.
See paragraph “Options” bellow on how to activate the Partial Volume Filter.

11-12
2271012-100.book Page 13 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Transferring VR Models Between Views

Overview
You can transfer a VR model from one view to another using the same “drag and
drop” principle as for icons described above.
Again,, when you transfer a VR model from one view to another, it will take on the
properties of the destination view. For example, if a VR model is transferred to a VR
view it will be displayed in VR, but if it is transferred to a coronal view it will be dis-
played as a coronal slice.
The display parameters (window level/width, cut planes, colors, rendering mode,
etc.) will be those of the view into which it is being transferred.
Objects can be transferred between views in one of two ways:
• Copying of an object into a view with deletion of the original object in that view,
• Merging together of one VR object from one VR view with another VR object in
another VR view.
Note: This function can only be executed if both objects belong to the same vol-
ume.

Copying a VR model from one view to another

To copy a VR model from one view to another, with deletion of the original VR
model in the target view, proceed as follows:
• Press and hold left on the view containing the VR model to be moved (but not on
the trackball, VR cursor or red annotations),
• Drag the VR model to the desired target view,
• Drop the VR model into the flashing box marked “Drop here to reassign view” by
releasing the left button.
The dropped VR model replaces the prior content of the target view.

11-13
2271012-100.book Page 14 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Merging VR models
To merge a VR model in one view with a VR model in another view, proceed as fol-
lows:
• Press and hold left on one of the views containing a VR model to be merged (but
not on the trackball, VR cursor or red annotations),
• Drag the model to the desired target view containing the other VR model to be
merged,
• Drop the model into the flashing box marked “Drop here to merge views” by
releasing the mouse button.
The dropped VR model is now merged with the prior content of the target view.
Also, the Focus status annotation appears in the upper left corner of the view. See
paragraph “Focus Status Annotation (Merged VR Views Only)” below for use of
merged VR model control functions.
Note: A maximum of eight VR models can be merged.

11-14
2271012-100.book Page 15 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Focus Status Annotation (Merged VR Views Only)

If you have merged two or more VR models together (see “Transferring Objects
Between Views” above), the focus status annotation appears in the upper left cor-
ner of the VR view containing the merged VR models. See illustration.
The default setting of the transparency status annotation for merging VR models is
Transparent.

Merged VR model selection functions can be accessed

by pressing right on this annotation and selecting one
of the choices from the menu that pops up:

Change Focus Object

Keep Focus Object Only
Remove Focus Object Only

[Change Focus Object] is used to select another merged VR model for display
mode manipulation (window level/width, cut planes, colors, rendering mode, etc.).
The newly-selected VR model flashes briefly in green to indicate that it is selected.
[Keep Focus Object Only] is used to delete all merged VR models EXCEPT the
selected one.
[Remove Focus Object Only] is used to delete ONLY the selected merged VR
model.
Note: The selected merged VR model by default following the merge operation is
the VR model that was in the TARGET view prior to merging. This remains
true until you either move the VR cursor or use [Change Focus Object] in
the menu described above.

MAKE SURE THAT PATHOLOGIES AND OTHER

WARNING! ESSENTIAL ANATOMICAL STRUCTURES ARE
CLEARLY SHOWN WHEN USING MERGED VR MODEL
VIEWS FOR DIAGNOSTIC PURPOSES.

11-15
2271012-100.book Page 16 Thursday, May 8, 2003 11:30 AM

Volume Analysis

3 USING VOLUME RENDERING

Overview
Within the Volume Analysis 2 application you can use volume rendering on 3D
views and on MPVR oblique views.
• To use volume rendering on 3D views, you can select the “VR” (Volume Render-
ing) protocol in the “General” category after startup of Volume Analysis 2.
• To use volume rendering on an MPVR oblique view (“thick” slice), you must first
set up the view, then switch the rendering mode to “volume rendering”.
Note: Any build protocol can be used with volume rendering, although the use of
the “VR”, “Reformat”, “Angio” or the optional “Navigator” and “Navigator
Smooth” protocols is recommended.
Setting the model mode in a "Custom" protocol to “Volume”, or using proto-
cols such as “CT Bone” that use the “Volume” model mode , may at times
result in very large volumes of data and poor system performance. Memory
requirements may be above system limits.
Note, that the “Volume” model mode adds data in the 3D model that are
used for surface shading, and that Volume Rendering does not require
these data.
See Chapter 5 “Building the 3D Model” for more details.

Volume Rendering on 3D Views

At startup
To obtain a 3D view in volume rendering mode at startup,
• Start Volume Analysis 2 from the Image Works Browser as usual,
• Select (General) as the anatomical region category,
• Select and apply the [VR] protocol.

11-16
2271012-100.book Page 17 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Main screen
On the main Volume Analysis 2 screen (control panel and views), to switch a 3D
view to volume rendering mode:

• On the view, click and hold left or right on the rendering mode
active annotation (near top left) and select [Volume Render-
ing] in the drop-down menu,
OR

• On the view, click and hold left or right on the view type active
annotation (top left) and select [VR] in the drop-down menu.
The view type will still indicate “3D” but the rendering mode
switches to “Volume Rendering”,
OR

1 2 • In the control panel, select (Display Tools) > [Viewport Lay-

3 4
out]) and set the view type to [VR] for the desired view.

Volume Rendering on MPVR Oblique Views

You can use volume rendering on MPVR oblique views. To do so you must first
have set up the MPVR oblique view as detailed in Chapter 7 “Reformatting”.
To switch an MPVR oblique view to volume rendering mode:
• On the view, click and hold left or right on the MPVR rendering mode active
annotation (top annotation at lower left, next to the slice thickness annotation)
and select [VR] in the drop-down menu.

11-17
2271012-100.book Page 18 Thursday, May 8, 2003 11:30 AM

Volume Analysis

4 CONTROLS
Overview
When you activate the volume rendering option on 3D or MPVR oblique views (see
previous paragraph), an additional (VR Tools) tool menu button is displayed on the
main Volume Analysis 2 control panel.
The VR Tools menu gives you access to the three VR (Volume Rendering) control
panels:
- VR Presets,
- VR Opacity,
- VR Colors.
You can switch between the three control panels either by means of the (Back) and
(Next) buttons (at bottom right on the panels) or by means of the VR Tools menu.
You can remove a Volume Rendering control panel temporarily from view by means
of its (Close) button, without affecting settings of the volume rendering mode on the
current 3D or MPVR oblique view. To bring the panel back into view, reselect the
corresponding menu item in the VR Tools menu.
To stop using volume rendering on a view and remove the (VR Tools) tool menu
button, simply select a different rendering mode or view type.

VR Presets
Overview
When using volume rendering on a view, you must set up several parameters:
opacity curve shape and values, type of shading, window width and level, etc.
You can do this manually, using the controls on the view and on the VR Opacity
and VR Colors panels (see further), or you can use the pre-defined settings avail-
able on the VR Presets panel.
Volume Rendering is supplied with a set of factory-defined presets.
You can also store settings you have defined yourself as new presets, and retrieve
them at a later time. You can add your own comments to each new preset. See
paragraph “Save Preset”.
You can delete any presets that are either not relevant to your work, or that are
unsatisfactory for any other reason. See paragraph “VR Presets Controls”.

11-18
2271012-100.book Page 19 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Factory-Defined Presets
The factory-defined presets are intended as examples, and contain detailed com-
ments that explain the rationale for each setting.
• Use them as an aid to explore various settings for visualizing different anatomical
structures, when you are familiarizing yourself with the software.
• Use them as as a starting point for the display of a particular structure, then use
the manual controls to optimize the display.

The factory-defined presets are provided as examples.

CAUTION You should be aware that actual results will vary for
individual patients and acquisition techniques.

VR Presets controls
To apply a preset to the current image:
• Select the anatomical category from the menu associated with the (Anatomy)
button, to display the corresponding list or presets (panel of icons).
To display the comments associated with a preset icon, leave the mouse pointer
steady on the icon for a few moments.
If more presets are defined than can be shown in the window, use the small {up}
and {down} buttons below the window to move through the available presets.
• Click on the preset icon (a border round the icon shows that it is selected). The
preset is applied automatically.
To delete a preset:
• Click on the preset icon to select it, then click on (Delete Preset).
Refer to paragraph 5 below for the functions of the (Color) and (Enhanced Reso-
lution) buttons.

11-19
2271012-100.book Page 20 Thursday, May 8, 2003 11:30 AM

Volume Analysis

5 VR OPACITY
VR Opacity control panel
You use the controls on this panel to define and adjust the opacity curve, to set cer-
tain options, and to save presets.

Define opacity curve

Three parameters define the opacity curve:
- Curve type.
- The low and high control points on the voxel value scale.
- Maximum opacity value.

11-20
2271012-100.book Page 21 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Curve Type
To select the opacity curve shape, click on the corresponding (Curve Type) button.
See paragraph above for the definition of “Max. Opacity”.

This is the most usual choice. It may be used to display "bright"

structures, such as bone or vessels in CT data sets.
− Voxels with values below the setting of the left control point
slider will be transparent.
− Voxels with values above the setting of the right control point
slider will be at “Max. Opacity”.
− The opacity of voxels with values between the two sliders will
increase linearly from transparent at the value of the left slider
to “Max. Opacity” at the value of the right slider.

This is a step curve with only one control point. It allows you to
obtain a surface shading rendering on the VR views.
− Voxels with values below the setting of the control point will be
transparent.
− Voxels with values above the setting of the control point will be
at “Max. Opacity”.

This curve type is used to display structures with voxel values

within a defined range. Liver lesions in CT or CTA bolus data sets
are examples of such structures.
Partial volume effects (see paragraph above) can also generate
voxel values within the range you have selected at the interface
between dark and bright regions. The Partial Volume Filter (see
paragraphs above) can be used to attenuate this effect.
− Voxels with values both below the setting of the left control point
slider and above the setting of the right control point slider will
be transparent.
− The opacity of voxels with values between the two sliders will
increase linearly from transparent at the value of the left slider
to “Max. Opacity” at the midpoint between the two sliders, then
decrease to transparent at the value of the right slider.

11-21
2271012-100.book Page 22 Thursday, May 8, 2003 11:30 AM

Volume Analysis

This curve type can be used to display dark structures, such as

airways, or vessels in “black blood” MRA techniques. It can also
be used with a cut plane to display the lumen of a vessel as
“etched” in the plane.
− Voxels with values below the setting of the left control point
slider will be at “Max. Opacity”.
− Voxels with values above the setting of the right control point
slider will be transparent.
− The opacity of voxels with values between the two sliders will
decrease linearly from “Max. Opacity” at the value of the left
slider to transparent at the value of the right slider.

This curve type can be used with a cut plane to display the lumen
of a vessel as “etched” in the plane, similar to the curve type
above.
− Voxels with values both below the setting of the left control point
slider and above the setting of the right control point slider will
be at “Max. Opacity”.
− The opacity of voxels with values between the two sliders will
decrease linearly from “Max. Opacity” at the value of the left
slider to transparent at the midpoint between the two sliders,
then increase to “Max. Opacity” at the value of the right slider.

Note: A range of pixel values may not exclusively define a single tissue or organ.
As a consequence, it may be difficult to visualize particular structures of
interest since they may be hidden by other objects with the same values.

11-22
2271012-100.book Page 23 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Low and High Control Points

• To adjust the low or high con-

trol point, press and drag the
20 240 left or right diamond-shaped
slider to the desired value.
When you move a control
0 100 200 300 point slider, the scale below
the sliders is resized automat-
ically. You can also resize the
scale manually by pressing
and dragging on the scale
markings.

The triangular mark on the scale continuously shows the voxel value at the current
position of the 3D cursor on the image window. This can be used as an aid to posi-
tion the control point sliders.
The selected curve type is displayed between the sliders (the illustration shows the
“low-to-high” curve type).
Moving one of the control point sliders also highlights the voxels on the reformatted
and MIP views that have an opacity value greater than 10% of the maximum opac-
ity value. This will help you to select the range of voxel values corresponding to the
anatomy that you want to display.

11-23
2271012-100.book Page 24 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Max. Opacity
The maximum opacity value (“Max. Opacity”) is set by means of the {Max. Opac-
ity} thumbwheel.

• Press on the {Max. Opacity} thumbwheel and drag it up

Max. Opacity : 80% and down to the desired value, or click on the triangles
above and below the thumbwheel to increase and
decrease the value by steps.
This defines the opacity (in percent) of the point of maxi-
mum opacity of the selected curve. A value of 100% cor-
responds to fully opaque, where a voxel reflects 100% of
all incident light. A value of 1% means that a voxel is
nearly transparent, and will reflect only 1% of the incident
light.

Setting the maximum opacity value to a low value may result in very dark images,
because only a fraction of the incident light is used to create the final image.
On black-and-white images, use the regular window width/level controls to achieve
a correct shading level.
On a color image, the brightness is adjusted automatically whenever the maximum
opacity value is adjusted. Additional manual adjustment using the {Brightness}
thumbwheel on the Color panel may be necessary for optimum results on the struc-
tures of interest.

11-24
2271012-100.book Page 25 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Options
Color or Black-and-white Shading
For details concerning the use of color or black-and-white shading, see paragraph
2 “Introduction”. For details of the available color controls, see paragraph 6 “VR
Colors”.
• To select color shading mode for the Volume Rendering images, press the
(Color) button. Press the button again to return to black-and-white shading
mode.

Enhanced Resolution
In enhanced resolution mode, both horizontal and vertical resolution are increased
by a factor of two. This results in a fourfold increase in image processing time,
therefore it is advised to perform initial set-up and adjustments in standard resolu-
tion mode, then switch to enhanced resolution at the end for final assessment of the
image.
• To select enhanced resolution mode, press the (Enhanced resolution) button.
Press the button again to return to standard resolution.
Note: On oblique views, enhanced resolution is selected automatically, regardless
of the setting of the enhanced resolution mode button.

Enhance Contrast
The Enhance contrat parameter allows you to automatically modify the W/L to
intensify contrast on black and white images. There are two levels of intensification,
depending on what the image requires: mild or strong.
• To modify the contrast on black and white images, click on the arrow of the
(Enhance Contrast) button. A pullright menu opens up. Click on the (Mild) or
the (Strong) button, depending on the level of contrat your image needs. Click
on the (None) button to return to a standard visualization.

This type of setting will alter the linear relationship

CAUTION between the original dataset and the VR volume (in par-
ticular, as far as W/L is concerned).

11-25
2271012-100.book Page 26 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Partial Volume Filter

The “partial volume” effect occurs at the interface between tissues with widely sep-
arated densities, as described in paragraph 2 “Introduction”. The Partial Volume Fil-
ter allows you to attenuate this effect.
• To activate the Partial Volume Filter, press the (Partial Vol. Filter) button. Press
the button again to turn the filter off.
This filter is not available on oblique views. Setting the Partial Volume Filter to “on”
has no effect on oblique views.

11-26
2271012-100.book Page 27 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Save Preset
To save the current Volume Rendering parameters as a new preset:
• Click on the (Save Preset) button to opens the Save Preset window.
• Select the anatomical category under which you want to save the preset from the
menu associated with the (Anatomy) button.
• Move the mouse pointer into the “Name” field and type the name for the new pre-
set. Use only alphanumeric characters or spaces and avoid long names.
• Move the mouse pointer into the text field and type any comments you may want
to add.
To move to a new line, use the Enter key. To delete text in front of the cursor, use
the BackSpace or Del key.
To move the text cursor (e.g. for corrections) move the mouse pointer to the new
position and click.
Do NOT use the characters / or \ or “ in the text.
• To cancel the save operation, press the (Cancel) button. The preset is not saved.
• To save the new preset, press the (Save) button.

11-27
2271012-100.book Page 28 Thursday, May 8, 2003 11:30 AM

Volume Analysis

6 VR COLORS, COLOR SHADING MODE

Shading Modes
As described in detail in paragraph 2 “Introduction”, two different shading tech-
niques can be used for Volume Rendering images:
• Black-and-white shading (voxel value averaging).
• Color shading (surface shading).
When the color shading mode is on (active), the image will be shaded based on the
orientation of the surfaces in the voxels that contribute to a given pixel, and colored
according to the value of these voxels. The colors are set up by means of the VR
Colors panel: see below.
When the color shading mode is off, the image will be shaded using the average of
the voxel values as weighed by their contribution to the pixel values.
To select the color shading mode:
• Click on the (Color) button either in the VR Presets or the VR Opacity panel.
Click on the button again to return to black-and-white shading.

11-28
2271012-100.book Page 29 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Color sliders

• You set the colors to be used for

color shading by means of the dia-
20 130 240
mond-shaped “sliders”. A color is
assigned to each slider. The
HU
0 100 200
HU
300
arrows indicate the currently
selected slider.
• The “tick” mark on the lower scale
continuously shows the voxel
value at the current position of the
3D cursor.

In the example above:

- All voxel values below 20 will be painted with the color assigned to the
first slider.
- Voxel values between 20 and 130 are painted with shades between the
colors assigned to the first and second slider.
- Voxel values between 130 and 240 are painted with shades between
the colors assigned to the second and third slider.
- All voxel values above 240 will be painted with the color assigned to the
third slider.
• To change the setting of a slider, simply click on it and drag it left or right. The
exact setting of the slider is continuously displayed above it.

• To assign a color to a particular slider, first click

on the slider to select it, then in the color palette
circle drag the color marker (cross and arrows)
to the desired color.

When you move a slider, the scale below the sliders is resized automatically. You
can also resize the scale manually by pressing and dragging on the scale mark-
ings.
While you move one of the color sliders, those voxels on the reformatted and MIP
views that are affected by that slider are highlighted by colored hatching. This will

11-29
2271012-100.book Page 30 Thursday, May 8, 2003 11:30 AM

Volume Analysis

help you to select the range of voxel values corresponding to the anatomy that you
want to display.

Brightness
The use of a low opacity setting on the VR Opacity panel (see paragraph 5) may at
times result in a dark image.

Brightness : 100%
• To modify the brightness of the image, press on the
{Brightness} thumbwheel and drag it up and down to the
desired value, or click on the triangles above and below
the thumbwheel to increase and decrease the value by
steps.
On color images, the brightness value is automatically
adjusted when the opacity value is modified. However,
additional manual adjustment using the {Brightness}
thumbwheel may be necessary for optimum results on the
structures of interest. With a high opacity setting, a high
brightness value may cause the brighter color values to
saturate to white.

Color Style
The color style setting determines the type of transition between color shades,
smooth or sharp. Smooth color shade transitions are useful to show subtle voxel
value differences. A sharp transition between color shades makes multiple struc-
tures easier to see (with a smooth transition, color coded borders between struc-
tures may appear blurry).
To select the appropriate color style:
• Click on the (Color Style:) button and select either [Ramp] or [Step] from the
menu.
“Ramp” mode produces smooth color transitions between the values corre-
sponding to the slider settings.
In “Step” mode the color changes abruptly at the half-way point between two slid-
ers.
Note: In Define Object mode (see paragraph 7), this feature is not available and
the option is displayed in grey.

11-30
2271012-100.book Page 31 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Creating Color Sliders

You can at all times create a new color slider:
• Click on the (Add Color) button to add a new color slider.
The new slider will be created at the position of the “tick” mark on the lower
scale, i.e. at the voxel value corresponding to the current position of the 3D cur-
sor.
The color of the new slider is defined by the location of the marker on the Color Pal-
ette.

Deleting Color Sliders

You can delete a color slider if required, but only if there are more than three color
sliders on the panel.
To delete a color slider:
• Select the slider to be deleted, then click on the (Delete Color) button.
OR
• Click on the slider and drag it upwards, out of the color slider panel.

11-31
2271012-100.book Page 32 Thursday, May 8, 2003 11:30 AM

Volume Analysis

7 VR COLORS, DEFINE OBJECTS MODE

Overview
The Define Objects mode allows you to define up to four separate “objects” on the
view. Each separate object is attached to a color slider. You define a separate
opacity curve for each object, using the controls on the VR Colors panel.
The contributions from the images created with each individual curve are blended
to form the final image.
In this mode you can:
• Adjust the range of voxel values that defines each object.
• Adjust the contribution of each object to the final image.
• Detach an object (remove it from the image).
Note: In this mode, the controls and settings on the VR Opacity panel (in particu-
lar the opacity curve setting) have no effect.

Defining an Object
To define an object:
• Select the color slider to be used by clicking on it.
• Attach an object to it by clicking on the (Attach Object) button.
• Set the associated color by means of the color palette.
The opacity curve for the object will be displayed on the color editor. Color display
mode is automatically selected, but you may revert to black and white mode.

11-32
2271012-100.book Page 33 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Adjusting the Range of Visible Voxels

20 240 Each object is defined by a range of

visible voxels. You can adjust this
range by dragging the slider to
HU HU
0 100 200 300 which the object is attached, and
the two neighboring sliders.

For an object attached to a slider that is located between other sliders, the opacity
curve is defined so that it falls off halfway between sliders, with a degree of overlap.
For an object attached to the extreme left or right slider, the opacity curve extends
to the beginning or end of the range of voxel values, respectively (see figure).

Adjusting the Opacity of an Object

To adjust the contribution of an object to the final composite image:

Opacity : 100% • Select the corresponding slider by clicking on it.

• Set the contribution of the object by means of the
{Opacity} thumbwheel. Press on the thumbwheel and
drag it up and down to the desired value, or click on
the triangles above and below the thumbwheel to
increase and decrease the value by steps.

An opacity value of 100 % will show the object fully, while a lower value will make it
progressively less visible.
Note: The {Brightness} thumbwheel controls the brightness of the entire image,
not of the individual attached objects.

11-33
2271012-100.book Page 34 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Detaching an Object
To detach an object (remove it from view):
• Select the slider to which the object is attached by clicking on it.
The legend on the (Attach/Detach Object) button will indicate (Detach Object).
• Detach the object by clicking on the (Detach Object) button.
The opacity curve associated with the object disappears from the VR Colors panel,
and the object disappears from the view.
Note: As long as one or more objects are attached to the color sliders, the controls
on the VR Opacity control panel (curve type, opacity curve editor, opacity
level) have no effect. If you change the setting of any of these controls, a
message is displayed, warning you that the changes will have no effect until
you detach all objects.

11-34
2271012-100.book Page 1 Thursday, May 8, 2003 11:30 AM

CHAPTER 12 Volume Analysis

CHAPTER 12 - ANNOTATIONS
We can distinguish three main types of annotations on Volume Analysis 2
views:
- -System annotations,
- User text annotations,
- Measurements.

It is the responsibility of the physician to keep written track

CAUTION of transformations made on the 3D model, using text
annotations, for example.

12-1
2271012-100.book Page 2 Thursday, May 8, 2003 11:30 AM

Volume Analysis

1 TYPES OF ANNOTATIONS
System annotations (including active annotations)
System annotations consist of:
- Information such as hospital and patient name, date, name of the
protocol, acquisition parameters, etc. These are part of the exam and
cannot be modified (the patient name can be shown or hidden but not
altered),
- Display data such as DFOV (display field of view), window width and
level, image number and orientation, etc. These are added to the views
and controlled by the application.
Some are displayed in red, indicating that they are active, which means the
user can modify them directly on the view. See paragraph “Active Annota-
tions” in Chapter 6 “Display”.

Text annotations
You can add notes and comments directly on the views, and use a marker with the
annotation to point out anatomical features.
User text annotations can be edited, moved or deleted as necessary. When
images are filmed or saved, any annotations shown on the images are filmed and
saved with them.

Measurement annotations
You can perform various measurements (voxel value, distance, angle, area, vol-
ume) on the views. The corresponding graphics and measurement values are dis-
played on the views. They can be moved and deleted, and filmed or saved, in the
same manner as the text annotations.
You can also create measurement annotations combined with explanatory text.
Measurements are described in detail in Chapter 13 “Measurements”.

12-2
2271012-100.book Page 3 Thursday, May 8, 2003 11:30 AM

Volume Analysis

2 TEXT ANNOTATIONS

To create a text annotation on a view, select (Display Tools) >

A [Annotation].

To create a basic text annotation, select [Annotation without

A arrow].

To create an annotation to point out a particular anatomical

A feature, select [Annotation with Arrow]. You will then be
prompted to mark a point on the view before entering the text
annotation. The mark and the annotation are linked by a dot-
ted line.

Note: The other buttons in the Annotation panel are used to create measurement
annotations combined with explanatory text. See Chapter 13 “Measure-
ments”.

12-3
2271012-100.book Page 4 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Text Entry
Because entering and editing text directly on a view is not very convenient, you use
a separate text entry window when creating a new text annotation or editing an
existing one. This window appears automatically when you select to create a new
annotation via the control panel menus or select an annotation for editing by point-
ing on it with the mouse.
The text entry window acts much as a standard text editor. You can type in text,
move to the next line with the <Enter> key, make corrections with the <BackSpace>
key, and move the text entry point (text cursor) by pointing and clicking with the
mouse.
When you have completed the text:
• Select (OK) to apply the text to the annotation.
Select (Save) to apply the text to the annotation and save it as a new preset
annotation (see paragraph 4).
Select (Cancel) to remove the annotation from the view.

12-4
2271012-100.book Page 5 Thursday, May 8, 2003 11:30 AM

Volume Analysis

3 MANAGING ANNOTATIONS
Note: The functions described in the following paragraphs apply to both text and
measurement annotations.

General
You can move annotations on the view, by clicking and dragging on them with the
mouse.
To change the font type of an annotation, click on it with the left or right mouse but-
ton and select the font type in the on-view menu attached to the annotation (you
have a choice of normal, italic, bold, small and large).
To delete an annotation, click on it and drag it outside the view, or click on it with the
left or right mouse button and select [Delete] in the on-view menu.

Preset Annotations

If you regularly process the same type of exams, you can save
frequently used annotations as ”presets”, then enter them again
A
by means of the (Display Tools) > (Recall Preset) button.
Once placed on the view, preset annotations can be moved,
edited or deleted like any other annotation.

Preset annotations can be text, measurement or combined annotations. See para-

graph 4.

You can also call the (Presets) menu by clicking on the

NOTICE

icon of the Review Controller.

12-5
2271012-100.book Page 6 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Show/Hide Annotations
At times, you may want to hide momentarily all annotations on the views (e.g., to
better examine a detail on a view, or to film a view without annotations for teaching
purposes).
To do so, select (Display Tools) > [Preferences] and switch off (Show Annota-
tions) button.
To return the annotations to the views, again select (Display Tools) > [Prefer-
ences] and switch on (Show Annotations) button.
Note: Switching off (Show Annotations) hides ALL annotations, i.e., both the sys-
tem annotations and the user annotations (text and measurements).

12-6
2271012-100.book Page 7 Thursday, May 8, 2003 11:30 AM

Volume Analysis

4 SAVING ANNOTATIONS AS PRESETS

Note: The functions described in this paragraph apply to both text and measure-
ment annotations.
If you regularly process the same type of exams, you can save frequently used
annotations as “presets”, then use them again later. Preset annotations can be
text, measurement or combined annotations.
To save an annotation as a preset:
• Create a new annotation, or select an existing one to modify.
• Click on (Save) in the New Annotation or Modify Annotation panel.
• In the Save Preset Annotation panel, enter the name for the new preset.
If you also want to be able to use the preset for anatomical categories or proto-
cols other than the current ones, switch on (Use for other anatomy) and/or (Use
for other protocol) as required.
• Select (Save) to save the preset, or (Cancel) to annul the operation

To place an existing preset annotation on a view, select (Display

Tools) > (Recall Preset) and select the required preset from the list
A
in the Preset Annotations panel.
Once placed on the view, preset annotations can be moved, edited
or deleted like any other annotation.

You can also call the (Annotation) menu by clicking on the A

NOTICE

icon of the Review Controller.

12-7
2271012-100.book Page 8 Thursday, May 8, 2003 11:30 AM

Volume Analysis

5 RECORDING THE ON-VIEW ANNOTATIONS

Note: The functions described in the following paragraphs apply to both text and
measurement annotations.

Overview
When a view is filmed, any annotations on the view will also appear on the film.
When a view is saved on the image disk using ScreenSave or the ScrapBook appli-
cation, any annotations on the view will be saved with it and will be shown when the
view is accessed at a later date. At this stage, the annotations become part of the
image: they can no longer be edited or deleted.
When saving reformatted images, the image type (ScreenSave or Reformat) deter-
mines how the annotations are saved. See paragraph bellow for details.

Show/Hide Annotations
If the annotations on the screen have been turned off via the (Display Tools) >
[Preferences] > (Show Annotations) control they will not be present on the filmed
or saved images either.

Although images without annotation may be suitable for

CAUTION teaching purposes, diagnosis should not be performed
with such images.

Note: This function can also be executed using the corresponding function of the
main on-view menu (refer to Chapter 6, Section 5 “Main On-view Menu”.)

Patient Name
While working on an exam, you can hide the patient name on the views (see Chap-
ter 6) for increased confidentiality. If you have done so, make sure to show the
patient name again on the views BEFORE filming or saving images for diagnostic
purposes.

When filming or saving images for diagnostic purposes,

CAUTION always make sure the patient name is displayed on all
views..

12-8
2271012-100.book Page 9 Thursday, May 8, 2003 11:30 AM

Volume Analysis

3D Cursor
At times, you may want to show the current position of the 3D cursor on filmed or
saved images (e.g., to point out the same feature on several different views).
To do so, select (Filming Tools) > (Film/Save Options) and turn off (Hide Cursor
on Copies).

Reformatted Images
When saving reformatted images, you can select the image type: ScreenSave or
Reformat.
Select (Filming Tools) > (Film/Save Options) and set (Image Type for Reformat)
to either [Save] or [Rfmt]. Also see Chapter 14 “Recording”.
If you use ScreenSave, all remarks in the earlier paragraphs are applicable. Only
the annotations present on the view at the moment that you save the image will be
saved with it, and become part of the image.
If you use Reformat, all system annotations are saved automatically with the view,
even if they are hidden at the moment that you save the image. When the view is
accessed at a later date using the Image Works Basic Display Viewer, you can
select which system annotations are to be displayed via the Viewer Preferences
controls. However, the user annotations (text and measurements) are saved only if
they are shown at the moment that you save the image.

Color Save
If you save ScreenSave images, you can choose between black-and-white images
or color images by activating the (Color Save) button of (Filming Tools) > (Film/
Save Options) panel.
Note: If you choose to save Color images, you will no longer be able to modify the
W/L on the saved images.
If you save Reformat images, using an [Rfmt] image type, the (Color Save) button
will still be active but its effect will be overridden by the image type.

12-9
2271012-100.book Page 10 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Blank page.

12-10
2271012-100.book Page 1 Thursday, May 8, 2003 11:30 AM

CHAPTER 13 Volume Analysis

CHAPTER 13 - MEASUREMENTS
You can measure voxel value, distance, angle, area and total volume on the
views, in a manner similar to the Image Works Basic Display application. Mea-
surement annotations are displayed directly on the views, and can be filmed or
saved with the views like any other annotation.
• “Report cursor” shows the RAS coordinates and voxel value at the 3D cursor
position.
• Distance, angle and area measurements are performed by tracing a line seg-
ment, angle or area trace on the view. “Volume” shows the total volume of the
3D model.
You can display a ruler (grid or tick marks) to better appreciate dimensions on
the views.
You can define and display profile graphs and histograms. These are displayed
on separate views, that also can be filmed or saved as required.
• A profile graph shows the voxel value along a 3D trace (profile).
• Histograms graphically show the voxel value distribution and associated sta-
tistics within an area (cross-section) or within a volume. They have a dual
purpose:
- They allow you to analyze the voxel value distribution of a 3D object
in various ways, as an aid in setting up 3D processing (e.g.,
thresholding),
- They can be used for cross-section and volume measurements of
specific anatomic features, on condition that the feature to be
measured can be clearly defined by a range (class) of voxel values.
Measurement accuracy depends on various factors. To assess the accuracy of
a measurement you want to perform, refer to paragraph 5 “Measurement Accu-
racy” in this chapter.

13-1
2271012-100.book Page 2 Thursday, May 8, 2003 11:30 AM

Volume Analysis

1 ON-VIEW MEASUREMENTS
Overview
You can measure voxel value, distance, angle, area and total volume on the views.
The procedures are closely similar to those in the Image Works Basic Display appli-
cation. However, there is an essential difference in that we are now dealing with
three-dimensional measurements.
For instance, to measure a distance we still need to place two points to define a line
segment (see paragraph “Distance” bellow for details). But these two points do not
have to be placed on the same view; they can be placed at entirely different loca-
tions in the 3D volume.
At all times, the views will only show the projection of the measurement (distance,
angle, area) onto the plane of the views. The displayed measurement value, how-
ever, can be either the true three-dimensional measurement (3D mode) or the mea-
surement of the projection (2D mode).
Note: In practice, rather than move through the image set to place measurement
points, you may find it easier to set up an oblique view (see Chapter 8) that
contains all the points of the feature you want to measure and perform the
measurement on this view. However, the use of such an oblique view is in
no way obligatory.
While in theory you can place measurement points on 3D views, you should NOT
use this technique, because on such views you do not have ANY indication of how
deep the point is located inside the 3D volume, without continuously correlating the
position of the 3D cursor on the baseline views.

DO NOT USE 3D VIEWS ONLY TO PERFORM VOXEL

WARNING! VALUE, DISTANCE, ANGLE OR AREA MEASURE-
MENTS. ALWAYS REFER TO 2D BASELINE VIEWS.

13-2
2271012-100.book Page 3 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Procedures
You can create measurement annotations either without or with attached text com-
ments, or use the existing preset measurement annotations.

Measurements without text

To perform measurements on the views without text comments:
• Select (Display Tools),
• Select the required type of measurement in the menu: voxel value, distance,
angle, area or total volume,
• Follow the instructions in the measurement control panel that appears.
Where necessary set the measurement mode: 2D or 3D, and the type: straight
line or curved.

Measurements with text

To perform measurements on the views and add a text comment:

• Select (Display Tools) > (Annotation).

A

• Select the required type of measurement: straight line distance, curve distanced,
angle, area or total volume.
• Follow the instructions in the measurement control panel that appears.
Where necessary set the measurement mode: 2D or 3D.
• Enter the text comments in the New Annotation panel.
• Select (OK) to add the text to the annotation.
Select (Save) to add the text to the annotation and save the combined measure-
ment/text annotation as a new preset annotation (see Chapter 12, paragraph 4).
Select (Cancel) to remove the measurement annotation from the view.

13-3
2271012-100.book Page 4 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Using presets
To perform measurements on the views using the existing preset measurement
annotations:

A • Select (Display Tools) > (Presets).

• Select the required type of measurement from the list in the Preset Annotations
panel.
Note: The preset measurement annotations available in the list may vary depend-
ing on the anatomical category of the current protocol.

13-4
2271012-100.book Page 5 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Measurements
Overview
You can perform measurements of voxel value, distance, curve, angle, area or total
volume using one of the procedures described below. Click on the Display Tools
button of the Control Pannel. The following menu is displayed:

Click on the button of the menu. The following window opens up on

the left-hand bottom view of the viewing area. You can then choose the type of
measurement you want to make (voxel value, distance, curve, angle,area or vol-
ume), by clicking on the corresponding button. Other windows then open up allow-
ing you to define the parameters you want for the chosen measure. These buttons
also allow you to enter customized annotations.

13-5
2271012-100.book Page 6 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Note: The other buttons on the drop down menu are shortcuts allowing you to take
measures with predefined parameters and annotations. The button on the
Review Controller is a preset to measure a distance.

Report cursor

To display the RAS coordinates and the voxel value of a

given location in the 3D volume, select (Report Cursor) and
place a point at the location you want to measure. The RAS
coordinates and the voxel value are displayed.

Distance
You can perform distance measurements either along a straight line (i.e., between
two points), or along a curve. The displayed value can be either the true distance in
the 3D volume (3D), or the length of the projection on the view where you perform
the measurement (2D).

Straight line distance measurement

To measure the distance between two points:

For a measurement without text, select (Distance) and

use the (Along) button to select [straight line].
For a measurement with text, select the (Straight line
distance) button.

Mark the two points. A line segment connecting the two

points, and the value in mm of the distance between
A
them, are displayed.

13-6
2271012-100.book Page 7 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Distance measurement along curve

To measure the distance along a curv
e (composed of straight-line segments):

For a measurement without text, select (Distance) and use

the (Along) button to select [curve].
For a measurement with text, select the (Curved distance)
button.

click left to mark the points. Click right to end marking

points. The curve and the value in mm of the distance
along the curve are displayed.
A

Angle

The displayed value of an angle measurement can be

either the true angle in the 3D volume (3D), or the angle of
the projection on the view where you perform the measure-
ment (2D).

To measure an angle on a view, select (Angle) and place

three points to define two successive line segments. The
line segments and the value in degrees of the angle
A
between them are displayed.

Area

Since an area measurement implies a 2D surface, such

measurements should be performed on 2D views (baseline
or oblique views). An area measurements performed on a
3D view is meaningless.

To measure an area on a view, select (Area) and place a

succession of points to outline the area you want to mea-
sure. The outline and the value in mm2 of the area are dis-
A
played.

13-7
2271012-100.book Page 8 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Volume

To measure the total volume of the current 3D model, select

(Volume) and click on a view. The value in mm3 of the vol-
ume of the 3D model is displayed.

Editing measurements
Straight-line distance, curve distance, angle and area measurements ar all shown
on the views as traces composed of one or more segments.
You can modify any of these traces:
• Click and hold left on an endpoint of a segment, and drag it to a new position.
You can do this at any time, even after closing the control panel that you used to
create the trace.
The corresponding measurement value on the view is modified in real time.
If you have added text to a measurement, you can edit it in the same way as a stan-
dard text annotation:
• Place the mouse pointer on the annotation.
• Start to type. This opens the Modify Annotation panel. Edit the annotation text.
• Select (OK) to modify only the text of the annotation.
Select (Save) to modify the text of the annotation and save the combined mea-
surement/text annotation as a new preset annotation (see Chapter 12, paragraph
4).
Select (Cancel) to annul the operation.

13-8
2271012-100.book Page 9 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Measurement annotations
Measurement annotations with or without added text are displayed directly on the
currently active view, with a dotted line linking it to the location that the measure-
ment corresponds to.
You manage measurement annotations in exactly the same way as text annota-
tions.
Refer to Chapter 12 “Annotations”, paragraphs 3 to 5 for full details on how to
- Move annotations,
- Change the font type,
- Delete an annotation,
- Save and use presets,
- Film and save the annotations with the view.

13-9
2271012-100.book Page 10 Thursday, May 8, 2003 11:30 AM

Volume Analysis

2 RULER
To better appreciate dimensions on a views, you can display a ruler, ether in the
shape of a grid overlaying the entire view, or in the shape of tick marks along the
sides of the view.
• Select the view as primary (red border).
• Select (Display Tools) > (Preferences) > (Ruler style) and set the ruler style as
required: [none], [grid] or [tick].

Spacing (mm): • Enter the grid or tick mark spacing in mm from the
keyboard.
10.0

Range (mm): • When you use tick marks as the ruler style, you can
also choose the distance (range) over which tick
220.0
marks appear.
Enter the desired range in mm from the keyboard.

Repeat as required for each view where you want to display a ruler.

13-10
2271012-100.book Page 11 Thursday, May 8, 2003 11:30 AM

Volume Analysis

3 PROFILE
A profile graph (see illustration on next page) shows the distribution of voxel values
along a 3D trace, in the form of a graph of voxel value against position along the
trace.
A profile graph is displayed in a separate view. To select the view you want to use
for the profile graph: open either the [view type] active annotation menu or the
(Display Tools) > [View Layout] panel (see Chapter 6) and select [Profile].
To display a profile graph, you define a 3D trace on the views (hold the <Shift> key
down and click to deposit points; see Chapter 6 for full details). The profile view will
indicate “Undefined profile” until you start defining the 3D trace.
The horizontal axis of the profile view is the position in mm along the trace, and the
vertical axis is the voxel value as a function of that position.
Statistical information about the voxel values along the trace is displayed at the bot-
tom of the profile view:
- Mean = average voxel value along the trace,
- Std. = standard deviation of voxel values along the trace.
Both the graph and the statistics are updated each time you extend or modify the
3D trace.
Click on the profile view to display the “voxel reference line” (thin vertical line on the
profile graph).
You can click and drag on the voxel reference line: as soon as you do so, the 3D
cursor is forced to the corresponding position on the trace, and the voxel value at
the position of the 3D cursor and voxel reference line is shown on the profile view.

13-11
2271012-100.book Page 12 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Voxel reference line

13-12
2271012-100.book Page 13 Thursday, May 8, 2003 11:30 AM

Volume Analysis

4 HISTOGRAMS
Overview
A histogram graph shows the percentage of occurrence of each voxel value, either
in a user-defined surface area on a reformatted slice (cross-section histogram) or in
a 3D object (volume histogram).
See paragraph “Cross-Section Histogram” on how to set up a cross-section histo-
gram, or paragraph “Volume Histogram” on how to set up a volume histogram.
A histogram is displayed in a separate view, that contains:
• The graph of percentage of occurrence against voxel value,
• A smoothing annotation that indicates how the percentage-of-occurrence values
are calculated
• Statistics about the voxel values (mean, standard deviation, maximum and mini-
mum), and the computed value of the total surface area or volume,
• A “voxel reference line” (thin vertical line, see paragraph bellow),
• Two “class boundary” lines (dotted lines, see paragraph bellow).
The class boundary lines mark the upper and lower limit of a "class" (range) of
voxel values around the voxel reference line whose values are specific to a cer-
tain anatomical feature (e.g. bone surrounded by muscle, or a tumor surrounded
by healthy tissue).
The class boundaries are initially calculated automatically by the application, but
you can move them manually to delimit an exact range of voxel values.
• Statistics about the voxel values (mean, standard deviation, maximum and mini-
mum) within the class defined by the boundary lines, and the computed value of
the corresponding surface area or volume.
Note: The histogram values and statistics are those of the current 3D model, not
those of the original exam. If the 3D model contains only a given range of
voxel values (either from the initial “build protocol”, see Chapter 5, or after
“thresholding”, see Chapter 9), only voxels within that range will appear in
the histograms.
Note: Statistics and computed values of surface area or volume displayed on the
histograms are subject to the same accuracy limitations as other on-view
measurements. See paragraph 5 “Measurement Accuracy”.

13-13
2271012-100.book Page 14 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Smoothing

The smoothing annotation (upper left on a histogram

view) indicates how the percentage-of-occurrence values
are calculated.
If the smoothing value is 0, each percentage-of-occur-
rence value on the histogram’s vertical axis is calculated
directly for each voxel value on the histogram°Øs hori-
zontal axis.

If the smoothing value is, for example, 10, a range of +/- 10 is taken around each
voxel point on the histogram’s horizontal axis. The percentage-of-occurrence val-
ues are averaged and this average value is attributed to the voxel point on the his-
togram’s horizontal axis. This creates a “smoothing” effect on the histogram curve,
and the greater the smoothing value, the greater the smoothing effect.To change
the smoothing value, you can directly modify the smoothing annotation. See Chap-
ter 6, paragraph “Active Annotations”.

Cross-Section Histogram
A cross section histogram view (see illustration on next page) shows:
- The percentage of occurrence of each voxel value in a user-defined
surface area on a reformatted slice,
- Numerical statistics about the voxel values in this same surface area
plus the calculation of this area,
- Boundaries around a class of similar voxel values in this area.
• Set up the reformatted slice that you want to use,
• Create a closed trace around the area that you want to use for the cross section
histogram (hold the <Shift> key down and click to deposit points; see Chapter 6
for full details). .
• Select the view where you want to display the cross section histogram:
open either the [view type] active annotation menu or the (Display Tools) >
View Layout panel (see Chapter 6) and select [X Sect.].
Note: If you open a cross section histogram view before creating the trace, the
view will indicate “Undefined histogram” until you start defining the trace.
Note: The view used to generate a cross section histogram must be a reformatted
slice view and not a 3D view.
Note: If you trace the surface area used for the cross section histogram on an
MPVR view (“thick” reformatted slice, see Chapter 8), the resulting histo-
gram will be that of the central (“thin”) slice.

13-14
 2271012-100.book
Page 15 Thursday, May 8, 2003

Smoothing
annotation
11:30 AM

Voxel class
statistics

Voxel reference line

Class boundaries

13-15
Volume Analysis
2271012-100.book Page 16 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Volume Histogram
A volume histogram view (see illustration on next page) shows:
- The percentage of occurrence of each voxel value in the current 3D
model,
- Numerical statistics about the voxel values in the 3D model plus the
total volume of the 3D model,
- Boundaries around a class of similar voxel values in the volume.
• Select the view where you want to display the volume histogram:
open either the [view type] active annotation menu or the (Display Tools) >
View Layout panel (see Chapter 6) and select [Histo.].
Note: Since a volume histogram refers automatically to the entire current 3D
model, no other setup is necessary.

13-16
 2271012-100.book
Page 17 Thursday, May 8, 2003

Smoothing
annotation
11:30 AM

Voxel class
statistics

Voxel reference line

Class boundaries

13-17
Volume Analysis
2271012-100.book Page 18 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Voxel reference line

The voxel reference line appears as a thin vertical line on the histograms.
• You can move the voxel reference line manually on the histogram view to select
a “class” of voxels in the histogram, as described in the next paragraph.
Either hold down the <Shift> key and move the mouse cursor on the view, or
click and drag on the voxel reference line.
• When you move the 3D cursor on one of the other views, the voxel reference line
follows the 3D cursor (i.e., moves to the position on the histogram corresponding
to the voxel value at the 3D cursor).
Note: If the voxel reference line is not displayed, move the mouse pointer onto the
histogram view and press the <Shift> key.
Also note that the voxel reference line will disappear if you move the 3D cur-
sor to a position where the voxel value is outside the range of the histogram
(e.g., outside the 3D object).

13-18
2271012-100.book Page 19 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Voxel Class Boundaries

Overview
A “voxel class” refers to a range (“class”) of voxels whose values are specific to a
certain anatomical feature (e.g. bone surrounded by muscle, or a tumor surrounded
by healthy tissue).
When the voxel reference line is displayed on the histogram, the boundaries of the
corresponding voxel class are computed automatically and displayed as thin dotted
lines.
You can manually “fine tune” the class boundaries if required.
Numerical statistics about the voxel values inside the class, plus the area or volume
constituting the class, are shown on the histogram.

Without going into the mathematics, the software analyzes

the “peaks” and “valleys” of the histogram curve and deter-
mines the voxel class around the current voxel reference
line from the changes in slope (the inflection points) of the
histogram curve.

The software analyzes the histogram “as displayed”, that is, it takes the smoothing
value (see paragraph above) into account when determining the class boundaries.
Increasing the smoothing value will tend to even out the histogram and small
“peaks” and “valleys” will tend to disappear, thereby changing the class boundaries.
Set the smoothing value as required before using the class boundaries.

Automatic
• Position the voxel reference line on the histogram at the desired voxel value
(hold down the <Shift> key and move the mouse cursor, or click and drag on the
voxel reference line).
The software automatically computes and updates the class boundary values, and
all voxels within the class are displayed in green on all views of the object for as
long as you hold down the <Shift> key or mouse button.

13-19
2271012-100.book Page 20 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Manual
• To modify the class boundaries manually, click and drag on them.
The voxels corresponding to the class are displayed in green on all views, and
remain so even after releasing the mouse button (to change the views back to nor-
mal, return to automatic mode by clicking on the voxel reference line or pressing
the <Shift> key).
Note: If the voxel reference line and the boundary lines are very close together,
clicking and dragging with the mouse button tends to move the boundary
lines rather than the voxel reference line. In that case, hold down the
<Shift> key to move the voxel reference line.

Remarks
Whether automatic voxel class boundary determination is effective will obviously
depend on the image data. If the image contains a range of voxels whose intensi-
ties are clearly specific to a certain anatomical feature (e.g. bone surrounded by
muscle, or a tumor surrounded by healthy tissue), the technique can be very useful.
The technique is likely to be more useful on CT exams, where the voxel values
(normally Hounsfield values) provide a better indication of tissue density and type
than on MR exams.
At all times, the manual “fine tuning” (moving the class boundaries on the histogram
with the mouse) allows you to define the voxel class in accordance with your own
perception of the views and your own requirements.

13-20
2271012-100.book Page 21 Thursday, May 8, 2003 11:30 AM

Volume Analysis

5 MEASUREMENT ACCURACY
Overview
This paragraph provides information about the accuracy of on-view measurements.
This accuracy depends on various factors, and in particular on the size of the
region of interest (ROI) being measured.
Most remarks in this paragraph also apply to measurements displayed in histogram
views.
A typical CT or MR acquisition has a DFOV (Display Field of View) from about
25cm to about 50cm. We will use 25cm for the examples below.
Inter-slice distances can vary from less than 1mm up to 10mm. For best results
with 3D imaging the optimum inter-slice distance is the one that results in isotropic
voxels (with the same dimensions in all three axes). However, considerations such
as type of exam and patient irradiation dose levels may lead to the choice of a
larger inter-slice distance. The choice of the optimum inter-slice distance for a
given case is obviously outside the scope of this manual.
Note: Regardless of the zoom factor being used to view images, region-of-interest
measurements and statistics are calculated based on the pixels from the
original, UNZOOMED image data as they arrived on the workstation.

Measurements are more reliable when done on 2D

CAUTION views; we recommend you to at least always check
on the 2D reformatted views where exactly the
points have been deposited.

Measurement resolution
The software calculates and displays measurements with a resolution of one deci-
mal place (such as 0.1 mm, 0.1 degree or 0.1 mm2). You should be aware that the
real measurement accuracy is generally considerably less for a number of different
reasons.

13-21
2271012-100.book Page 22 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Geometrical accuracy
Geometrical accuracy is limited by display resolution (pixel size). When four views
are displayed, each view equals 512x512 pixels, hence with a display field of view
(DFOV) of 25cm a pixel is equivalent to 0.5x0.5mm, so you cannot place a mea-
surement point with a precision better than this. As a result:
- For a distance measurement, the geometrical accuracy is equal to the
displayed length +/- image pixel size.
- For an angle measurement, the geometrical accuracy is equal to the
displayed angle value +/- 10 degrees for an angle measured between
segments which are five times larger than the image pixel size.
Accuracy improves as the length of the segments increases.
- For an area measurement, the geometrical accuracy is equal to the
displayed area value +/- the circumference of the region of interest
multiplied by (image pixel size)2 / 2.
Note : region-of-interest measurements and statistics are based on the
pixels INSIDE the graphic defining the region.
The geometrical accuracy defines a lower bound on the overall accuracy that can
be obtained. Further limiting factors are image set resolution, acquisition accuracy,
display settings and partial volume effects.

Image set resolution

The image set resolution is determined by the size of the field-of-view, the matrix
size and the inter-slice distance.
In the acquisition plane, for a field-of-view of about 25cm, the smallest detail in an
image acquired with a 512x512 matrix will be about 0.5x0.5mm.
With a 256x256 matrix the smallest detail will be 1x1mm.
In the acquisition plane, the measurement accuracy can obviously not be better
than the size of the smallest element. In the same way, the accuracy in a direction
perpendicular to the acquisition plane cannot be better than the inter-slice distance.

Distance, angle and area measurements are valid

CAUTION only if all trace segments are longer than the inter-
slice distance.

13-22
2271012-100.book Page 23 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Acquisition accuracy
Any errors in the original image set resulting from the acquisition process (calibra-
tion, slice interpolation) will be added to the same extent to the measurement error.
As an example, the spatial accuracy of MR images can vary, depending on the
patient, the pulse sequence and the MR system itself. Metallic implants or air-bone
interfaces may lead to susceptibility artifacts and spatial distortions greater than
those observed when calibrating the system with a Quality Assurance phantom,
even on a perfectly tuned MR system.

Display settings
Since anatomical features are rarely of a uniform density, the apparent dimension
of an anatomical feature can change when you change the display settings (win-
dow width and level), thereby adding another factor of uncertainty to an on-view
measurement.

Partial volume averaging

In CT exams, the value of a voxel is the weighted average

for all materials in the voxel. Because of the high attenua-
tion coefficient of calcium, even a small amount of calcium
A
(bone) in a voxel will weigh its value towards that of bone,
so that the entire voxel appears to be bone.
B
This affects the imaging, because the “bone” partial volume
voxels tend to mask adjacent features, but also the mea-
surements: the apparent bone thickness at “A” is more than
double the thickness at “B” while in reality the bone thick-
ness at both locations is the same. Volume measurements
of either the bone or the adjacent tissue will be less accu-
rate for the same reasons.

Note: The effect as illustrated is more pronounced with oblong voxels (inter-slice
distance larger than the voxel cross-section). However, even with isotropic
voxels (with the same dimensions in all three axes) the partial volume effect
will tend to make bone appear thicker than it is in reality, and hence affect
the measurement accuracy.

13-23
2271012-100.book Page 24 Thursday, May 8, 2003 11:30 AM

Volume Analysis

13-24
2271012-100.book Page 1 Thursday, May 8, 2003 11:30 AM

CHAPTER 14 Volume Analysis

CHAPTER 14 - RECORDING
You can record the results of your work in various ways.
You can print them on film (for diagnostic purposes) or on paper (for administra-
tive purposes only), or you can save them on the image disk of the workstation.
To record single images or individually selected sets of images, you use the
same ScrapBook application and Screen Save function that you also use with
AW Basic Display.
When working with three-dimensional image sets, you will often have a require-
ment for recording a set of regularly spaced images (such as a set of reformat-
ted images, or a set of images showing the rotation of a 3D object in successive
steps).
The Batch function of Volume Analysis 2 allows you to rapidly set up such a set
of images. You can then preview the set as an animated sequence (movie loop),
film it, and/or save it on the image disk.
You can also save the current 3D model “as is” on the image disk, to be reloaded
and further processed at a later time. A saved 3D model cannot be displayed or
processed by other Image Works applications.

14-1
2271012-100.book Page 2 Thursday, May 8, 2003 11:30 AM

Volume Analysis

1 FILMING, MOVIE LOOP, SCRAPBOOK AND SCREEN-

SAVE
Overview
To record single images or individually selected sets of images, you use the same
ScrapBook application and ScreenSave function that you use with Image Works
Basic Display.
Controls in the Volume Analysis 2 application allow you to open the ScrapBook
control panel and set up the application, then film or save selected images as
required.
To save individual images, you use the ScreenSave function via the on-view menu,
or the <F1>, <F2> and <F3> keyboard keys, or the <S> keyboard key as in Image
Works Basic Display.
Note: An oblique (reformatted) image can be saved either as an image of type
REFORM or as a secondary capture of type SCPT (depending on the set-
ting of the (Image Type for Reformat) option in the Film/Save menu).

14-2
2271012-100.book Page 3 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Annotations
Annotations on the views (system annotations, text annotations, measurements)
will appear on the filmed or saved images exactly as displayed on the screen.
If the annotations on the screen have been turned off (via the Display Tools >
Show/Hide Annotations control) they will not be present on the filmed or saved
images either.

Although images without annotation may be suitable for

CAUTION teaching purposes, diagnosis should not be performed with
such images.

While working on an exam, you can hide the patient name on the views (see Chap-
ter 6) for increased confidentiality. If you have done so, make sure to show the
patient name again on the views BEFORE filming or saving images for diagnostic
purposes.

When filming or saving images for diagnostic purposes,

CAUTION always make sure the patient name is displayed on all
views.

At times, you may want to show the current position of the 3D cursor on filmed or
saved images (e.g., to point out the same feature on several different views).
To do so, select (Filming Tools) > (Options) and turn (Save with 3D cursor) on or
off as required.

Image Size on Film

To simplify measurements on filmed images, it is obviously preferable that a filmed
image is either real size (scale 1:1) or scaled to a simple multiple or sub-multiple of
real size. This means that the DFOV should preferably be set to a multiple or sub-
multiple of the size of the film slot used by the ScrapBook.
See Chapter 6 paragraph "DFOV and filming" for details.

14-3
2271012-100.book Page 4 Thursday, May 8, 2003 11:30 AM

Volume Analysis

2 BATCH
The Batch function of Volume Analysis 2 allows you to rapidly set up a set of reg-
ularly spaced images, then preview this set as an animated sequence (movie loop)
and film it and/or save it on the image disk.

Batch Types
A batch can be one of the following types:
• Parallel oblique batch: a series of parallel oblique slices along a common cen-
ter line,
• Radial oblique batch: a series of oblique slices generated radially around a
common axis,
• 3D batch: a series of 3D images obtained by rotating a 3D object around an axis.

14-4
2271012-100.book Page 5 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Batch Protocols
Overview
To set up a batch you use a batch protocol. A default batch protocol is provided for
each of the three types of batch. You can use these as a starting point for your own
custom batch protocols which you can save and re-use.
A batch protocol defines:
• Batch mode: oblique or rotation
• Number of views and spacing between views (mm. or degrees)
• Display field of view
• Slice thickness (and render mode for “thick” slices)
• Output mode: this can be none (preview only), film, archive or film+archive.
If you select film mode you can also define the film layout (e.g., 2x3).
When creating and saving your own custom batch protocol, you have two options:
• Add it to the list of existing batch protocols for the current loading/processing pro-
tocol,
• Combine it with the current loading/processing protocol to create a new loading/
processing protocol.
The exams that you work with may systematically require the use of more than one
batch protocol (e.g., a rotation of the 3D view, and a corresponding set of radial
oblique slices).
Volume Analysis 2 allows you to create a custom multiple-batch protocol by com-
bining two or more batch protocol setups.

14-5
2271012-100.book Page 6 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Procedures
First select the view that you want to use to set up the batch and make it the pri-
mary view (red border), then select the batch protocol from the Batch Protocols list.
For a parallel oblique batch you can use almost any kind of view: axial, sagittal,
coronal, oblique and even 3D, and a protocol with the batch mode set to Oblique.
For a radial oblique batch you use an axial, sagittal, coronal or oblique view (but
not 3D), and a protocol with the batch mode set to Rotation.
For a 3D batch you use a 3D view, and a protocol with the batch mode set to Rota-
tion.
Note: Most loading/processing protocols already have a default batch protocol
attached to it which is shown in the Batch Film panel of the protocol, or as
soon as you select (Filming Tools) > [Batch Film].
If required, you can use (Protocols) in the Batch Film panel to display the
Batch Protocols list and select another protocol.

Oblique batches
With the parallel and radial oblique batches, a set of reference lines is displayed
on the primary view, indicating the position of the images that will be included in the
batch.
These reference lines serve as on-view controls: you can drag on them to position
the slices, to place the first and last slice and define the slice spacing (which deter-
mines the number of slices).
Some of the parameters, such as number of views and spacing, can also be set up
by typing the corresponding values in the Batch command window.
For an initial alignment of the reference lines (left, right, up or down) you can use
the cursor keys on the keyboard.

14-6
 2271012-100.book

Spacing annotation
(mm between views)
Cross-reference marks
(number of views+mm between
views)
Page 7 Thursday, May 8, 2003

Drag on any one to change number

of views and distance between
them.
11:30 AM

Slice center reference line

Drag here (central red square)

to move all markers together

Drag here (red squares) to rotate

all markers together

Drag here (red arrows) to move

end markers (number of views)

Batch end markers

View with parallel oblique

batch prompts
Volume Analysis

14-7
14-8
2271012-100.book

Volume Analysis

Rotation annotation
(angle between views)
Cross±reference marks
(number of views+angle between
views)
Drag on any one to change number
of views and angle between them
Page 8 Thursday, May 8, 2003
11:30 AM

Drag here (red square on first

batch marker) to rotate all markers
together

Drag here (central red square)

to rotate all markers together

Drag here (red square on last batch

marker) to change number of views

View with radial oblique

batch prompts
2271012-100.book Page 9 Thursday, May 8, 2003 11:30 AM

Volume Analysis

3D batch
With a 3D batch, a set of arrows on the view allows you to select the direction of
rotation: left, right, up or down.
The 3D object can rotate only around a vertical axis (left/right) or horizontal axis
(up/down), but you can move, tilt and rotate the 3D object as required before start-
ing the batch.
A 3D batch always consists of a full 360 degrees rotation of the 3D object.
You can define either the number of views in the batch or the angle of rotation (the
two are interdependent) in the Batch command window.
To select the direction of rotation, click on the batch rotation arrows (see illustration
on the next page), or use the cursor keys on the keyboard.

14-9
14-10
2271012-100.book

Volume Analysis

Number of views
annotation
Drag here to move batch
rotation selector
Page 10 Thursday, May 8, 2003

Batch rotation arrows

Rotation annotation
11:30 AM

(angle between views)

3D view with
batch prompts
2271012-100.book Page 11 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Preview, filming, saving

Once you have defined the batch protocol:
• You can preview the resulting batch as an animated sequence (movie loop). The
controls allow you to set the display rate (frames per second), to pause and
restart the sequence, and to move through the set step-by-step.
• You can film the batch, i.e. send the image set to a hardcopy device such as a
laser camera. The format (image layout on the film) is defined in the batch proto-
col, or can be selected in the Batch command window.
• You can save the batch on the image disk of the workstation as a new series in
the current exam, for later viewing and/or processing using other AW applica-
tions. An oblique (reformatted) batch can be saved either as a set of reformatted
images (type REFORM) or as secondary captures (type SCPT). A 3D batch is
always saved as a set of secondary captures (type SCPT).

14-11
2271012-100.book Page 12 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Custom Protocols
Creating a custom batch protocol
You can create custom batch protocols from the existing batch protocols.
Select an existing protocol. Use (Modify) to change the setup to match your
requirements.
You can now (Accept) the setup and return to the main Batch Film panel to use the
new setup, or you can use (Save Protocol) to save it with a new name.

Saving a custom batch protocol

To save a custom protocol, select (Save Protocol). You now have two options:
- You can add it to the list of existing batch protocols for the current
“loading” protocol (i.e., the loading/processing protocol used to build the
3D model):
select (Use) > [with current loading protocol],
- You can combine it with the current loading protocol to create a new
loading protocol:
select (Use) > [as new loading protocol].
You will be asked for the name of the protocol. Enter the new name, using only
alphanumeric characters and spaces. Avoid long names.
Note: If you use the name of an existing protocol, you will be given the option
either to overwrite the existing protocol or to cancel and change the name.
If you have selected Use as new loading protocol you will also be asked to select
the anatomical category for the new protocol.
Finally you can (Save) the protocol, or (Cancel) the operation.

Multiple batch protocols

If you regularly work on exams that require the use of more than one batch proto-
col, Volume Analysis 2 allows you to combine two or more batch setups in a single
custom multiple-batch protocol.
You can use existing protocols, or use (Modify) to define the setups to your require-
ments.
To assemble a multiple-batch protocol, you select and/or modify each setup as nec-
essary, then click on (Add Step) to add it to the protocol.
The (Previous Step) and (Next Step) buttons allow you to move through the suc-
cessive setups in the protocol, and make corrections as necessary.
Use (Save Protocol) to save the multiple-batch protocol as before (see above).

14-12
2271012-100.book Page 13 Thursday, May 8, 2003 11:30 AM

Volume Analysis

3 MOVIE LOOP
To initiate the Movie Loop mode, click on the [Movie Loop] option of the drop-down
menu displayed by clicking on the (Filming Tools) button of the Control Panel. The
following command window is displayed on the bottom left-hand view of the viewing
area:

The various buttons and parameters in this command window during movie looping
are described below:

First View button

This button allows you to specify the first view to be used in

the movie loop and the Target view of the Movie.

Move and/or rotate the view to the desired first position.

Click left on the (First View) button to enter the first view. The name of the first view
appears on the upper left corner of the Movies command window and the parame-
ters that are specific to that view type become visible in the Movies Definition panel
here above.

14-13
2271012-100.book Page 14 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Final View button

This button allows you to specify the final view to be used in

the movie loop.

Move and/or rotate the view to the desired final position.

Click left on the (Final View) button to enter the final view.
Note: In the case of a 3D image, specifying only the first view results in a default
360 degree rotation around the vertical axis of the image.

Creating a movie loop on oblique images

The Movies command window can be used to generate a movie loop sequence of
oblique views between two defined oblique views as follows:
• Prescribe the desired first oblique view in the sequence,
• Select this oblique view as primary,
• Click left on the (First View) button,
• Prescribe the desired final oblique view in the sequence,
• Select this oblique view as primary,
• Click left on the (Final View) button
Note: The first and final oblique views do not have to be parallel.

Number of views and Angle between views fields

Number of views These fields are used to specify either a desired num-
ber of images or an angle step size (or a distance step
10
size for reformatted slices) between images.

Angle between views: Move the mouse pointer into this field, delete the exist-
ing value by pressing the Back Space key once for
10.0
each digit to be deleted and enter either the number of
images (minimum possible number of images is five)
or the angle step size in degrees (or distance step size
in mm for reformatted slices) from the keyboard.

14-14
2271012-100.book Page 15 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Output Select Buttons

These buttons are used to select the des-

tination of the movie loop when you click
left on the (START) button.The (Pre-
view) button is used to view the movie
loop sequence on the monitor. With this
option, no film output is generated .

The (Film) button is used to send the movie loop to the laser imager for film gener-
ation.
Note: The (Film) button can only be activated if a hard copy output device is con-
nected to the AW workstation.
The (Save) button is used to store the movie loop in the data base. The movie loop
will subsequently appear in the Browser window as a new, separate series which
can be viewed and manipulated via the AW Viewer or Mini Viewer application.
Click left on the desired button.

Format button

When you click left on this button, a list of film formats appears for the movie loop.
Once you have selected the desired format (no. of rows x no. of columns), it
appears on the surface of the (Format) button.
Note: The list of available formats depends on the type of laser imager being used.
Note: The (Format) button is disabled (gray lettering) when the (Film) or (Film/
Save) Output select button is NOT enabled.

START button

To launch movie loop generation, click left on the (Start) button.

Note: The movie loop will be sent to the monitor, laser imager or data base
depending on the choice of (Output) select button. See “Output select but-
tons” above.

14-15
2271012-100.book Page 16 Thursday, May 8, 2003 11:30 AM

Volume Analysis

CLOSE button

To close the Movies command window, click left on the

(CLOSE) button.

14-16
2271012-100.book Page 17 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Previewing a film batch or movie loop

When you have set up a movie loop or film batch, activated the (Preview) button,
and clicked on the (Start) button to launch it on-screen, the Movies command win-
dow appears as shown above.
The various buttons and parameters in the Movies command window are described
below:

Display speed slider

10
The Display speed slider allows you to adjust the preview
loop frame rate.
Display speed

Hold left on the slider and drag left or right to obtain the desired display speed. Or
place the mouse pointer to the right of the current slider position and click left to
increase the value by one, or place the mouse pointer to the left of the current slider
position and click left to decrease the value by one.

14-17
2271012-100.book Page 18 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Image index slider

1
The image index slider allows you to freeze the preview
loop and go to a desired frame in the loop.
Image index

Hold left on the slider and drag left or right to obtain the desired frame in the loop.
Or place the mouse pointer to the right of the current slider position and click left to
increase the value by one, or place the mouse pointer to the left of the current slider
position and click left to decrease the value by one.

Stop button

To exit the preview display mode and return to the normal

Movies command window, click left on the (Stop) button.

Pause/Restart button

To freeze the preview loop during viewing, click left on the

(Pause) button. The button becomes the (Restart) button.
Click left on the (Restart) button to continue displaying the
preview loop. The button again becomes the (Pause) but-
ton.

Step button

To advance a frozen preview loop by one frame, click left

on the (Step) button.

CLOSE button

To close the Movies command window, click left on the

(CLOSE) button.

14-18
2271012-100.book Page 19 Thursday, May 8, 2003 11:30 AM

Volume Analysis

4 LINK WITH DATA EXPORT APPLICATION

You can record images or sets of images as JPEG images or MPEG movies, using
the GE Data Export application.

This feature is only available on AW 4.0 systems, with

CAUTION the Data Export 1.0.151 or higher versions.

Recording images as JPEG images

To record single images, or the entire screen either as 4 different images or as a
single image, you use the <Alt F1>, <Alt F2> or <Alt F3> keyboard keys as in AW
Basic Display. Saved images are thus sent to the current Data Export session as
JPEG images (see Data Export User Guide, GE reference 2285816-100, for more
information about Data Export sessions).

Recording batch sequences as JPEG series or MPEG movies

To record batch sequences as JPEG series or MPEG movies, you use respetively
the JPEG series or MPEG movie ouput modes. Saved sequences are thus sent to
the current Data Export session (see Data Export User Guide, GE reference
2285816-100, for more information about Data Export sessions).

Recording movie loops as JPEG series or MPEG movies

To record movie loops as JPEG series or MPEG movies, you use respetively the
JPEG series or MPEG movie ouput modes. Saved loops are thus sent to the cur-
rent Data Export session (see Data Export User Guide, GE reference 2285816-
100, for more information about Data Export sessions).

Recording fly-through sequences as MPEG movies

To record Navigator fly-through sequences as MPEG movies, you use the Save
MPEG movie button. Saved sequences are thus sent to the current Data Export
session (see Data Export User Guide, GE reference 2285816-100, for more infor-
mation about Data Export sessions).

14-19
2271012-100.book Page 20 Thursday, May 8, 2003 11:30 AM

Volume Analysis

Managing Data Export sessions

You can export the current Data Export session, made of JPEG images and/or
MPEG movies, either to a given address (using the ftp protocol), or to the http
server of your AW station, or on a CDROM, using the (Filming tools) > [Export
Session] option.
You can get information about the current Data Export session (number and com-
position of the pages), using the (Filming tools) > [Session Info] button.
See Data Export User Guide, GE reference 2285816-100, for more information
about how to handle Data Export sessions.

14-20
2271012-100.book Page 21 Thursday, May 8, 2003 11:30 AM

Volume Analysis

5 SAVING THE 3D MODEL

You can save the current 3D model on the image disk, using the (Save/Recall)
function, so that you can reload it at a later time for further processing.
The 3D model is saved as a new separate series in the current exam. It is shown in
the Browser Series list as a series of Type 3D OBJ.
Even though it is listed on the Browser as a series containing only one “image”, a
saved 3D model can take up very substantial amounts of image disk space. You
can only save the current 3D model if there is sufficient space on the image disk.
Saved 3D models can be reloaded for further processing by Volume Analysis 2.
They can also be archived, or networked to another workstation that has the AW3.1
software (or later version) installed. They cannot be viewed using other viewing
applications such as the AW Basic Display Viewer or Mini Viewer.
Note: Volume Analysis 2 can also directly load and process 3D models that were
generated by other applications such as Advantage 3DXR.
Also see Chapter 5 “Building the 3D Model”, paragraph 6 “Loading a 3D Model”.

14-21
Chapter14.fm Page 22 Thursday, May 8, 2003 1:33 PM

Volume Analysis

Blank page.

14-22
Glossary.fm Page 1 Thursday, May 8, 2003 1:29 PM

Volume Analysis

GLOSSARY
Note: Words in italics refer to terms defined elsewhere in the Glossary.

3D Model - The representation of the 3D (three-dimensional) image data in the

workstation computer memory. 3D models can be saved, archived and networked
(3D OBJ format).
Usually, a 3D model is created by the software at startup from a CT or MR
image set. However, saved 3D models, or 3D models generated by Advan-
tage 3DXR can also be loaded and processed by the Volume Analysis 2
application.
3D OBJ - Computer file format (“Type” in Browser series list), used to store a 3D
Model on the workstation. Also used by Advantage 3DXR to store 3D data directly.
3DXR (3D X-ray) - Process of deriving anatomical information by computer synthe-
sis of X-ray data acquired at multiple angles by means of a conventional X-ray sys-
tem.
Active Annotation - An system annotation on a view that can be modified by the
user to control certain viewing parameters (e.g., window width and level). Active
annotation are displayed in red.
Advantage 3DXR - Software application used to derive a 3D data set from X-ray
data acquired at multiple angles by means of a conventional X-ray system. A data
set generated by Advantage 3DXR uses the 3D OBJ format (3D object).
Annotation - Generally, workstation-supplied text which accompanies an image
when it is displayed on-screen, describing when and how that image was acquired,
with what parameters. Also, text and measurement information added on a view by
the user.
Artifact - Feature in an image resulting either from the initial data acquisition or
subsequent computer processing that does not correspond to a real feature in the
original anatomical structure. Also see Partial Volume Effect.
Baseline view - A basic axial, coronal or sagittal view, aligned parallel to the main
axes of the RAS coordinate system.
CT (Computed Tomography) - Process of deriving anatomical information by
computer synthesis of X-ray data, acquired by means of a CT scanner in the form
of parallel “slices”.
CTA (CT Angiography) - The use of CT techniques optimized for angiography.
DFOV (Display Field Of View) - The real dimensions of a view (width and height)
with reference to the RAS coordinates.

1
Glossary.fm Page 2 Thursday, May 8, 2003 1:29 PM

Volume Analysis

DICOM - Abbreviation for Digital Imaging and Communications in Medicine. Stan-

dard for the formatting and exchange of medical images and associated informa-
tion.
Exam - In MR, a single study, including all its component series and scans. In CT,
all images made from data taken of a patient after entering a particular scan cycle.
In 3DXR, a 3D model generated by Advantage 3DXR.
Free-hand Trace - Graphical tool that allows you to define an ROI by “drawing” a
free-hand trace on a view.
HU (Hounsfield Unit) - Scale unit denoting voxel density in a CT data set.
Image - In this document the term “image” is used to designate the part of the exam
data being processed and displayed on the workstation screen. Depending on the
display settings, a view (q.v.) can display an entire image, or part of it (zoom).
Layout Preset - Custom procedure allowing the user to memorize in a file the view-
ing parameters best adapted to the current exam, and equivqaent to a custom pro-
tocol.
Lumen view Cross-reference view showing an unfolded linear representation of
any kind of tubular structure (i.e. colon, vessel...)
Measurement Annotation - A user annotation on a view that shows the result of a
measurement.
MPVR (Multi-Projection Volume Reconstruction) - A technique that allows you
to define and display a “thick” slice that encompasses a feature of interest, instead
of using baseline and oblique views that represent slices that are only one voxel
thick.
MR, MRI (Magnetic Resonance Imaging) - Creation of images using the magnetic
resonance phenomenon. Most of the current applications involve imaging the distri-
bution of hydrogen nuclei (protons) in the body.
MRA (MR Angiography) - The use of MR techniques optimized for angiography.
On-view menu - Menu displayed either on a view or on a particular feature of a
view such as a trace, a reference image, etc. by pressing the right mouse button.
Partial Volume Effect - The appearance of voxels with intermediate values at the
interface (separating surface) between two tissue types with clearly distinct and dif-
ferent densities, where these voxels do not correspond to a real feature in the origi-
nal anatomical structure.
Pixel - Abbreviation for “picture element,” the smallest unit a computer screen can
display.
RAS - Abbreviation for Right/Anterior/Superior. Designation for the patient-linked
coordinate system used in CT and MR data sets.
Reference Image - Small image (normally displayed in the bottom right corner of a
view) that graphically indicates the orientation of the image displayed in the view
with respect to a baseline image (axial, coronal or sagittal).

2
Glossary.fm Page 3 Thursday, May 8, 2003 1:29 PM

Volume Analysis

Rendering - Techniques used to represent a three-dimensional object on a two-

dimensional surface.
Report Cursor - Graphical tool that allows you to read the value and 3D location of
any voxel on the image being viewed.
Review Controller - Square on-view tool appearing around the active view, on
which the user will find basic control buttons needed with the specific current soft-
ware.
ROI (Region Of Interest) - User-defined area of an image to be analyzed.
Segment Trace - Graphical tool that allows you to define an ROI by marking a set
of points on a view that are connected by the software with straight-line segments.
Smart Cursor - Graphical tool used to interactively fly through an aerial way (i.e.
colon, vessel...).
Spline Trace - Graphical tool that allows you to define an ROI by marking a set of
points on a view that are connected by the software with a smooth curve (spline).
Standard Deviation - Statistical measurement that provides a measure of variabil-
ity of voxel values within an ROI.
System Annotation - An annotation on a view added by the system software, con-
taining data concerning the displayed image. Certain system annotations can be
active, i.e., they can be modified by the user.
Text Annotation - A user annotation on a view containing text. Text annotations
can be used to add comments, or to add a legend to an anatomical feature on a
view.
Title Bar - A bar at the top of a window that provides information about that window,
and also allows you to move the window to a different position on the screen by
clicking and dragging on it with the mouse.
Trace - Graphical tool that allows you to define an ROI on a view, or perform mea-
surements (distance, angle, area).
User Annotation - An annotation on a view added by the user, containing either
text or the result of a measurement.
View - Part of the workstation screen, used to display image data. The view area of
the Volume Analysis 2 screen normally contains four views. A view can display an
entire image, or part of it (zoom).
View Area - During use of a viewing application, the portion of the screen(s) where
images are displayed. The view area normally contains four views, but a single
view can be enlarged so as to take up the entire view area.
Voxel - Abbreviation for “volume element,” the basic element in a CT or MR data
set.

3
Glossary.fm Page 4 Thursday, May 8, 2003 1:29 PM

Volume Analysis

Window Width and Level (W/L) - In this context, “window” refers to the range of
pixel values within the image data, that is assigned a shade of grey for display.
“Level” refers to the center value, “width” refers to the range of pixel values dis-
played around this central level.
The adjustment is marginally similar to adjusting brightness and contrast
controls: a narrow window (low width) translates to a high contrast of the dis-
play, and similarly a low level translates to a high value of brightness.

4
 Table of Contents Perfusion
(Optional)

CHAPTER 1 - GENERAL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1 SYSTEM REQUIREMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Patient Confidentiality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
2 HOW TO USE THIS DOCUMENT. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
3 CONVENTIONS FOR THIS MANUAL . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

CHAPTER 2 - INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1 OVERVIEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2 BASIC PRINCIPLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
3 PROTOCOLS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
4 DISPLAY LAYOUT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
5 CONTROL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
6 SUMMARY OF OPERATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Start . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Protocol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
ROIs and Text Annotations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Recording . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
7 PIXELS AND VOXELS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

CHAPTER 3 - GRAPHS AND FUNCTIONS . . . . . . . . . . . . . . . . . . . . . . . . . 17

1 OVERVIEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
Loading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
Views and Controls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
Perfusion (Optional)

2 SERIES VIEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
3 GRAPHS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
Graph View . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
4 FUNCTIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Function View . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
5 COMPOSITE VIEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

CHAPTER 4 - PROCEDURES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
1 IMAGE REQUIREMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
2 START PERFUSION 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
3 PROTOCOLS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
Help . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
Image Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
Image Registration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
Processing Thresholds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
Artery Input . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
Vein Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
Automatic Reference ROIs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
Defining Artery and Vein ROIs at a Different Image Location . . . . . . . . . . . . . 48
Enhancement Image Range . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
Final Settings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
Advanced Settings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
Compute . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
4 IMAGE IMPORT PROTOCOL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
Protocol Panels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
 Perfusion (Optional)

CHAPTER 5 - CONTROL PANEL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57

1 OVERVIEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
2 CONTROL PANEL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
3 TASK BAR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
4 MAIN CONTROL PANEL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
Protocols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
Graph View Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
Central Panel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
View Control and Mainpulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
Mouse Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
Review Controller . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
Image Orientation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
Display Normal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
Grid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
Smoothed or Unsmoothed Images . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
Annotations and ROIs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
Preferences/Settings Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
Graph View . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
Saving . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
Regions Of Interest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
Film/Save Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71

CHAPTER 6 - ON-VIEW CONTROLS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73

1 OVERVIEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
2 ON-VIEW CONTROLS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
Adjusting Window Width and Level . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
Manual . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
Auto W/L . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
W/L Presets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
Perfusion (Optional)

Active Annotatios . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
Patient name . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
Scan Location . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
DFOV (Display Field Of View) . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
Roam (Scroll) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
ROI annotation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
Window W/L annotations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
3 SERIES VIEW CONTROLS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
Series View Active Annotations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
Series View On-View Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
Review Controller . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
4 GRAPH VIEW CONTROLS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
Graph On-view Controls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
Graph View Active Annotations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
Graph View On-View Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
Auto Scaling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
5 FUNCTION VIEW CONTROLS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
Function View Active Annotation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
Color Ramp Controls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
Function View On-View Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
Composite View . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
Transparency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
Window W/L . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
Selection of the Reference Image . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
Saving Composite Views . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
 Perfusion (Optional)

CHAPTER 7 - REGIONS OF INTEREST AND TEXT ANNOTATIONS . . . . 101

1 OVERVIEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
2 DEFINING A REGION OF INTEREST (ROI) ON THE VIEWS . . . . . . . . . 103
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
Cursor Region of Interest (ROI) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
Cursor size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
Cursor annotation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
User-defined Regions of Interest (ROIs) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
Accuracy of on-view measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
Single-pixel ROI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
Ellipse and Rectangle ROIs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
Spline/Polygon ROI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
Freehand ROI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
Density mask ROI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
Pixel List ROI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
Mirror ROI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
Split/Merge ROI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
Group/Ungroup ROI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
3 IMAGE ANNOTATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
4 GRAPH VIEW ANNOTATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
5 MANAGING USER ANNOTATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124

CHAPTER 8 - RECORDING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125

1 OVERVIEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
Controls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
Hardcopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
Saving Views . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
Saving Processed Images . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
Export . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
Perfusion (Optional)

Annotations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
2 CONTROL. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
Film/Save Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
Film Composer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
Series Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
Graph Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
Functional Map (s) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
On-View Menus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
Keyboard . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
3 FILMING AND PRINTING. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
4 SAVING VIEWS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
5 SAVING PROCESSED IMAGES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
Series Data Subtraction Images . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
Series Data Reformatted Images . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
Functional Maps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
Pixel Values in Saved Images . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
6 EXPORT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
Save View in TIFF Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
7 ANNOTATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
Show/Hide Annotations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
Saving Annotations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
Patient Name . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145
Cursor Annotation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145

APPENDIX 1- FUNCTIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147

1 OVERVIEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148
General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148
Algorithms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148
Input Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
2 CT PERFUSION ALGORITHM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150
Computed Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
 Perfusion (Optional)

Blood Flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151

Mean Transit Time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
Blood Volume . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152
Permeability Surface Area Product . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152
3 OTHER ALGORITHMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153
Positive Enhancement Integral . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153
Time To Peak . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153
Maximum Slope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154
4 INPUT PARAMETERS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
Input Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156
Pre-enhancement image(s) and Post-enhancement image . . . . . . . . . . . . 157
Processing Thresholds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157
Artery ROI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157
Vein ROI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158
Advanced Settings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159
5 PIXEL VALUES IN SAVED IMAGES. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
Functional Maps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160

GLOSSARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161
Perfusion (Optional)

Blank page.
 CT Perfusion

CHAPTER 1 - GENERAL
The CT Perfusion 2 package is an optional software package for CT systems.
It allows the analysis of CT data sets that contain multiple images for each scan location
in the data set, representing changes in image intensity over time (“time series” or
“dynamic” data sets).
Results are in the form of graphs and functional maps, that can be saved and printed to
film or a color paper printer.

1 SYSTEM REQUIREMENTS
This software application runs on the standard Linux platform. It uses the platform’s sys-
tem for film output. It accepts image data acquired on any CT system compatible with
Image Work.

Patient Confidentiality
The equipment on which the CT Perfusion 2 application runs
CAUTION
includes one or more hard disk drives which may hold medical data
related to patients. Such equipment may in some countries be sub-
ject to regulations concerning the processing of personal data and
the free circulation of such data.
It is strongly recommended that access to patient files be protected
from all persons not in medical attendance.

1
CT Perfusion

2 HOW TO USE THIS DOCUMENT

• For a general overview of the features and functions of the CT Perfusion 2 software,
read Chapter 2 “Introduction”.
• For a description of the two main analysis tools: graphs and functions, available in CT
Perfusion 2, read Chapter 3 “Graphs and Functions”.
• To start CT Perfusion 2 from the Browser/Patient List, load the exams and use the CT
Perfusion 2 protocols to process the exams, refer to Chapter 4 “Procedures”.
• To familiarize yourself with the tools and controls in CT Perfusion 2, consult Chapter 5
“Control Panel” and Chapter 6 “On-view Controls”.
• For a list of the keyboard controls available in CT Perfusion 2, refer to Chapter 7 “Key-
board”.
• To add user annotations on the images displayed by CT Perfusion 2, see Chapter 8
“Regions of Interest and Text Annotations”.
• To save and/or film the results, refer to Chapter 9 “Recording”.
• For more information on the theoretical background of the functions (algorithms and
derived functions) used by CT Perfusion 2, see Appendix 1 “Functions”.
• For a list of the terms in this manual that are specific to the application, refer to the
Glossary at the back of the manual.

2
 CT Perfusion

3 CONVENTIONS FOR THIS MANUAL

Throughout the text in this manual, certain type styles and symbols are used to differenti-
ate between one tool or graphic and another:
• Menu and control panel titles appear in bold face: Function menu,
• Menu options appear in bold face, within square brackets: [Exit],
• Graphical buttons with text legends appear in bold face, within parentheses: (New Pro-
tocol),
• Graphical buttons with icons appear in bold face and lower case, within parentheses:
(rotate right),
• On-screen tools appear in bold face and lower case, within braces: {scroll bar},
• On-screen prompts and messages appear in italics: Enter key value: ,
• Mouse buttons are underlined: left.
Whenever this manual refers to “clicking”, “selecting”, “pressing a button”, etc. with the
mouse, this always means using the left mouse button, unless the middle or right mouse
button is specifically mentioned.
Safety notice legends:
THIS INDICATES A POTENTIALLY HAZARDOUS SITUATION WHICH, IF NOT AVOIDED,
W ARNING
COULD RESULT IN DEATH OR SEROUS INJURY.

This indicates a potentially hazardous situation which, if not avoided, may result in minor
CAUTION
or moderate injury
This indicates a non±hazardous situation which, if not avoided, could result in equipment
NOTICE damage, lost time, or reduced image quality

3
CT Perfusion

CHAPTER 2 - INTRODUCTION
CT Perfusion 2 has been developed to analyze CT data sets that contain multiple images
for each scan location, representing changes in image intensity over time (“time series” or
“dynamic” data sets).
CT Perfusion 2 provides you with two main tools for analysis:
• Graphs: these show the changes in pixel value over time at a given point on the
images, either at the cursor position or within a region of interest (ROI) defined by the
user.
• Function images: the pixel values from the data set are used to compute a characteris-
tic parameter for every pixel location by means of a function. This can be a basic algo-
rithm or a function derived from such an algorithm.
A “functional map” is then constructed by displaying the value of this parameter for
every pixel location.
CT Perfusion 2 is supplied in the form of protocols. Each consists of a succession of pro-
tocol panels that contain the necessary information and the controls required to set up an
analysis.
You also control CT Perfusion 2 by means of the on-screen control panel, on-view con-
trols, and the keyboard. You can place both text and graphic (ROI) annotations on the
views. The results, in the form of graphs, tables and images, can be saved on disk and
printed to film or a color paper printer.
This chapter provides you with a general overview of the operations and controls avail-
able in the CT Perfusion 2 software.

4
 CT Perfusion

1 OVERVIEW
The CT Perfusion 2 application is started from the Image Works Browser List and has the
following main features:

Restart Browser

Restart Display

Image File Search

Application Selection Remove Sort Network Archive PPS Queue Utilities Messages
Examinations : Exam #47, May 05 92, SMITH, JON
Exam Name Date Description Mod Fmt A Ser Type Imgs Description Mod VA

3145 J.Herman Jan 08 98 CT 1 PROSP 18 CT Add/Sub

3512 B.Fox Dec 23 97 CT

Edit Patient

Film

Mini Viewer
2 examinations one series
Series
Img Img Ctr Thck/SP Gntry Img Ctr Img Ctr SFOV DFOV Perfusion2
(deg) Res Matrix Midscn Archive
S–I (mm) (mm) R–L A–P (cm) (cm)
Viewer
1 S 50.0 10.0/ +0.0 R5.0 A0.0 43 Lung 512 No
2 S 45.0 10.0/ +0.0 R5.0 A0.0 43 Lung 512 No
3 S 40.0 10.0/ +0.0 R5.0 A0.0 43 Lung 512 No
4 S 35.0 10.0/ +0.0 R5.0 A0.0 43 Lung 512 No
5 S 30.0 10.0/ +0.0 R5.0 A0.0 43 Lung 512 No
6 S 25.5 10.0/ +0.0 R5.0 A0.0 43 Lung 512 No

60 images

- Setup of processing parameters from protocol panels,

- Checking the exam for possible patient motion using a “cine” function,
- Selecting and loading of the image data from the Image Works Browser List,
- Display and plotting of image intensity information, using the cursor and
regions of interest (ROIs),
- Creation of functional maps,
- Creation of composite images (overlay of functional map onto reference
image),
- Addition of text annotations,
- Saving, printing and transferring of the processed data.
Some of the main characteristics of the CT Perfusion 2 application:
- Allows selection of any time series, or “dynamic” data set,
- Automatic loading of the selected series or images, including all slice
locations and time points,

5
CT Perfusion

- Capable of loading image data sets of up to 1,000 images,

- Image matrix sizes ranging from 32x32 to 2048X2048,
- Allows multiple series and images.

6
 CT Perfusion

2 BASIC PRINCIPLES
CT Perfusion 2 has been developed for the analysis of CT data sets that contain multiple
images for each scan location in the data set. They are referred to as “time series” or
“dynamic” data sets, and are acquisitions with a sequential change in scan time for each
scan image (such as contrast take-up studies acquired using cine imaging).
Such data sets contain for each scan location a series of N sequentially ordered images,
where the interval 1..N represents time.
The pixel value in an area of interest at a given scan location may have a different value
for EACH image. To analyze these changes in pixel value, CT Perfusion 2 provides two
tools:
- Graphs: the pixel values at a given pixel location p for the images 1..N are
plotted as a graph of pixel value over image number or time,
- Function images: for every pixel location p the pixel values from the images
1..N are used to compute a characteristic parameter by means of a function.
A functional map is then constructed by displaying the value of the parameter
for every pixel location.
Also see illustration on next page.
For the analysis of blood perfusion, CT Perfusion 2 provides functions to compute quanti-
tative functional maps for regional blood volume (BV), mean transit time (MTT), regional
blood flow (BF) and permeability surface area product (PS).
Additionally, general-purpose functions are available to generate functional maps for pos-
itive enhancement integral, time to peak, maximum slope of increase and maximum
slope of decrease.

Series of N images Extract the intensity curve Use a function to compute a Set intensity of p
at the same location. of pixels p. characteristic parameter v in the function image to v.

N HU

F
2
1 f(v1, v2, ..., vN)=v p
p

1 2 N v
v1 time

For full details about the theoretical background of each function, refer to Appendix 1
“Functions”.

7
CT Perfusion

3 PROTOCOLS
CT Perfusion 2 is supplied in the form of protocols.
After starting the application and selecting a protocol, each protocol displays a succes-
sion of protocol panels that contain the necessary information, steps and controls
required to set up the analysis.
Overall, the protocols use the same sequence, but not all panels and controls are present
in all protocols.
When setup is complete, the software uses the functions contained in CT Perfusion 2
(see above) to compute and display the functional maps appropriate to each protocol.
CT Perfusion 2 is supplied with the following protocols, each optimized for generating
quantitative functional maps for a specific type of analysis:
- The Brain Stroke protocol generates maps for blood volume, blood flow and
mean transit time,
- The Brain Tumor and Body Tumor protocols generate an additional map for
permeability surface area product (used in examinations of tumors),
- The CT Standard protocol generates maps for various other perfusion-related
parameters: positive enhancement integral, time to peak, maximum slope of
increase and maximum slope of decrease.
An Image Import protocol allows you to reload saved functional maps for further review
and processing.

8
 CT Perfusion

4 DISPLAY LAYOUT
The CT Perfusion 2 screen display consists of a main control panel at the left, and a view-
ing area at the right. The viewing area contains four views:

The series view (upper left) displays the reference

image,
The graph view (upper right) displays the image intensity
curves,
The two function views (lower left and right) display the
functional maps (these views are empty until the func-
tional maps have been computed).
The protocol panels used to set up an analysis (see
above) are normally displayed on top of the lower left
view, but they can be moved or closed temporarily if
required.

Series view
You use the series view to display the currently selected original image from the exam
series, to control and keep track of the image location (scan location and rank), to define
regions of interest (ROIs), and to examine the image data for possible patient motion
prior to analysis.
Graphs
Graphs can be plotted for the pixel values at the current cursor position, or for the pixel
values within one or more regions of interest (ROI) defined on the views by the user.
Alternatively, CT Perfusion 2 can display information in the form of a histogram or as a
table.
Function views
The functional maps computed by the software are displayed in the function views. The
user selects which map to display in which view. Each function view can be set to display
either a single full-size map or four small half-size maps.
Composite view
To correlate a functional map and the original (reference) image, you can set up a func-
tion view to display an overlay of the functional map and the reference image.

9
CT Perfusion

5 CONTROLS
You control CT Perfusion 2 by means of the on-screen main control panel, on-view con-
trols, and the keyboard, in addition to the controls available in the protocol panels during
the setup of the analysis.
The buttons and menus available in the main control panel are described in full in Chap-
ter 5.
On-view controls are controls that you access and use directly on the views, such as:
- Window width/level adjustment,
- Active annotations: these are annotations displayed in red, that can be
modified to adjust view parameters,
- On-view menus that contain functions specific to each view. You access
these by clicking right anywhere on the view (except on the active
annotations).
- A review controller that lets you move through the images using sliders and
buttons or by means of a cine loop.
The on-view controls are described in full in Chapter 6.
Certain functions are controlled from the keyboard. These are described in the relevant
sections of this manual, and summarized in Chapter 7.
The <Help> key on the keyboard, or the <Ctrl-H> key combination, allows you at all times
to view the full list of available keyboard commands, without having to refer to the man-
ual.

10
 CT Perfusion

6 SUMMARY OF OPERATIONS
A typical session using CT Perfusion 2 for analysis consists of the following operations:
Start
Select the exam to be analyzed in the Image Works Browser and start CT Perfusion 2,
then select the appropriate protocol.
Prototol
Step through the successive panels displayed for the selected protocol. The summary
below lists all panels that you may encounter (not all panels are present in all protocols).
You can move forward and backwards through the panels until you are satisfied with the
settings.
Image Review and Image Registration
For proper results from the CT Perfusion 2 analysis tools, the image data must be coher-
ent, i.e., not degraded by patient motion during the acquisition.
Use the review controller (cine loop, sliders) to examine the image data for possible
patient motion before analysis. An Image Registration function is available in the Brain
Stroke and Brain Tumor protocols that can remove patient motion in the slice plane auto-
matically by registering (aligning) the images.
Processing Thresholds
By excluding non-relevant data from the processing, you can significantly reduce pro-
cessing time and avoid the appearance of spurious noise on the functional maps.
On the first image of the series, set the lower threshold to exclude air spaces (including
the empty space around the patient), and the upper threshold to exclude bone. Only the
selected range of Hounsfield Unit values between the thresholds will be processed.
Artery Input and Vein Output
The functions used to compute blood volume, mean transit time, blood flow and perme-
ability surface area product require ROIs to be placed inside an artery, to account for the
injection rate of the contrast agent, and inside a large vessel to normalize the results for
the HU value for 100% blood (usually a vein, but any vessel large enough to be fully rep-
resentative of the HU value for 100% blood is acceptable even if it happens to be an
artery).
Follow the instructions in the panels to place these ROIs, either automatically or manu-
ally.

11
CT Perfusion

Enhancement Image Range

The pre-enhancement image range is used by the software to determine a baseline.
The post-enhancement image range setting is used primarily to limit processing time.
This may be necessary in the case of long-duration acquisitions (several minutes) as
used in tumor studies. Otherwise, for best results it should be left at the default value
(last image of the data set).
Use the sliders on the panel together with the curves on the graph view to set the pre and
post enhancement image ranges.

Advanced Settings
The protocols use default settings for certain parameters (such as maximum blood flow,
hematocrit ratio, tissue density, resolution used by the algorithm, image data smoothing).
For specific cases, you may want to adjust one or more of these parameters. E.g., the
hematocrit ratio may be higher for an exam of a small infant.

Compute
After you have set up all parameters for the analysis to your satisfaction, select (Com-
pute) in the last protocol panel to compute and display the functional maps.
Once the functional maps are displayed you can close the protocol panel (without losing
the settings) to remove it from view. You can recall the protocol and make changes to the
settings if required, then select (Compute) again to recompute the functional maps.
Both algorithm resolution and image data smoothing (noise filtering) affect the final
results (functional maps). Lowering the algorithm resolution decreases the resolution of
the functional maps in proportion, but also speeds up the processing. Increasing the spa-
tial smoothing factor (size of the gaussian filter) smooths the image data used for the
computation, reducing noise on the functional maps.

12
 CT Perfusion

Review
A full range of features is available to view and analyze the functional maps. These are
summarized here, and described in more detail elsewhere in the manual.
- Selection of the functional map to be displayed in each view,
- Display of either one or four functional maps per view (full-size or reduced-
size views),
- Image manipulation such as zoom, scroll, rotation and flip of the images,
- Selection of different color ramps (or a range of gray levels) for the display of
the functional maps,
- Adjustment of the pixel value range represented by the color ramps or gray
levels (equivalent to window W/L adjustment),
- Creation of composite views, consisting of the overlay of a functional map on
a reference image with user-adjustable transparency (only on full-size views),
- Creation of ROIs (regions of interest) and text annotations on the views, that
can be shown or hidden as required,
- Using ROIs to select only part of a functional map for display.

ROIs and Text Annotations

You can add both ROI (Region Of Interest) graphic annotations and text annotations on
the views.
ROIs are used to perform measurements or to define a specific area to be displayed on
the function view, Text annotations on the images or on a graph can be used to draw
attention to a specific area of interest.
Both types of user annotations are managed in the same way. You can move, edit, copy
or delete annotations, and either show or hide them on the views.
Note: Automatically created artery and vein ROIs (see page 30) cannot be moved or
edited.

13
CT Perfusion

Recording
To document the results of an analysis, you can film or print the images, or save the con-
tents of any of the views (images or graphs) on the workstation disk in screen save for-
mat.
You can also save images or image sets as processed images, that can be re-loaded for
further processing with CT Perfusion 2 and in particular with 3D image processing appli-
cations such as Reformat and 3D Analysis.
To export the results to other (non-AW) systems and applications for reports or documen-
tation, you can save lists of numerical data from the graph view in text format, or save
view contents in TIFF format.
Filmed or saved images to be used for diagnostic purposes should always be correctly
annotated (system and user annotations).

14
 CT Perfusion

7 PIXELS AND VOXELS

The documentation and literature concerning image processing is not always fully consis-
tent in the use of the terms “pixel” and “voxel”. This paragraph attempts to clarify the use
of these terms in the context of this User Guide.
The definitions are simple enough:
- A “pixel” is a “picture element”, the basic element from which a two-
dimensional picture is constructed.
- A “voxel” is a “volume element”, the basic element from which a three-
dimensional volume is constructed.
However, an ambiguity occurs in CT image processing, because CT data sets represent
a 3D volume at one or more locations (slices). If we consider each image as a “picture”
we can say it is made up of “pixels”. But, strictly speaking, each image is a representa-
tion of a 3D volume (the slice), hence made up of “voxels”.
The basic elements that make up the display on the workstation monitor screen are also
referred to as “pixels”. The typical workstation color monitor display consists of 1280 by
1024 pixels, and when the viewing area is divided into four views, each view consists of
512 by 512 pixels.
• Within the context of CT Perfusion 2, we are dealing with image processing rather than
3D analysis. In this context, documentation and literature habitually refer to “pixel”
even where formally “voxel”might be more correct. For clarity, the User Guide adheres
to this convention.

15
CT Perfusion

Blank page.

16
 CT Perfusion

CHAPTER 3 - GRAPHS AND FUNCTIONS

CT Perfusion 2 has been developed for the analysis of CT data sets that contain multiple
images for each scan location in the data set. They are referred to as “ time series” or
“dynamic” data sets, such as acquisitions with a sequential change in scan time for each
scan image (as in contrast take up studies acquired using cine acquisition mode).
Such data sets contain for each scan location a series of N sequentially ordered images,
where the interval 1..N represent time.
The pixel value in an area of interest at a given scan location may have a different value
for EACH image. To analyze these changes in pixel value, CT Perfusion 2 provides two
tools :
• Graphs: the pixel values at a given pixel location p for the images 1..N are plotted as a
graph of pixel value over image number or time,
• Function images: for every pixel location p the pixel values from the images 1..N are
used to compute a characteristic parameter by means of a function. A functional map
(“ parametric image”) is then constructed by displaying the value of this parameter for
every pixel location.
A function can be either an algorithm, or it can have been derived from one of the basic
algorithms. A function usually requires one or more input parameters.
You use the CT Perfusion 2 protocols to set up the input parameters and then invoke the
appropriate algorithm(s).

17
CT Perfusion

1 OVERVIEW
CT Perfusion 2 has been developed for the analysis of CT data sets that contain multiple
images for each scan location in the data set, referred to as “time series” or “dynamic”
data sets.
Loading
When the data selected for analysis are loaded, CT Perfusion 2 sorts the data according
to scan location and sorts the multiple images for each scan location in order of time.
If the image set contains data for more than one scan location, CT Perfusion 2 initially
displays the first image at the central scan location in the set, but controls within the appli-
cation allow you to access and analyze all locations.

18
 CT Perfusion

Analysis
For each scan location we now have a series of N sequentially ordered images, where
the interval 1..N represents time.
We can now use the two main analysis tools available with CT Perfusion 2:

Series of N images Extract the intensity curve Extract a Set intensity of p

at the same location. of pixels p. characteristic parameter v in the function image to v.

N HU

F
2
1 f(v1, v2, ..., vN)=v p
p

1 2 N time
v
v1

1. Extract the pixel values at a given pixel location p for the images 1..N. These can
now be plotted as a graph of pixel value in Hounsfield units over image number or
time.
2. Extract the pixel values for the images 1..N, compute a characteristic parameter v
from these N pixel values by means of a function (algorithm), and construct a func-
tional map by displaying the value of the parameter v for every pixel location p of
the image. In other words, the graph of pixel values for the images 1..N for each
pixel location is “collapsed” into a single value, so that the result for all pixel locations
can now be displayed in a single image.
Note: For the purposes of recording and further analysis, graphs and images can be
filmed or printed (laser camera or color printer) and saved on disk.
With CT Perfusion 2 you can only display and analyze data for one scan location at a
time.
However, if the image set contains data for more than one scan location, you can save
the processed images for all locations as a new series with the same modality and image
parameters as the original images, and with the current function (algorithm) applied to all
images at all locations.
Such series can be further processed with other AW applications if required. In particular,
they can be displayed and analyzed using 3D image processing applications such as
Reformat and 3D Analysis.

19
CT Perfusion

Views and Controls

The CT Perfusion 2 screen display consists of a main control panel at the left, and a view-
ing area at the right. The viewing area contains four views:
• The series view (upper left)
displays the reference image,
• The graph view (upper right)
displays the image intensity
curves,
• The function views (lower
half) display the functional
maps,
• A function view can also dis-
play a composite image that
consists of the overlay of a
functional map on a reference
image.
Note: Only the two upper views
are displayed initially.

The controls (buttons) available in the main control panel are described in Chapter 5.
The on-view controls (i.e., the controls that you access and use directly on the views such
as the red active annotations and pop-up menus) are described in Chapter 6.

Series view
You use the series view to display the currently selected original image from the exam
series, to control and keep track of the image location (scan location and rank), to define
regions of interest (ROIs), and to examine the image data for possible patient motion
prior to analysis using the “cine” function.
Refer to section 2 for details.

Graphs
Graphs can be plotted for the pixel values at the current cursor position, or for the pixel
values within one or more regions of interest (ROI) defined on the views by the user.
Alternatively, CT Perfusion 2 can display the information in the form of a histogram or as
a table.
Refer to section 3 for details.

20
 CT Perfusion

Function
To perform an analysis, you select a protocol, use the controls to set up the input param-
eters and start computation. The quantities computed by the algorithm(s) in the protocol
are displayed as functional maps in the function views.
Refer to section 4 for details.
Composite view
To correlate the functional map and the original (reference) image, you can use a “com-
posite” view that displays an overlay of the functional map on the reference image.
Refer to section 5 for details.

21
CT Perfusion

2 SERIES VIEW
The series view (upper left) displays the reference image, i.e., the currently selected orig-
inal image from the exam series.

Initially the series view displays the first

image at the center location of the image
set.
To move to another image at the same
location:
• Use the rank active annotation or the
<left> and <right> cursor keys on the
keyboard,
OR
Move the vertical white bar on the graph
view to the desired location and click on
the middle mouse button.
To move to a different location:
• Use the scan location active annotation
or the <up> and <down> cursor keys on
the keyboard.
You can also display and use the {review
controller} to move through the image
set.

For proper results from the CT Perfusion 2 analysis tools, the image data for a given scan
location must be coherent, i.e., not degraded by patient motion during the acquisition.
ALWAYS use the series view “cine” function to examine the image data at the current
scan location for possible patient motion prior to analysis.

22
 CT Perfusion

3 GRAPHS
Graphs can be plotted for the pixel value at the current cursor position (“cursor ROI”
graph), or for the pixel values within one or more regions of interest (ROI) defined on the
views by the user.
The cursor itself can best be described as a small “dynamic” square ROI with a size that
can be set between 1x1 and 10x10 pixels (note that the cursor shape reflects the ROI
size). It plots the mean value of the pixels at the cursor position to the graph view (white
graph), and the graph is updated in real time whenever the cursor is moved.
Note: The cursor ROI graph is not displayed if another ROI has been selected.
For user-defined ROIs, you have a range of options: single-pixel ROI, freehand trace,
resizeable box, ellipse, polygon and spline, density mask, pixel list and mirror ROI.
An ROI can define four types of graphs: average value (AVG), minimum value (MIN),
maximum value (MAX) or standard deviation (DEV) of the value of the pixels within the
ROI.
If more than one ROI is defined, the corresponding graphs can be displayed either super-
imposed in a single frame, or individually as reduced-size graphs, each in their own
frame.
You can select from different units to scale the graph’s X and Y axes, such as image
number or time for the X-axis, and absolute, relative or percentage values for the Y-axis.
You can also scale all graphs to the same height (Y axis range) for comparison of the
curve shapes.
You can adjust the X and Y-axis scale factors to view a subset of the information if
required.

23
CT Perfusion

Graph View
The graph view (upper right) is used to display the image intensity information and statis-
tics of regions of interest. Three types of display of the image intensity information are
available: time graph, histogram and ROI list.

Time graph
The time graph displays the graphs of the
image intensity information corresponding to
the cursor ROI and user-defined ROIs (see
above).
The cursor time intensity curve represents
the changes in pixel value of the pixel(s)
under the cursor for every image in the data
set.
The ROI time intensity curves represent the
same information for the pixels within the
area of each ROI defined on the series view.
The cursor ROI graph is displayed in white,
the other ROI graph are annotated with the
corresponding ROI number.
An ROI can be selected (active), e.g. to
move or edit it. A selected ROI is displayed
in green, inactive ROIs are displayed in pur-
ple.
The vertical axis can represent pixel intensi-
ties in HU units, relative values or percent-
ages. The horizontal axis can represent
image number or time.

The moveable white bars can be used to identify the pixel intensity (vertical axis) and
image number or time (horizontal axis) at any point on the graph.
The scale of the horizontal or vertical axis can be modified by means of the red active
annotations. The auto red annotation, or the <Space> bar of the keyboard can be used
to scale the graph automatically.

24
 CT Perfusion

With more than one ROI, the correspond-

ing graphs can be displayed superimposed
in one graph (1x1 annotation), or as sepa-
rate reduced-size graphs.

Histogram
The histogram is a bar graph showing the
relative frequency with which each pixel
value occurs in a selected ROI.
The horizontal axis represents pixel value.
The vertical axis represents the number of
occurrences of each pixel value.
The scale of the horizontal or vertical axis
can be modified by means of the red active
annotations. The auto red annotation, or
the <Space> bar of the keyboard can be
used to scale the graph automatically.
Note: A histogram can only be displayed
if an ROI is present and selected on
the views.

25
CT Perfusion

List Values
The list shows the numerical values for
each ROI in a table
For documentation purposes this table can
be saved as a text file on the workstation
disk.

26
 CT Perfusion

4 FUNCTIONS
Overview
As noted in section 1, we have a series of N sequentially ordered images for each scan
location, where the interval 1..N represents time. The basic principle is illustrated again
below:

Series of N images Extract the intensity curve Extract a Set intensity of p

at the same location. of pixels p. characteristic parameter v in the function image to v.

N HU

F
2
1 f(v1, v2, ..., vN)=v p
p

1 2 N time
v
v1

1. We extract the pixel values for the images 1..N,

2. We use a function to compute a characteristic parameter v from these N pixel val-
ues,
3. We construct a functional map by displaying the value of the parameter v for every
pixel location p of the image.
In other words, the graph of pixel values for the images 1..N for each pixel location is “col-
lapsed” into a single value, so that the result for all pixel locations can now be displayed
as a single image, referred to as the functional map.
In CT Perfusion 2, a function can be either a basic algorithm, or a function derived from
one of these algorithms.
To create the functional maps, you select a protocol that contains the appropriate func-
tion(s).
The algorithm resolution determines how many pixels from the reference image are used
for computing the functional maps. As an example, a resolution setting of 2 will compute
only every second line and every second column (one pixel in four). Using a lower reso-
lution speeds up the computation.
The image data are smoothed before computation by means of the algorithm. This
reduces noise in the original image data, and the resulting functional map will be corre-
spondingly smoother. The spatial smoothing factor determines the strength of the gauss-
ian filter function, or spatial averaging, applied to the image data.
The protocols use default values for resolution and spatial smoothing factor, but they can
be adjusted by the user.

27
CT Perfusion

The functional maps are displayed in the function views (lower half of the view area), typ-
ically in color
The function views are not displayed until you have set up the input parameters and
selected (Compute) in the protocol.
The available functions (algorithms and derived functions) are briefly listed below. They
are described more fully, with their mathematical background and input parameters, in
Appendix 1 “Functions”.

Function View

The function views (lower half of the view

area) display the functional maps created
using the algorithm(s) and/or derived func-
tions in the selected protocol.
The maps are typically in color and display
the result of the computations. They are
not displayed until you have set up the
input parameters and selected (Compute)
in the protocol.
Various color ramps are available. The
color ramp defines how the pixel values
are mapped to the colors in the functional
map.
ROIs other than the “artery” and “vein”
ROIs used by the CT perfusion algorithm
have no effect on the calculation of the
functional maps.
However, you can use ROIs to eliminate
part of a functional map from the display,
either inside or outside the ROIs, e.g., to
show only a specific feature, or to elimi-
nate noise from the display.

You control the range of pixel values that is displayed in the function view (window W/L)
either by adjusting the H and L active annotations of the color ramp on the view, or by
using the same window W/L controls as for the other views.
Pixels displayed in black (pixel value = 0) indicate that the function did not return a signif-
icant value for the location in question. This can be for several reasons such as: location
outside the anatomy, use of a threshold (values below the threshold are set to zero), fail-
ure of the function to fit a curve to the data at the location because of excessive noise,
etc. Also see Appendix 1 “Functions”.

28
 CT Perfusion

Functions
This paragraph briefly lists the available functions (algorithms and derived functions).
They are described more fully, including their theoretical background and input parame-
ters, in Appendix 1 “Functions”.
The descriptions and explanations of the functions below apply to their use with time
course data. While some of these functions can be invoked with other data types
accepted by CT Perfusion 2, the results are unlikely to have any physical significance.

CT Perfusion algorithm
This algorithm is specifically intended for analysis of CT blood perfusion scan data follow-
ing the injection of a contrast agent.
The algorithm uses the data from a reference arterial ROI to deconvolve the time curve.
This means that the algorithm takes into account the actual injection rate of the contrast
agent.
The CT perfusion algorithm is used to compute absolute values for Blood Volume (BV),
Mean Transit Time (MTT), Blood Flow (BF) and Permeability Surface area product (PS).
Computed results:
Blood Flow is computed and displayed in ml per 100g of wet tissue per minute,
Mean Transit Time is computed and displayed in seconds.
It characterizes the temporal position of the transient increase in the time course data.
Transient increases that occur later have a larger transit time.
The notion of a Mean Transit Time assumes that the injection of the contrast agent is
instantaneous, which of course can never be the case in practice. The algorithm take
into account the injection rate by using the data from a reference ROI located in an artery,
to obtain a quantitative estimate of MTT and BF.
Blood Volume is computed and displayed in ml per 100g of wet tissue.
It is defined as the product of blood flow and mean transit time.
Permeability Surface Area Product is computed and displayed in ml per 100g of wet
tissue per minute.
It is used in tumor studies to characterize the retention of some of the contrast agent in
the tissues due to brain-blood-barrier (BBB) leakage or metastases.
The computed values for blood flow, blood volume and permeability surface area product
are normalized relative to the pixel value for 100% blood in a large vessel (vein) using a
reference ROI and relative to the average tissue density to obtain values expressed per
unit mass.

29
CT Perfusion

Other algorithms
Positive Enhancement Integral
Time course data acquired during the injection of a contrast agent may have image inten-
sity variations, which result in positive enhancement. The positive enhancement integral
function calculates the area underneath the time intensity curve within a user-specified
range.
Time To Peak
Time-to-peak (in seconds) is the time between the onset of the enhancement transient
and the peak value of the time curve. It is computed directly from the time course data by
the algorithm.
Maximum Slope of Increase
For time course data that exhibit a transient increase, the Maximum Slope of Increase
algorithm will return a value that characterizes the maximum slope of increase. Transient
increases that occur rapidly will have larger maximum slope of increase values and those
occurring gradually will have smaller maximum slope of increase values.
Maximum Slope of Decrease
By analogy, the Maximum Slope of Decrease algorithm will return a value that character-
izes the maximum slope of decrease of the time course data.

30
 CT Perfusion

5 COMPOSITE VIEW

A composite view is a function view that

displays the overlay of the functional map
onto a reference image.
This allows you to correlate the functional
map with the reference image.
In the composite view the color scale and
pixel intensities are taken from the func-
tional map except when the pixels are
black. The remaining pixels in the com-
posite view are the gray scale and pixel
intensity of the reference image.
To facilitate the correlation, the functional
map overlay can be displayed as transpar-
ent to a greater or lesser degree. At 0%
transparency the overlay is fully opaque, at
100% it is fully transparent, i.e., no longer
visible.
Initially, the reference image used in the
composite view is the same image that is
displayed in the series view.

Certain exams may contain a separate anatomical reference image for the same location.
Usually, this is a separate high resolution image for the same location as the lower reso-
lution time course images.
If the exam contains such a separate anatomical reference image you can use this image
as a reference in the composite view.

31
CT Perfusion

Blank page

32
 CT Perfusion

CHAPTER 4 - PROCEDURES

A CT exam to be processed with CT Perfusion 2 has to meet certain basic requirements.

You select the Image Works Brower then start CT Perfusion 2 and select a protocol.
This chapter describes the image requirements, the startup procedure, and the contents
and use of the CT Persion 2 protocols.

33
CT Perfusion

1 IMAGE REQUIRMENTS
CT Perfusion 2 accepts CT data sets that contain multiple images for each scan location
in the data set, representing changes in image intensity over time (“time series” or
“dynamic” data sets).
A CT image set to be used with CT Perfusion 2 should meet at least the following require-
ments:
• Field of view, matrix size and display center must be the same for all images in the set,
• Orientation and gantry tilt should be the same for all images in the set,
• Tilted acquisitions are not supported for right/left decubitus patient orientations,
• The set should include only AXIAL, SAGITTAL, CORONAL or OBLIQUE images.
Other types such as screen saves, etc. are not supported,
Note: When loading an image set, the CT Perfusion 2 software uses the FIRST selected
image in the Browser/Patient List as a basis for using/discarding the other images.
For example, any images having a matrix size or gantry tilt different from that of
the first selected image in the Browser/Patient List are discarded.
Note: Functional maps generated by CT Perfusion 2 that have been saved as processed
images can be re-loaded, displayed and further processed by means of the CT
Perfusion 2 Import protocol.
Note: Under certain conditions, CT Perfusion 2 may accept to load CT data sets that are
not time series. However, the results from attempting to process such a data set
with CT Perfusion 2 will be meaningless.
For use with CT Perfusion 2 the following requirements should also be taken into
account:
- To obtain satisfactory results from the CT Perfusion 2 analysis tools, it is
important that the image data are coherent, i.e., not degraded by patient
motion during the acquisition. The review functions in the protocols allow you
to check the image data for patient motion. The brain perfusion protocols also
contain a function allowing you to remove the effects of patient motion
automatically (registration).

Always check the image data for the presence of patient

motion before analysis.
CAUTION It remains the responsibility of the user to determine
whether the amount of patient motion in the exam is
acceptable for the purpose of the analysis.

34
 CT Perfusion

- The quality and accuracy of the results for blood flow, mean transit time and
permeability surface area product from the blood perfusion protocols in CT
Perfusion 2 is critically dependent on the time resolution (interval between
images) of the data set. A time resolution of 1 second is generally adequate.
Accuracy will degrade rapidly for larger time resolution values. This applies in
particular to the first minute of the data. For long acquisitions (several
minutes, as used in tumor studies) the time interval may increase towards the
end of the data set.

It remains the responsibility of the user to determine

CAUTION whether the time resolution in the exam is acceptable
for the purpose of the analysis.

35
CT Perfusion

2 START CT PERFUSION 2
To start using CT Perfusion 2:
• To start CT Perfusion 2 from the Image Works Browser.

Restart Browser

Restart Display

Image File Search

Application Selection Remove Sort Network Archive PPS Queue Utilities Messages
Examinations : Exam #47, May 05 92, SMITH, JON
Exam Name Date Description Mod Fmt A Ser Type Imgs Description Mod VA

3145 J.Herman Jan 08 98 CT 1 PROSP 18 CT Add/Sub

3512 B.Fox Dec 23 97 CT

Edit Patient

Film

Mini Viewer
2 examinations one series
Series
Img Img Ctr Thck/SP Gntry Img Ctr Img Ctr SFOV DFOV Perfusion2
(deg) Res Matrix Midscn Archive
S–I (mm) (mm) R–L A–P (cm) (cm)
Viewer
1 S 50.0 10.0/ +0.0 R5.0 A0.0 43 Lung 512 No
2 S 45.0 10.0/ +0.0 R5.0 A0.0 43 Lung 512 No
3 S 40.0 10.0/ +0.0 R5.0 A0.0 43 Lung 512 No
4 S 35.0 10.0/ +0.0 R5.0 A0.0 43 Lung 512 No
5 S 30.0 10.0/ +0.0 R5.0 A0.0 43 Lung 512 No
6 S 25.5 10.0/ +0.0 R5.0 A0.0 43 Lung 512 No

60 images

• Then click Perfusion 2

36
 CT Perfusion

3 PROTOCOLS
After starting the application, the Select a Protocol panel is displayed. The panel con-
tains icons for the CT Perfusion 2 protocols available to process the selected exam and
series. When you move the mouse cursor onto an icon, a help pop-up will show the func-
tion of the protocol.
• To select and start a protocol and load the exam, click on the corresponding icon.
CT Perfusion 2 is supplied with several protocols, each optimized for generating quantita-
tive functional maps for a specific type of analysis:
- The Brain Stroke protocol generates maps for regional blood volume,
regional blood flow and mean transit time.
- The Brain Tumor and Body Tumor protocols generate an additional map for
permeability surface (used in examinations of tumors).
- The CT Standard protocol generates maps for various other perfusion-related
parameters: positive enhancement integral, time to peak, maximum slope of
increase and maximum slope of decrease.
After starting the application and selecting a protocol, each protocol displays a succes-
sion of protocol panels that contain the necessary information and controls for each step
of the analysis.
The protocol panels and their use are described below.
Overall, these protocols use the same sequence, but not all panels and controls are
present in all protocols. Differences are described below, and will also be mostly self-evi-
dent during the use of the protocols.
A separate Import protocol allows you to use CT Perfusion 2 to re-load and display func-
tional maps that were saved earlier as processed images.

37
CT Perfusion

Help
Several protocol panels contain a (Help) button, to display additional information about
the current function.
• To display the information, click on the (Help) button in the protocol panel.
• Use the {scroll bar} on the right of the panel to move through the contents of the Help
panel.
• Use the (Back|Forward) buttons if more than one panel is available.
• To remove the panel from view, click on the (Close) button.
You can access a certain number of CT Perfusion 2 functions by means of the worksta-
tion keyboard.
The <Help> key, or the <Ctrl-H> key combination, allows you at all times to view a panel
with the full list of available keyboard commands, without having to refer to the manual:
• To display the list, press the <Help> key or the <Ctrl-H> key combination.
• Use the {scroll bar} on the right of the panel to move through the list.
• To remove the panel from view, click on the (Close) button.

38
 CT Perfusion

Image Review
Protocols: all
To obtain satisfactory results from the CT Perfusion 2 analysis tools, it is important that
the image data are coherent, i.e., not degraded by patient motion during the acquisition.
The most effective way to detect patient motion is by displaying the images in rapid suc-
cession by means of a “cine loop”.
In the Brain Stroke and Brain Tumor protocols, an Image Registration function is avail-
able that can remove patient motion in the slice plane automatically by registering (align-
ing) the images.
Procedure

• Display the {review controller} along the edges of the series

view, by clicking on the corresponding button in the main con-
trol panel. For a full description of this tool, see page 117.
• Use the (loop) or (rock) button and the {jog shuttle} to dis-
play a cine loop. The loop will first run slowly, while the
images are being loaded, then at a speed controlled with the
shuttle. Check for patient motion. Note the image(s) or image
range.

• Use the horizontal {time/rank slider} to view the images individually and check for
anomalies (such as acquisition artifacts).
• With the Brain Stroke and Brain Tumor protocols, you can use the Image Registration
function to remove the motion. With other protocols, make sure that the results of the
processing are not adversely affected by patient motion in the image set.

39
CT Perfusion

Image Registration
Protocols: Brain Stroke, Brain Tumor

The Image Registration function automatically

removes patient motion in brain CT image sets by
registering (aligning) the images.
As noted above, the presence of patient motion
can seriously degrade the results of an analysis.
You may prefer to use registration systematically
on brain CT image sets since it does not take
much time (less than a minute on a typical set of
50 images).

Procedure
• Click on the (Apply Registration) button to start registration. The process takes a few
moments to complete, as indicated by the {progress bar} at the lower left of the
screen.
• After performing registration, again use the cine loop to check that all motion has been
eliminated from the images. If residual motion is still visible, repeat the registration.

40
 CT Perfusion

Registration Settings

If required, you can modify the parameters used by the software to perform registration.

• Click on the (Settings) button to display the

Registration Settings panel.
• Use the Low Threshold and High Threshold
sliders to adjust the range of pixel values (rep-
resenting bone) taken into account to perform
the registration.
• Use the Control Points Count slider to adjust
the number of points used.
• You can click on (Show Points) to display the
actual control points. With the right settings for
the parameters, these control points should out-
line the boundaries of a bone region that will
result in satisfactory registration.

Note: Whenever you modify the settings, always select (Apply Registration) again to
apply the new seettings.

41
CT Perfusion

Processing Thresholds

Protocols: all

The low and high threshold sliders allow you to

define the range of HU values (pixel values) that
will be processed by the protocol (shown by green
hatching on the top left view).
All pixels with HU values outside the defined
range will be masked out, i.e., they will not be pro-
cessed and will appear black in the function views.
This has a dual purpose:
- Non-relevant image data such as air
spaces (including the empty space
outside the patient) and bone are
excluded from the processing. This
can amount to as much as half of the
data, and the processing time is
reduced accordingly.
- Noise resulting from such data (in
particular outside the patient) will not
appear on the functional maps, making
for easier interpretation of the resulting
image.

Procedure
• On the series view, display the first image of the series (or at least one of the pre-
enhancement images).
• Adjust the {left slider} to set the low (“air”) threshold. The default setting may at times
exclude some of the soft tissue. Adjust as necessary.
• Adjust the {right slider} to set the high (“bone”) threshold. The default setting may at
times exclude part of contrast-enhanced blood vessels. Adjust as necessary.
• The buttons on either side of the sliders allow you to reset the sliders to the default val-
ues.

42
 CT Perfusion

Artery Input

Protocols: Brain Stroke, Brain Tumor, Body Tumor

When computing Mean Time of Transit, Blood

Flow and Permeability Surface, the protocol uses
a reference ROI inside an artery to determine the
Artery Input Function (AIF), to account for the
actual injection rate of the contrast agent. This
reference ROI can be determined automatically or
manually.
In automatic mode, the software defines the artery
ROI by selecting among all pixels with a sufficient
enhancement (all vessels) the region of pixels with
the smallest peak time, i.e., the point where the
contrast agent arrives first.
In manual mode, the user identifies a suitable
artery and defines an ROI inside it.
See Chapter 8 for details on how to define ROIs.

Automatic Procedure
• If the approximate location of the artery where the contrast agent arrives first is known,
define an ROI for the corresponding area and select it: the search for the artery will be
limited to the zone inside this ROI. If no ROI is selected, the software will search the
entire image.
• Click on the (Automatic) button. The software defines the artery ROI and displays it
on the series view. If a “search” ROI was defined beforehand, it is deleted automati-
cally.
• Check that the ROI is placed satisfactorily, then select (Next).

43
CT Perfusion

Manual Procedure
• On the series view, identify an artery and define an ROI inside it . Check the corre-
sponding time-intensity curve on the graph view. The size of the ROI should be in the
order of 2 to 6 pixels. In particular, it is important that the entire ROI is located inside
the artery.
• With the ROI selected (displayed in green), click on the (Manual) button to define the
ROI that will be used to determine the Artery Input Function.
• Select (Next).
Note: For best results in automatic mode, use an ROI to limit the search zone, if the
approximate location of the artery where the contrast agent arrives first is already
known. If the results in automatic mode are not satisfactory, you can switch to the
Enhancement Image Range panel (see page 71), provisionally set the enhance-
ment image range to the approximate range of the first intensity transient, then
return to the Artery Input panel and repeat the procedure. In this manner only
this enhancement image range is taken into account in the search for the artery.
Note: In automatic mode, the reference ROI is identified by an “Artery” text annotation,
to help identify it if several ROIs are already present on the views. If you place the
reference ROIs manually, it may be advisable to add the corresponding text anno-
tation for the same reason.
Important: To compute the functional maps, the software uses the reference information
from the artery ROI created or designated by the last click on the (Automatic) or
(Manual) button. When you modify or delete such an ROI, always be sure to re-
define the artery ROI to be used by the software by means of the (Automatic) or
(Manual) button in the Artery Input panel. Not doing so may lead to erroneous
results. The same remark applies to the vein ROI.

44
 CT Perfusion

Vein Output

Protocols: Brain Stroke, Brain Tumor, Body Tumor

When computing Blood Volume and Blood Flow,

the protocol uses a reference ROI inside a large
vessel (vein) in order to normalize the results rela-
tive to the HU value for 100% blood. This refer-
ence ROI can be determined automatically or
manually.
In automatic mode, the software uses the region
of pixels with the largest blood volume (which is
usually located inside a vein) as a reference ROI.
In manual mode, the user identifies a suitably
large vessel and defines an ROI inside it.
See Chapter 8 for details on how to define ROIs.

Automatic Procedure
• If the approximate location of the vessel with the largest blood volume is known, define
an ROI for the corresponding area and select it: the search for the vein will be limited to
the zone inside this ROI. If no ROI is selected, the software will search the entire
image.
• Click on the (Automatic) button. The software defines the vein ROI and displays it on
the series view. If a “search” ROI was defined beforehand, it is deleted automatically.
• Check that the ROI is placed satisfactorily, then select (Next).

45
CT Perfusion

Manual Procedure
• On the series view, identify a suitable large vessel (usually a vein) and define an ROI
inside it . The size of the ROI should be in the order of 2 to 6 pixels. In particular, it is
important that the ROI is located inside the vessel. The corresponding time-intnesity
curve is displayed on the graph view.
• With the ROI selected (displayed in green), click on the (Manual) button to define the
ROI that will be used to determine the reference HU value for blood.
• Select (Next).
Note: Since the largest vessel is usually a vein, it is referred to as the “vein ROI”. How-
ever, any ROI located in a vessel large enough to be fully representative of the HU
value for 100% blood and not affected by partial volume effects, is acceptable
(even if it happens to be located in an artery).
Note: In automatic mode, the reference ROI is identified by a “Vein” text annotation, to
help identify it if several ROIs are already present on the views. If you place the
reference ROIs manually, it may be advisable to add the corresponding text anno-
tation for the same reason.
Important: To compute the functional maps, the software uses the reference information
from the vein ROI created or designated by the last click on the (Automatic) or
(Manual) button. When you modify or delete such an ROI, always be sure to re-
define the vein ROI to be used by the software by means of the (Automatic) or
(Manual) button in the Vein Output panel. Not doing so may lead to erroneous
results. The same remark applies to the artery ROI.

46
 CT Perfusion

Automatic Reference ROIs

The software can compute the reference artery and/or vein ROI(s) automatically as
described above. They are displayed on the views, so that you can check immediately if
they are placed satisfactorily.

Using automatically generated reference ROIs can be faster

NOTICE and more accurate than placing the ROI(s) manually, but you
should always check that the automatically generated refer-
ence ROI(s) are placed correctly.

Automatically generated reference ROIs are unlike other ROIs in that they cannot be
moved or edited, and in that the linked “Artery” or “Vein” text annotation is part of the ROI.
If an automatically generated artery or vein ROI is not satisfactory:
• Return to the Artery Input or Vein Output panel. The corresponding reference ROI is
selected and displayed in green (unless it has already been deleted).
• Delete the ROI using the (cut) button in the control panel, the <Cut> key on the key-
board, or the <Ctrl-X> key combination.
• Re-define the ROI either automatically or manually as described above.
Automatically generated reference ROIs are identified by an “Artery” or “Vein” text anno-
tation on the views. This annotation is part of the ROI. If for any reason you want to mod-
ify or remove the annotation:

• Select the ROI, then “ungroup” it by displaying and selecting

the (ungroup ROI) button (see page 169). You can now sep-
arately edit or delete the text annotation as required.

Note: You can move the annotation of a reference ROI to a different location on the view
by selecting the ROI and dragging on the annotation, without needing to ungroup
the ROI.

47
CT Perfusion

Defining Artery and Vein ROIs at a Different Image Location

Usually, you will define the reference artery and vein ROIs at the image location you are
analyzing.
However, if you are working with an exam with more than one location (i.e., more than
one slice), you may find at times that you can define one or both reference ROIs more
clearly on a slice different from the one you are using for the analysis. To do so:
• Move to the location (slice) you want to use for the reference ROI(s). You can use the
active annotation, {review controller} or <up> and <down> keyboard keys.
• Define the ROI(s) using the controls in the Artery Input and Vein Output panels in the
same manner as described above.
• Return to the location (slice) to be analyzed.
The function will now use the reference ROI(s) that you have defined, even though they
are not displayed on the current slice or in the graph view.
Note: The last panel of the protocols will show the numbers of the ROIs used by the pro-
tocol as the reference artery and vein ROIs, allowing you to check that the refer-
ence ROIs are defined correctly.

48
 CT Perfusion

Enhancement Image Range

Protocols: all
By default the full range of images in the data set will be processed. In some circum-
stances, you may want to limit the range of images to be processed.
Use the sliders on the panel together with the curves on the graph view to set the pre and
post enhancement image ranges.
Protocols: Brain Stroke, Brain Tumor, Body Tumor

Set the pre and post-enhancement image number

sliders to define the range of images that will be
processed.
The pre-enhancement image range (from first
image to last pre-enhancement image) is used by
the software to define a baseline.
The setting of the post-enhancement image num-
ber determines the end of the image range that
will be processed.
Procedure
• Set the top slider (“last pre-enhancement
image”) to the number of the last image before
arrival of the bolus of contrast agent. This cor-
responds to the point of the artery ROI curve on
the graph view just before the onset of the tran-
sient.

• For stroke studies, it is preferable to limit processing (integration) to the image range
that contains the first transient.
Set the bottom slider (“first post-enhancement image”) to the number of one of the
first images after the first pass of the bolus of contrast agent, before the onset of
recirculation.
The post-enhancement image range (between first post-enhancement image and last
image) will be excluded from the processing.
• For tumor studies, it is preferable to include as much of the post-enhancement image
range as possible in the processing.
Leave the bottom slider (“last post-enhancement image”) at the default value (last
image), unless you need to limit processing time. This should only be necessary in
the case of long-duration acquisitions (several minutes).
The images between the last post-enhancement image and the last image will not be
processed.

49
CT Perfusion

Protocol: CT Standard

The function of this panel in the CT Standard pro-

tocol is the same as that for the other protocols:
except that two separate sliders are available to
define the pre-enhancement image range.
This is used to limit the pre-enhancement range
of images to those just before the onset of the first
transient, and eliminate any anomalies that may
be apparent in the first few images of the data set.
Procedure
• Create one or more ROIs at significant loca-
tions on the series view to determine the evolu-
tion of the intensity transient.

• Set the “first pre-enhancement image” slider to the number of the first image of the pre-
enhancement image range to be taken into account for processing. If no anomalies
are present, you can leave this slider at the default setting (first image).
• Set the “last pre-enhancement image” slider to the number of the last image before
arrival of the bolus of contrast agent.
• Leave the “post-enhancement image” slider at the default value (last image), unless
you need to limit processing time (in the case of long-duration acquisitions) or if you
specifically want to exclude part of the images from the processing (e.g., to process
only the first transient).

50
 CT Perfusion

Final Settings

Protocols: Brain Stroke, Brain Tumor, Body Tumor

The Final Settings panel lists the current settings

of the input parameters for the protocol.
In particular, it shows the numbers of the ROIs
used by the protocol as the reference artery and
vein ROIs. This can be useful if you have defined
these ROIs manually, or if any of the ROIs is not
at the same image location.
The panel also allows you to access certain
advanced settings.

Protocol: CT Standard
The Final Settings panel lists the current settings of the input parameters for the protocol.
The protocol does not use any advanced settings.

51
CT Perfusion

Advanced Settings
Protocols: Brain Stroke, Brain Tumor, Body Tumor
The protocols use default settings for certain parameters, such as maximum blood flow
value, hematocrit ratio, tissue density, resolution and degree of spatial smoothing.
For specific cases, you may want to adjust one or more of these parameters. E.g., the
hematocrit ratio may be higher for an exam of an infant.

For a detailed description of these parameters, refer to Appendix 1 “Functions”.

• To adjust any of the parameters, click on

(Advanced Settings), then click on the [Blood
Flow], [Hematocrit], [Tissue] or [Display]
“tab” to display the individual panels (“Display”
panel illustrated).
• Set the options as required.
Some parameters will show a type-in field to
enter custom values. Move the mouse pointer
inside the field, delete the existing value with
the <BackSpace> or <Del> key, type in the
new value and confirm with <Enter>.
• Move between the individual panels by means
of the tabs. When all parameters as set as
required, click on the (Done) button in any of
the panels to return to the Final Settings panel.
(The (Back) button in these panels takes you
back to the last-but-one protocol panel, the
(Close) button closes the entire protocol.)

52
 CT Perfusion

Compute
Once you have set all parameters to your satisfaction:
• Click on (Compute) in the final panel to start computing the functional maps.
This may take a few moments because all functional maps for the current location are
computed at the same time. This is indicated by the {progress bar} at the bottom of the
main control panel. You can use the (Stop) button at the left side of the {progress bar} to
interrupt the computation.
When computation is complete and the functional maps are displayed, you can remove
the protocol panel from view (without losing the settings) by clicking on the (Close) but-
ton. This allows you to use both function views to display the functional maps.
You can now use the tools available in the control panel, in the on-view menus and on the
views to review and analyze the results of the computation. See Chapter 5 “Control
Panel” and Chapter 6 “On-View Controls”.
To modify any of the input parameters, you can recall the protocol.
• Click on the button with the name of the protocol in the main control panel. The proto-
col panels are again displayed.
• Modify the parameters as required, then click again on (Compute) in the final panel to
re-compute the functional maps.

Always select (Compute) again to re-compute the func-

CAUTION tional maps after making changes to the input parameters.
The changes are not taken into account automatically.

To use a different protocol with the same image set:

• Click on the (New Protocol) button in the main control panel, and select the protocol in
the Select a Protocol panel.

53
CT Perfusion

4 IMAGE IMPORT PROTOCOL

Overview
The Import protocol differs from the other CT Perfusion 2 protocols in that it is not used
for the analysis of CT image data sets, but to re-load and display functional maps that
have been saved as processed images.
The protocol is shown in the Select a protocol panel only when one or more series of
saved functional maps have been selected in the Browser/Patient List.
A saved functional map can be identified in the Browser/Patient List by its name (e.g.,
Blood Volume) in the Description column of the series list. The series type can be PRO-
CESS or REFORMAT, depending on how the map was saved.
To use the Import protocol:
• Select the series that contains the desired image(s) in the Browser/Patient List. You
can select more than one series if required.
The series should contain a functional map saved as a processed image. Other
image types, including functional maps saved as screen saves, are not accepted.
• Start the CT Perfusion 2 application.
A message will alert you that the selected series is not a time series. Click on (OK) to
acknowledge the message.
• In the Select a protocol panel, select the Import protocol.
The images are loaded and displayed in the series view and in the function views. The
protocol panels are displayed superimposed on the bottom left view. These panels show
image data and parameters of the loaded images; they can be closed and recalled as
necessary (see below).
The series view displays one image per location, using gray levels. The function view(s)
display the image(s) in color with the appropriate color ramp.
Once loaded, you can explore and manipulate the functional maps using the same con-
trols and tools as during a normal analysis (see Chapter 5 and Chapter 6). In particular:
• If more than one functional map has been loaded, you can select which maps to dis-
play using the Import: (function name) active annotation (near top left on the function
views).
• You can select color ramps from the on-view menu, and adjust H/L (or W/L) as
required.
• You can create ROIs on the views for measurements.
• You can display composite views using a reference image selected in the original
acquisition series.
Note: You cannot use the graph view because the original series data are not loaded.

54
 CT Perfusion

Protocol Panels

The import protocol comprises two panels.

The first panel (“Import Functional Map”) shows a

list of the functional maps that have been loaded.
Select a functional map in this list to display the
corresponding Group name, Function name, Bias
value and Scale value.
The Group Name refers to the type of function
used to create the functional map (e.g., CT Perfu-
sion), the Function Name indicates the actual
function used (e.g., Blood Volume).
Bias and Scale refer to the bias (offset) and scale
factor that were used to “fit” the numerical values
returned by the function into the available range
of pixel values (integers from 0 to 4095), when the
functional map was saved. This process is used
to avoid loss of information whenever the values
returned by the function would have been too
small to be meaningful when simply rounded off to
integer values.
Also see Appendix 1, section 5 “Pixel Values in
Saved Images”.

55
CT Perfusion

The second panel shows the parameters that

were used to compute the selected functional
map.

You can close the panels at any time with the (Close) button, and recall them using the
(Import) protocol button in the main control panel.

56
 CT Perfusion

CHAPTER 5 - CONTROL PANEL

You access the CT Perfusion 2 control through:
The control panel (displayed at the left of the CT Perfusion 2 screen), which is
described in this chapter.
On-view controls,
Active annotaitons (displayed in red on the views).
These are system annotations that can be modified by the user, e.g., to
change window width or level, or to move to a different location in the exam.
On-view menus,
The review controller, to move throug the images using sliders and buttons or by
means of a cine loop. These controls are described in Chapter 6.
Also, the <Help> button on the keyboard, or the <Ctrl-H> key combination, allow
you at all times to view the full list of available keyborad commands, without having
to refer to the manual.

57
CT Perfusion

1 OVERVIEW
The CT Perfusion 2 screen display consists of a control panel at the left, and a viewing
area at the right that contains four views.

• The series view (upper left) dis-

plays the reference image,
• The graph view (upper right) dis-
plays the image intensity curves,
• The two function views (lower
left and right) display the func-
tional maps (these views are
empty until the functional maps
have been computed).
The protocol panels used to set up
an analysis are normally displayed
on top of the lower left view, but
they can be moved or closed tem-
porarily if required.
The controls in the control panel
are listed in section 2, and
described in sections 3 and 4.

Each view also has its on-view controls, i.e., controls that you use directly on the views
such as active annotations (displayed in red) and on-view (pop-up) menus.
See Chapter 6.

58
 CT Perfusion

2 CONTROL PANEL

The CT Perfusion 2 control panel is displayed at the left of the

1
screen.
It contains:
1. A header showing the software version and the current
patient name,
2
2. The Task Bar (see section 3),
3
3. The Main Control Panel (see section 4),
4. A {progress bar} showing the progress of computing
tasks that take more than a few moments to complete.
The controls in this panel are listed below, and described in
the following sections in the order they appear in the panel.
4 Certain functions are also described in more detail in other
chapters, where relevant.

3 TASK BAR
The controls in the Task Bar allow you to return to the Browser when required, and to quit
the application.
To return to the Browser without losing work in progress, e.g., to check information in the
lists or to review saved images:

• Click on the (Browser) button

• If required, you can now also select and use certain other viewing applica-
tions (e.g., Viewer and Mini Viewer).
• Return to the CT Perfusion 2 application by re-selecting CT Perfusion 2 in
the Browser (as when starting the application,).
This returns you to CT Perfusion 2 in the same state as you left it.

59
CT Perfusion

To quit the CT Perfusion 2 application and return to the Browser, either to start process-
ing a new exam or to start another application:

• Click on the (Close) button.

A message will ask: “Do you really want to exit CT Perfusion 2?”.
• Select (Yes) to confirm, (No) to cancel.

Note: Any work in progress (images or graphs) that you have not saved previously will
be lost.

4 MAIN CONTROL PANEL

The main control panel contains the following categories of

controls:
1
1. Protocol buttons,
2. Central panel,
3. Menu buttons, to open the Pref./Settings and Film/
Save menus,

60
 CT Perfusion

Protocols

Two Protocol buttons are located at the top of the Main Control
Panel.
• Use the (New Protocol) button to return to the Select a Protocol
panel and select a different protocol to analyze the currently
loaded exam series.
The second Protocol button shows the name of the currently
selected protocol.
• Use this button to recall the panels of the protocol, if you have
temporarily removed them from view.

Graph View Mode

The graph view (upper right) can display:
- Image intensity curves,
- Histogram,

- Image intensity data in list format.

• Use the graph view mode buttons to select (Graph), (Histo-

gram) or (ROI List).

Note: You can also set the graph view mode from the graph on-view menu.

61
CT Perfusion

Central Panel

The buttons in the upper half of

this panel are used for various
view control and manipulation
functions.
The buttons in the lower half are
used for the creation and manage-
ment of user annotations (text and
ROIs).
The panel does not display all
available buttons.
Mostly this concerns buttons that
are associated with another button
that is already displayed: e.g., a
(rectangle ROI) button is associ-
ated with the (ellipse ROI) button.
• To use the associated buttons,
first display them by clicking on
the narrow arrow buttons to the
left or right edge of the dis-
played buttons.

View Control and Mainpulation

These functions allow you to:

- Modify the functions of the left and middle mouse buttons
to scroll and zoom the images or adjust W/L,
- Flip and rotate the images,
- Return the images to the default settings,
- Create a grid on the views,
- Switch between smoothed and unsmoothed display of
the images.

62
 CT Perfusion

Mouse Functions

You can modify the function of the left and middle mouse buttons.

This is the default mode (cross-hair pointer). Use this mode for all
Selection the standard functions as described elsewhere in this manual: dis-
play of the image intensity curve at the cursor position, ROI selec-
tion, adjustment of active annotations, etc.

In this mode, you use the left mouse button for selection. The middle mouse button con-
trols window width and level. Clicking the right mouse button on a view or on an active
annotation displays the corresponding on-view menus.
Also, clicking on this button when another mode is selected will deselect that mode and
return you to selection mode.

Roam (Scroll)

To move the images around within the viewports , e.g., to re-center a

particular feature in the views after enlarging the images (“zoom”), select
this mode, then click and drag with the left mouse button on a view.

When you move one of the images (reference image or functional map), the other images
also move automatically as soon as you release the mouse button.
You can also use the image orientation active annotations on the sides of the images.

Window Width/Level

To adjust window width and level with the left mouse button, select this
mode, then click and drag with the left mouse button as you would nor-
mally with the middle mouse button in selection mode.

63
CT Perfusion

Zoom

Continuous zoom: to zoom in or out on the images (i.e.,

change the field of view) in a gradual fashion, select this
mode, then click and drag to the left or right with the left
mouse button on a view.
Zoom in steps: to zoom in or out on the images in steps,
display and use the (step zoom) buttons to select zoom
factors of 1, 1.4, 2 ... until 8 relative to the original DFOV.
The (+) button increases the zoom by a factor /2 at each
click, the (-) button reduces it by the same factor.

When zooming in or out on one of the images (reference image or functional map), the
other images are automatically re-sized accordingly. The maximum zoom factor is 8.
You cannot zoom out beyond the original size (DFOV) of the image.
You can also use the DFOV active annotation to change the field of view.

Review controller

To show/hide the {review controller} (“image reviewer”) on the sides of

the series view, use this button (toggle). For a description of the {review
controller}.

64
 CT Perfusion

Image Orientation

Display and use these buttons to, respectively:

- Flip the images left/right,
- Flip the images up/down,

- Rotate the images left by 900,

- Rotate the images right by 900.

When rotating or flipping one of the images (reference image or functional map), the ori-
entation of the other images is automatically changed accordingly. You cannot rotate or
flip the graph view.
The L, R, A, P image orientation annotations are updated on the images when you flip or
rotate the views.
• To return the image to normal orientation, use the Display Normal function (see
below).

65
CT Perfusion

Display Normal

To return the images to the default setup, display and use this
button.
This function returns the series view (reference image) to a “best
estimate” (auto) W/L setting and removes all changes on all
images made by means of the scroll, zoom, rotate and flip func-
tions.

Grid

To display a grid overlay on the images or remove it from view, use this
button (toggle).
By default, the grid lines are spaced at 2.5cm.

Grid spacing: to change the spacing of the grid lines, dis-

play and use the (grid spacing) buttons. The (increase
spacing) button (left) increases the spacing by a factor /2
at each click, the (decrease spacing) button (right)
reduces it by the same factor.

Smoothed or Unsmoothed Images

Use this button to toggle between smoothed and unsmoothed images.

Smoothed images use a bi-linear interpolation for the display,
unsmoothed images use pixel replication.
Note: This is a display function only. It does not affect the resolution of
the computation of the functional maps. See pages 75 and 221
for the use of the algorithm resolution and spatial smoothing set-
tings to determine functional map resolution.

Annotations and ROIs

For the use of the buttons in the lower half of the central panel to cre-
ate and manage user annotations (ROIs and text), see Chapter 8
“Regions of Interest and Text Annotations”.

66
 CT Perfusion

Preferences/ Settings Menu

To set up your preferences for display layout, save

options and ROI statistics display, click on (Pref./
Settings) and make your selection in the menu.
Use [Graph view] to set up the graph, histogram
and list displays (3 panels). See page 95.
Use [Saving] to select whether to use color to
save screensave images, and the format to be
used when saving an image series or functional
maps as processed images. See page 97.
Use [Regions Of Interest] to select which ROI
statistics should be displayed on the series and
function views. See page 98.

Note: Certain functions and settings are also accessible by means of the on-view con-
trols and on-view menus. See Chapter 6 “On-View Controls”.

67
CT Perfusion

Graph View
Use the [Graph view] menu item to set up the graph and list displays. Move between the
panels by means of the (Back|Next) buttons.

Graph Preferences
As soon as more than one ROI is defined, the cor-
responding curves can be displayed superim-
posed in one single graph or as multiple
reduced-size graphs.
The curves can be plotted using different units for
the vertical axis:
• Absolute: uses the actual pixel values for each
ROI,
• Relative: assumes image 1 for each ROI is at
point 0 and all other images are plotted relative
to the first image,
• Percentage: plots the change in pixel value rel-
ative to the first image, expressed in percent of
the pixel value in the first image,
• Shape comparison: allocates 0 to the lowest
pixel value of each ROI and 1000 to the highest,
to scale all curves to the same vertical range.

The horizontal axis can indicate either rank (image numbers), or time.
The standard deviation of the pixel values within each ROI can be added to the curves
(shown as vertical lines).
The cursor size setting can be varied between 1 (1 pixel) and 10 (10x10 pixels). This
determines the number of pixels taken into account for the cursor ROI curve.
The (Autoscale) button automatically scales the graph view.

68
 CT Perfusion

ROI List Preference

The list can be displayed in order of either rank
(image numbers), or time (shown as “horizontal
unit” in the panel).
The values displayed in the list can be:
• Average: shows the average pixel value for
each ROI and each image,
• Max. or Min.: shows the maximum or minimum
pixel values,
• Std. deviation: shows the standard deviation of
the pixel values within each ROI for each image.
The four arrow buttons allow you to scroll through
the list (the graph view switches to List mode as
soon as any of these buttons is clicked). This may
be more convenient than using the active annota-
tions on the list view.

69
CT Perfusion

Saving
Use the [Saving] menu item to select whether to save screensave images in color (if
applicable) or as gray levels, and to select the format (“Type” in the Browser/Patient List
series list) to be used when saving an image series or functional maps as processed
images.

You can select whether screensave images are

saved in color or using gray levels, by turning the
Save ScreenSave images in color check box on or
off.
See Chapter 9 “Recording”, for the use of this
function.
If you use CT Perfusion 2 on a workstation that
does not support the DICOM color format, screen
save images will always be recorded using gray
levels, and the option is not shown.
Image series and functional maps can be saved as
processed images using the following formats:
- REFORMAT
- PROCESS
- SCPT

When saving an image series as processed images, the default setting is REFORMAT.
When saving functional maps as processed images, the default setting is PROCESS.
See Chapter 9 “Recording”, page 192 onwards for details.

70
 CT Perfusion

Regions Of Interest
Use the [Regions Of Interest] menu item to select the ROI statistics to be displayed on
the series and function views.

The following statistics can be displayed for each

ROI:

Area: the area of the ROI in mm2,

Average: the average of the pixel values within
the ROI,
Percentage of selected ROI: the average of the
pixel values within the ROI, expressed in percent
of the average of the pixel values of the selected
ROI. Displayed only if one of the ROIs is selected.
Min. and Max.: the minimum and maximum pixel
values within the ROI,
Std. deviation: the standard deviation of the pixel
values within the ROI.
By default both the series and function views show
only the area, average pixel value and standard
deviation.

Film/Save Menu

For the use of this menu to film and save the results of an analy-
sis, refer to Chapter 9 “Recording”.

71
CT Perfusion

Blank page.

72
 CT Perfusion

CHAPTER 6 - ON-VIEW CONTROLS

You access the CT Perfusion 2 control through:
− The main control panel, displayed at the left of the CT Perfusion 2 screen and
described in Chapter 5.
− On-view controls,
− Active annotaitons (displayed in red on the views).
These are system annotations that can be modified by the user, e.g., to change
window width or level, or to move to a different location in the exam.
− On-view menus,
− The review controller, to move throug the images using sliders and buttons or by
means of a cine loop. These controls are described in this chapter.

The user can create various types of regions of interest (ROIs) on the views. Their defini-
tion and use are described in Chapter 8.

73
CT Perfusion

1 OVERVIEW
The CT Perfusion 2 screen display consists of a main control panel at the left, and a view-
ing area at the right that contains four views.
The four views are illustrated and briefly described in Chapter 3. The controls (buttons
and menus) available in the main control panel are listed and described in Chapter 5.
You can also access a certain number of CT Perfusion 2 functions by means of the work-
station keyboard. See Chapter 7.
Each view has its on-view controls, i.e., controls that you access and use directly on the
views such as active annotations (displayed in red) and on-view (pop-up) menus.
Controls that are common to most if not all views are listed and described in section 2.
Controls that are specific to a particular view are described separately for the series view
(section 3), graph view (section 4) and function views (section 5).
The definition and use of regions of interest (ROIs) on the views is described in Chapter
8.

74
 CT Perfusion

2 ON-VIEW CONTROLS
Overview
Each view has a certain number of on-view controls:
− Interactive window width and level adjustment.
− Active annotations: these are system annotations on the views that you can modify to
adjust view parameters. Active annotations are displayed in red.
− On-view menus (also referred to as pop-up menus): you use these to access functions
specific to each view. Click right anywhere on the view except on the red annotations,
and select the required function in the menu that appears.
The contents of the on-view menus are described separately for each view further in
this chapter.

75
CT Perfusion

Adjusting Window Width and Level

Manual

In Selection mode, you can adjust window width (“contrast”) and win-
dow level (“brightness”) of the images interactively with the middle
mouse button (as in AW Basic Display), to control which image data val-
ues are visible on images and with what degree of contrast.

• When in Selection mode, use the mouse to modify

Decrease level
window W/L in the same way as on the AW Basic
Display Viewer: click and hold the middle mouse
button on a view, then move the mouse left or right
to modify window width, and up or down to modify
Decrease width Increase width window level.

Increase level

In W/L mode, you can adjust window W/L with the left mouse button (as
for the “zoom” and “roam” functions).
• Select (W/L mode) in the main control panel, then adjust window W/L
with the left mouse button as described above for the middle mouse
button.

The red W/L annotations at the bottom of the views are updated automatically whenever
you modify window W/L. These are active annotations, so you can also modify them
directly.

76
 CT Perfusion

Auto W/L
To automatically apply a “best estimate” W/L setting on the series view or on one of the
function views:

• Click and hold with either the left or the right mouse button on the W/L
active annotation on the view and drag to [Automatic] in the drop-
down menu.
or
• Move the mouse cursor onto the view and press <W> on the key-
board.

W/L Presets

To apply one of six predefined W/L settings on the series view:

• Click and hold with either the left or the right mouse button on the W/L
active annotation on the series view and drag to the corresponding
name in the menu.

77
CT Perfusion

Active Annotations
Certain on-view annotations are shown in red, to indicate they are active.
To use a numerical active annotation:
• Move the mouse pointer onto the annotation, and
- Click on the left or right mouse button to increase or decrease the value in
steps, or
- Click and drag left or right with the middle mouse button to change the value
in a continuous fashion (cursor changes to a <=> symbol), or
- Type the new value on the keyboard and validate with <Enter> (a small
numerical entry field appears on the view as soon as you start typing).
To use the text active annotations:
• Click with the right mouse button on the annotation and, where applicable, select the
desired function or setting from the pop-up menu.
To use the orientation annotations (L, R, A, P, S, I) on the sides of the images to roam
(scroll) the images.
Note: When you move the mouse pointer onto an active annotation, the cursor changes
shape, indicating that the annotation is active.
Active annotations that are available on most, if not all, views are described in the follow-
ing paragraphs:
− Patient name,
− Scan location,
− DFOV (Display Field View),
− Anatomical reference annotations,
− ROI annotation,
− Window width and level
The other active annotations are specific to each view and are described separately.

78
 CT Perfusion

Patient name
The patient's name is displayed in red. It can either be shown or hidden on the views.
• To hide the patient's name, move the mouse cursor onto it, click right and select [Hide
patient name]. The patient's name is replaced by "****" on all views.
• To return the patient's name to the views, move the mouse cursor onto the "****" leg-
end, click right and select [Show patient name].

When filming or saving images for diagnostic purposes,

CAUTION always make sure the patient name is displayed on all views.
See Chapter 9 "Recording".

Scan Location
The location of the currently displayed images in terms of RAS coordinates is shown by
the scan location annotation at top left on the views.
If the data set contains images for more than one scan location, the annotation is dis-
played in red and can be used to move to a different scan location in the set.
Note: If the data are for only one scan location, the annotation is displayed in yellow, and
is not active.
See page 106 on how to use the active annotation (click left mouse button to increase,
right mouse button to decrease, or drag middle mouse button).
Note: To move to an image with a different rank number, but at the same location, you
use the rank active annotation at bottom left on the series view or move the
vertical white bar on the graph view to the desired rank number and click on the
middle mouse button. Alternatively, you can use the <left> and <right> arrow keys
on the keyboard.

79
CT Perfusion

DFOV (Display Field Of View)

At startup, the DFOV (Display Field Of View) of the displayed images is the same as
those of the original acquisition (as shown in the Images list in the Browser/Patient List).
You can "zoom in"on the images by reducing the DFOV, using the red DFOV annotation
on the views.
• Move the mouse pointer onto the annotation, and
- Click on the left or right mouse button to decrease or increase the value in
steps. Each step corresponds to a sub-multiple of the original DFOV, or
- Click and hold the middle mouse button and drag to the left or to the right to
change the value in a continuous fashion, or
- Type the new value on the keyboard and validate with <Enter>.
The zoom factor (ratio between acquisition DFOV and actual DFOV) can be varied
between 1.0 and 8.0.

You cannot "zoom out" (increase the DFOV beyond the original value).

You can also use the (Zoom) button in the main control panel

Roam (Scroll)
You can use the red anatomical reference annotations (L, R, A, P, S, I) on the sides of the
images to roam (scroll) the image, in particular if you have "zoomed in" on an image (see
above) and want to move to a part of the image that you are interested in.
Click and hold a mouse button on any of these annotations, and drag the image around
as required.
When you move one of the images (reference image or functional map), the other images
also move automatically as soon as you release the mouse button.

You can also use the (Roam) button in the main control panel

80
 CT Perfusion

ROI annotation
For each ROI defined on the images the corresponding statistics (such as the area and
average pixel value) are displayed.
This annotation may at times mask part of the image. If this happens:
• Click on the red ROI annotation legend and drag the annotation to a different position.
Window W/L annotations
The red W/L annotations at the bottom of the views are updated automatically whenever
you modify window W/L. These are active annotations, so you can also modify them
directly.
To adjust either window width or level using the annotations:
• Click and hold the middle mouse button on the annotation and drag to the left or to the
right to change the value in a continuous fashion,
or
• Move the mouse pointer onto the annotation, type the new value on the keyboard and
validate with <Enter>.
To use the W/L annotation drop-down menu to select a W/L preset (series view) or adjust
window W/L automatically (all image views):

• Click and hold with either the left or the right mouse button on the W/
L annotation and drag to the corresponding function in the menu.

81
CT Perfusion

3 SERIES VIEW CONTROLS

The series view (upper

left) displays the
1
reference image.
2
3

4
5 7

82
 CT Perfusion

Series View Active Annotations

No. Annotations in Red Function

1 Patient name Show/hide patient's name
2 Scan location To change the scan location
3 DFOV To magnify the image
4 ROI To move the ROI measurement annotation
5 Rank To change the rank number
6 Anatomical reference (RAS) To scroll the image
7 Window W/L To change window width and level

Note:
- For the Patient name, DFOV, ROI, Anatomical reference (RAS) and
Window W/L annotations.
- Rank: The rank number is the number assigned to the image when the
images are sorted by increasing time and is displayed in the lower left hand
corner of the image in the series view. To move to an image with a different
rank number, but at the same location, use the rank active annotation at
bottom left on the series view.
- You can also move the vertical white bar on the graph view to the desired rank
number and click on the middle mouse button, or use the <left> and <right>
arrow keys on the keyboard. Either way, the rank number annotation is
updated automatically.
- Scan location: The location of the currently displayed images in terms of
RAS coordinates is shown by the scan location annotation at top left on the
views. If the data set contains images for more than one scan location, you
can use this annotation to move to a different scan location in the set.

83
CT Perfusion

Series View On-View Menu

To access the series views pop-up menu, click right on the view:

Pop up menu Function

[Display Normal] To display normal the image from a scrolled
or magnified state.
[Save View] To screen save the image to a screen save
series.
[Creat Annoataion] To add a text annotation on the view.

Note:
- [Display Normal] does not erase the graphics. Use the (cut) button, the
<Cut> key, or the <Ctrl-X> key combination, to erase annotations, graphs and
ROIs, or use the (hide annotations) button to hide them.
Review Controller

The {review controller}

consists of a narrow
Image location slider frame on the right and
bottom sides of the
series view that contains
tools for rapid and
comprehensive review of
the exam:
• Sliders and arrow but-
tons to move through
the images, either by
image location or by
Image location scroll buttons time/rank,
• Cine controls.
Loop mode Time/rank scroll buttons
Rock mode
Volume mode
Jog shuttle Time/rank slider

84
 CT Perfusion

To show or hide the {review controller} on the sides of the series view,
use this button (toggle) in the control panel.

To hide the controller, you can also click on the cross mark in the bottom
right corner.

Image location slider and arrow buttons

To move through the images for the different scan locations (only if
the exam contains images for more than one scan location):
• Click and drag on the vertical {location slider} to move through
the images in a continuous manner.
• Use the (up) and (down) arrow buttons to move through the
images in steps,
or
• Use the <up> and <down> arrow keys on the keyboard.
You can also use the scan location active annotation to move
through the images.

Rank slider and arrow buttons

To move through the images for the successive time points at the current scan location:

• Click and drag on the horizontal {rank slider} to move through the images in a contin-
uous manner.

• Use the (left) and (right) arrow buttons to move through the images in
steps,
or
• Use the <left> and <right> arrow keys on the keyboard.
You can also use the rank active annotation to move through the images.

85
CT Perfusion

Cine function

For proper results from the CT Perfusion 2 analysis tools, the image data for a given scan
location must be coherent, i.e., not degraded by patient motion during the acquisition.
You use the “cine” function to examine the image data at the current scan location for
possible patient motion before starting analysis.

To display a “cine” sequence of the images at the current scan location:

• Select an image at the desired location.

• Select (loop) or (rock).
Loop mode displays the images from first to last, then loops
around to start again with the first. Rock mode displays the
images forward from first to last, then back from last to first.
When you select either (loop) or (rock), the images are loaded
first, then displayed at low speed.

• Click and drag on the {jog shuttle} and move it away from the
center position to increase the display rate. Move the shuttle
to the right to display the images in sequence, to the left to dis-
play them in reverse order.
You can also directly click and drag on the shuttle to start a
cine loop. The loop will first run slowly, while the images are
being loaded, then at speed.
• De-select (loop) or (rock), or release the shuttle, to stop the
cine sequence.

You can also display a cine sequence of all locations with a given rank number:

• Select an image with the desired rank number.

• Select volume mode.
• Select (loop) or (rock) and use the {jog shuttle} as described
above to control the cine loop.

Note: Rapidly moving the {rank slider} or holding the middle mouse button on the rank
active annotation and rapidly moving the pointer back and forth will also produce a
semblance of a cine loop. You can use this to check the images around a suspect
location, but you should not use it as a substitute for the”cine” function, because
the display will skip images when the slider or mouse pointer is moved rapidly.

86
 CT Perfusion

4 GRAPH VIEW CONTROLS

Overview

The graph view (upper right) dis-

plays the image intensity curves
and statistics of regions of inter-
est in either a graph, table or his-
togram format.

Note:
The cursor ROI image intensity curve represents the change in pixel values for the
pixel(s) under the cursor for every image in the data set. This curve is automatically dis-
played and updated with each cursor movement, as long as no other ROI is selected.
The user ROI image intensity curves represent the same information for the pixels within
the area of each ROI defined on the image views (series, function and composite views).
Each image intensity curve is annotated with the corresponding ROI number.
The vertical axis represents the signal point spread plotted to the graph.
The horizontal axis (number scale) represents the image range (image numbers or time).
You can change this to view a subset (fewer images or less time) of the data set.

87
CT Perfusion

Graph On-View Controls

If you have defined an ROI on the views and selected it (ROI and image intensity curve
displayed in green), you can use the white moveable bars (crosshairs) on the graph view
to identify the image intensity value (horizontal bar) for a given image number (vertical
bar) at any point.
• You select the image number (or time) by simply moving the mouse cursor to the left or
right on the graph view. The vertical bar follows the cursor, and the horizontal bar auto-
matically indicates the corresponding image intensity value.
• To correlate a point on an image intensity curve with the reference image, move the
vertical white bar to the desired point on the curve and click on the middle mouse but-
ton.
The reference image corresponding to that rank point (time) is displayed in the series
view.
Graph View Active Annotations

The graph view

(upper right) dis-
1
plays the image
2 intensity curves
and statistics of
regions of inter-
est in either a
graph, table or
histogram for-
mat.

5 7
6 8 9 10

88
 CT Perfusion

No. Annotations in Red Function

1 Patient name Show/hide patient’s name.
2 Scan location To change the scan location.
Vertical scale To change the scale values of the graph’s
vertical axis. The vertical axis can represent
3 pixel intensities either in HU units (absolute,
relative or as a percentage) or re-scaled for
curve shape comparison.
Horizontal scale To change the scale values of the graph’s
horizontal axis. The horizontal axis can rep-
4 resent either image numbers from 1 to a
maximum of 1024 or image time in seconds
or milliseconds.
Rank To change the rank number (histogram view
5
only).
Cursor size To change the cursor size between 1x1 and
6
10x10 pixels (not on histogram view).
Region of Interest (ROI) Number A selected time curve is indicated by the cor-
7
responding ROI number.
8 Auto scale To adjust the vertical scale automatically.
Graph view layout To switch between a single graph view with
9 all time curves superimposed, or multiple
reduced-size graphs.
Region of Interest (ROI) Type Defines the ROI calculation as AVG (aver-
10 age), MAX (maximum), MIN (minimum), or
DEV (deviation).

89
CT Perfusion

Note:
- The graph view (upper right) can display the image intensity data in either a
graph, table or histogram format. Not all of the active (red) annotations listed
above appear on all three types of view.
- For the Patient name and Scan location annotations
- Vertical Scale and Horizontal Scale : the scale values of the horizontal and
vertical axis can be modified by means of the red annotations. Click left or
right, or hold the middle mouse button and drag to the left or right. Use the
keyboard’s <Space> bar or the red auto annotation to auto scale or center the
graph.
- Rank : changing the rank annotation on the histogram view automatically
changes the reference view. On the list view, if the image set contains more
images than can be listed on the screen, the rank annotation (left column)
becomes active (red) and you can use it to move through the list.
- Cursor size : you can adjust the cursor size (square) between 1x1 and 10x10
pixels.
- ROI number : the ROI number annotation is displayed whenever an ROI is
selected. You can use this annotation to select an ROI.
You can also select an ROI by clicking on the curve in the graph view. Note that each
time curve is annotated with the corresponding ROI number.
- Auto scale : to automatically scale the graph(s) displayed in the graph view:
click on the red auto annotation (bottom right). Alternatively, you can press
the <Space> bar on the keyboard, or use the (Autoscale) button in the (Pref./
Settings) > [Graph View] panel.
- ROI type : this annotation (bottom right, not on the histogram view) defines
what type of graph is displayed in the graph view: average value (AVG),
minimum value (MIN), maximum value (MAX) or standard deviation (DEV) of
the value of the pixels within the ROI(s).

90
 CT Perfusion

• Graph view layout : as soon

as more than one ROI is
defined, the corresponding
graphs can be displayed
superimposed in one graph
(1x1 annotation), or as sepa-
rate reduced-size graphs.
Click on the annotation to
toggle between 1x1
(superimposed graphs) and
multiple graphs.
You can also move the mouse
pointer on the view and press
the </> (“divide by”) key on
the keyboard.

Graph View On-View Menu

To access the graph view pop-up menu, click and hold right on the view, then drag down
to the menu item.
To make a selection in a submenu, drag right onto the item in the submenu.

91
CT Perfusion

Pop up menu Function

[Set X Unit]- rank or time Toggles between the rank number(images) or the
time (in sec. or msec.) for the X-axis or horizontal
axis of the graph.
[Set Y Unit] - absolute, relative, The Y-axis (vertical axis) can display the HU units
%, shape comparison as absolute, relative, in percentages or rescaled for
curve shape comparison.
Absolute gives the actual mean pixel values for
each region on the Y-axis.
Relative assumes image 1 is at point 0 and all
other images are plotted relative to the first image.
Percentages are relative to the pixel value in the
first images.
For shape comparison, all curves are rescaled to
the same Y unit range.
In all cases, there are as many data points as there
are images.
[Show ROI List] To display the ROI time curves in HU list format.
[Show Histogram] (or press Displays the histogram in number of pixels per HU
<H> on the keyboard) within an ROI.
[Show Graph] To display the time curve in graph format.
[Save View ] (or press <S> on To save the displayed graph, histogram or ROI list
the keyboard) to a screen save series.
[Create Annotation] To add a text annotation on the view.
[Show Annotations] or [Hide Displays or hides user-created text annotations.
Annotations.]
[Show Deviation] or [Hide Devia- Displays or hides the graphic display (vertical lines)
tion] of the standard deviation of the pixel values within
the selected ROI.
[Multiple Graphs] or [Single Displays the ROI time curves as separate graphs
Graph] or superposed in a single graph view.

92
 CT Perfusion

Note:
- The pop-up menu contents depend on the type of view (graph, histogram or
list values) and whether or not annotations are present and/or deviation is
displayed. Only the graph on-view menu is illustrated.
- A ROI must be present and selected for the [Histogram] option to be
available.
- [Create Annotation] : also see Chapter 8, page 170 onwards. The (aA)
button on the control panel cannot be used to create an annotation on the
graph view.
- [Multiple Graphs] or [Single Graph] is shown in the menu only when more
than one ROI is present.

Auto Scaling
To automatically scale the graph(s) displayed in the graph view:
• Press the <Space> bar on the keyboard,
or
• Click on the red auto annotation (bottom right on the graph view),
or
• Select (Pref./Settings) > [Graph View] and use the (Autoscale) button in the panel
that is displayed.
Rescaling is done according to the following rules:
- If an ROI is selected (displayed in green), the rescaling is based on that ROI
only.
- If no ROI is selected and the mouse cursor is on the graph view (i.e., no
cursor ROI curve is displayed), the rescaling is based on all ROIs present.
- If a cursor ROI curve is present on the graph view (i.e., no user-defined ROI
selected, and mouse cursor on the series view or on a function view), the
rescaling is based on the cursor ROI curve (displayed in white).

93
CT Perfusion

5 FUNCTION VIEW CONTROLS

The two function views

(lower half of the
1 screen) display the
2 functional maps cre-
3 ated by the current
4 protocol.

5
8

7 9

94
 CT Perfusion

Function View Active Annotations

No. Annotations in Red Function

1 Patient name Show/hide patient’s name.
2 Scan location To change the scan location.
3 DFOV To magnify the image.
Functional map To select the functional map to be displayed in
4
the view.
Color ramp max/min To change max and min values represented by
5
the color ramp.
6 ROI To move the ROI measurement annotation.
Transparency To adjust the transparency of a functional map
7
overlay.
8 Anatomical reference (RAS) To scroll the image.
9 WIndow W/L To change window width and level.

Note:
− For the Patient name, Scan location, DFOV, Anatomical reference (RAS) and Win-
dow W/L annotations.
− Functional map: This annotation shows the currently displayed functional map. Click
with the right mouse button on the annotation to show the list of functional maps com-
puted by the current protocol. Select the desired map from the menu.
− Color ramp max/min: You can use the scale value annotations (max and min) next to
the color ramp to modify the range of pixel values represented by the range of colors of
the color ramp.
Note that window width/level and color ramp max/min are simply two different ways of
presenting the same information:
- Color ramp max scale value = window level + 0.5 x window width,
- Color ramp min scale value = window level - 0.5 x window width.
Changing one of these automatically changes the other.
− Transparency: allows you to set the transparency on a composite view (overlay of a
functional map on a reference image).

95
CT Perfusion

Color Ramp Controls

Various color ramps are available from the on-view menu

In the function views, pixel values are displayed as a range of colors

(unless you have selected the gray-levels “color” ramp).
You control the range of pixel values that is displayed with the same
window W/L controls as for the reference view: middle mouse button
or the W/L active (red) annotations. You can also use the color ramp
max/min active annotations. On the function views these are generally
more practical than the standard window W/L controls.

The top of the color ramp corresponds to large values and the bottom to small values.
By default, values above the maximum value of the window are displayed with the top
color of the ramp.
• The (H) button at the top of the ramp allows you to toggle between displaying these
values either with the top color of the ramp or in black.
Values within the range of the window are distributed along the color ramp.
By default, values below the minimum value of the window are displayed in black.
• The (L) button at the bottom of the ramp allows you to toggle between displaying these
values either in black or with the bottom color of the ramp.

96
 CT Perfusion

Function View On-View Menu

To access the function view pop-up menu, click and hold right on the view, then drag
down to the menu item.
To make a selection in a submenu, drag right onto the item in the submenu.

Pop up menu Function

[Select Reference Image] - To select the reference image in a composite view.
original, selection or none
[Color Ramps] To select a color ramp for the function image.
[Display Normal] To display normal the image from a scrolled or magni-
fied state.
[Save View] To save the function image to a screen save series.
[Create Annotation] Create a text annotation on the views.
[Clear Region] To clear all of the function image displayed either out-
side or inside a selected ROI.
[Split View] or [Enlarge View] To split the function view into four reduced-size views or
to return it to full size.

Note:
− [Set Reference Image]: see page 136-137 for details.
− Various color ramps are available from the [Color Ramps] pull down menu. See
page 133 for details.
− [Display Normal] will not erase the graphics. Use the (cut) button, the <Cut> key, or
the <Ctrl-X> key combination, to erase annotations, graphs and ROIs, or use [Hide
Graphics] to hide them.
− [Save View] saves the function image as a screen save series (type SCPT). All anno-
tations and ROIs present on the image (including the color ramp) are saved with it,
except for the cursor and cursor position/value annotation.

97
CT Perfusion

The color information can only be saved and reviewed on systems that support the
DICOM color format. If this is not the case, the image and the color ramp are saved
and displayed using levels of gray.
− [Create Annotation]: creates a text annotation on all image views. See Chapter 8,
page 170 onwards.
− [Clear Region] is only listed in the pop-up menu if an ROI is present on the view, and
has been selected (displayed in green).
You can select [Outside ROI] or [Inside ROI] to clear all of the function image
located either outside or inside the selected ROI.
To restore the full image, re-select (Compute) in the last panel of the current protocol.
− [Split View] allows you to split a full-size function view into four reduced-size views.
The menu item changes to [Enlarge View] to return the reduced-size view currently
pointed to by the mouse cursor to a full-size view.

Composite View

A composite view displays an overlay of a

functional map onto a reference image.
The reference image can be either the origi-
nal reference image displayed in the series
view, or a separate image selected in the
Browser/Patient List.
A composite view can only be displayed on a
full-size function view, not on a reduced-size
(split) view.
To display a composite image:
• Click and hold right on the function view
that you want use for the composite view,
to open the on-view menu.
• Drag to the right on the [Set Reference
Image] menu item.
• Click on [Original] in the submenu to use
the original reference image,
or
• Click on [Selection] to use a different
image
Note: To return to the original function view,
select [None] in the submenu.

98
 CT Perfusion

Transparency
The Transparency active annotation on composite views allows you to set a varying
degree of transparency for the overlay of the functional map on the reference image (ini-
tially 50%). At 0% the overlay is fully opaque, at 100% it is fully transparent, i.e., no
longer visible.
Window W/L
To adjust window width and level on a composite view:
• To change the W/L or H/L of the functional map overlay, use the controls on the func-
tion view.
• To change the W/L of the underlying reference image, use the W/L controls on the
series view.
Selection o f the Reference Image
Normally, the reference image used for the composite view is the same (original) image
that is displayed in the series view.
However, certain exams may contain a separate anatomical reference image for the
same location. Usually, this is a separate high resolution image for the same location as
the time course images, when these are using a lower resolution.
If the exam contains such a separate anatomical reference image you can use this image
as a reference in the composite view:
• Click on the (Patient List) button to return the Browser/Patient List to the front,
• Select the anatomical reference image in the Browser/Patient List,
• Reselect the CT Perfusion 2 application,
• Click right on the composite view to display the pop-up menu, and select [Set Refer-
ence Image] > [Selection].
Note: A warning message is displayed if the location of the reference image that you
have selected does not match the current image set location.
The new reference image should be part of the current exam. If you select an image
in a different exam, the CT Perfusion 2 application will restart and load that exam.
The new reference image should be an acquisition image. Image formats such as
screen saves are not accepted.
• To return to the original reference image, again click right on the composite view to dis-
play the pop-up menu, and select [Set Reference Image] > [Original].

99
CT Perfusion

Saving Composite Views

When you film a composite image or save it as a screen save, the image is filmed or
saved exactly as it is displayed, i.e., using the current transparency setting
If you save a composite image as a processed image, only the functional map is saved.
See Chapter 9 “Recording”.

100
 CT Perfusion

CHAPTER 7 - REGIONS OF INTEREST AND TEXT ANNOTATIONS

You can add both graphic annotations (ROIs) and text annotations on the views.

Both types of user annotations are managed in the same way. You can move, edit, copy
or delete the annotations, and either show or hide them on the views.

101
CT Perfusion

1 OVERVIEW
We can distinguish two types of user annotation:
− You can add graphic annotations (ROIs) on the views to perform measurements or to
define a specific area to be displayed on the function view.
To create a graphic annotation on the views refer to section 2.

− You can add text annotations on the images or on a graph to draw attention to a spe-
cific area of interest.
To create a text annotation on the images refer to section 3.
To create a text annotation on a graph refer to section 4.

You manage graphics (ROIs) and text annotations in the same way. You can move, edit,
copy or delete annotations, and either show or hide them on the views. See section 5.

102
 CT Perfusion

2 DEFINING A REGION OF INTEREST (ROI) ON THE VIEWS

Overview
CT Perfusion 2 uses two types of “regions of interest” (ROIs): the cursor ROI and user-
defined ROIs.
− The cursor ROI coincides with the current cursor location and size. The corresponding
image intensity curve is displayed continuously on the graph view, unless a user-
defined ROI is selected.
− The user can define single-pixel, elliptic, box-shaped, polygon, spline, freehand, den-
sity mask, pixel list and mirror ROIs on the views.

User-defined ROIs have three distinct purposes:

− To define a specific area of interest on the views, that can be analyzed by means of the
corresponding image intensity curve displayed on the graph view and the ROI statistics
such as area and average pixel value inside the ROI.
− To show only the significant part of a functional map or remove irrelevant detail, by
defining an ROI and then removing the part of the functional map either inside or out-
side the ROI.
− To be used by the perfusion algorithms as references for the computation of parame-
ters such as blood volume, mean transit time and blood flow. Such reference ROIs
can be either generated automatically, or defined by the user.

103
CT Perfusion

Cursor Region of Interest (ROI)

Overview
The cursor ROI coincides with the current cursor location and size.
The ROI is considered to cover the same areas as the cursor (1 to 100 pixels). For
example, if the cursor size is set to 3 pixels, the ROI covers a square of 9 pixels (3x3).
The corresponding image intensity curve is displayed continuously on the graph view, but
only if no user-defined ROI is selected. It is updated in real time each time the cursor is
moved.
Cursor size
To adjust the cursor size:
• Place the mouse pointer on the red Cursor size annotation at bottom left on the graph
view, then click left or right to increase or decrease the cursor size in steps, or hold the
middle mouse button and drag to the left or to the right,
or
• Select (Pref./Settings) > [Graph View] and use the Cursor size slider.
You can adjust the cursor size between 1x1 and 10x10 pixels.
If the cursor size is larger than 1 pixel, the image intensity curve reflects the average of
the pixel values under the cursor ROI.
Cursor annotation
When you move the cursor on any of the image views (series view or function view), a
“cursor annotation” is displayed on the corresponding view (upper right side). This anno-
tation shows the RAS coordinates and the pixel value at the current cursor position.
Important: On the series view, the pixel value shown is that of the reference image at
the cursor position.
On the function view (functional map), the pixel value shown is that of the functional
map (i.e., the value at the cursor position of the parameter computed by the current
function).
On a composite view (overlay of a functional map on a reference image), the pixel
value shown is that of the functional map.
Note: The cursor ROI image intensity curve on the graph view is identical for the same
cursor position on the three image views, because it represents the changes in
pixel value (image intensity) across the depth of the image set at the current loca-
tion.

104
 CT Perfusion

User-defined Regions of Interest (ROIs)

The user can define different types of ROI on the views:
− Single pixel,
− Ellipse,
− Rectangle (box),
− Polygon/spline,
− Freehand,
− Density mask,
− Pixel list,
− Mirror.
ROIs can be split or merged, and grouped or ungrouped as required.
You can select ROIs to move and resize them, and to cut, copy and paste them.
The procedures for defining and managing ROIs on the views are described below (page
151 onwards).

The buttons used to create and manage the user annotations

(ROIs and text) are located in the lower part of the central panel of
the main control panel.

105
CT Perfusion

General
ROIs are shown at the same position on all three image views (series view and function
views), so that you can define, move and modify them as required on any of the three
views.
The graph view shows the image intensity curves corresponding to the ROIs. You can
select to display the curves corresponding to the average (AVG), maximum (MAX), mini-
mum (MIN) or standard deviation (DEV) of the pixel values enclosed by the ROI graphics.
You can also use ROIs to show only the significant part of a functional map or remove
irrelevant detail, by defining an ROI and then removing the part of the functional map
either inside or outside the ROI.
To select an existing ROI, click left either on the ROI in one of the image views or on the
corresponding curve in the graph view. A selected ROI (and the corresponding curve) is
displayed in green.
To select more than one ROI at the same time, hold down the <Ctrl> key on the keyboard
and either click left on each ROI in turn or draw a frame around the ROIs holding the left
mouse button.
To de-select the current ROI, click left anywhere within the image except on the ROI. The
color of the ROI changes from green to purple.
The cursor ROI (see page 148) is displayed in the graph view only if NO user-defined
ROI is currently selected.
An ROI can be cut, copied or pasted using the (Cut), (Copy) or (Paste) buttons on the
control panel, the <Cut>, <Copy> or <Paste> special function keys or the <Ctrl-X> (cut),
<Ctrl-C> (copy) or <Ctrl-V> (paste) combinations of keyboard keys.

106
 CT Perfusion

Measurements
For each ROI, a number of measurements (statistics) is computed:

− Area in mm2,
− Average of the pixel values within the ROI,
− The same average, expressed as a percentage of the average of the pixel values of a
selected ROI (only if one of the ROIs is selected),
− Minimum, maximum and standard deviation of the pixel values within the ROI.
On the series view the pixel value statistics are those of the reference image, on the func-
tion view those of the functional map.
By default, area, average and standard deviation are displayed on the series view and
function views, but you can use (Pref./Settings) > [Regions of Interest] to select the
ROI statistics to be displayed on these views.

The graph view shows the area in mm2 and in pixels, and the standard deviation, for the
currently selected ROI only.

107
CT Perfusion

Accuracy of on-view measurements

Measurements are calculated and displayed by the software to one decimal place (0.1
mm2). You should be aware that the real measurement accuracy is generally consider-
ably less for several reasons.
The main factors are image set resolution, display resolution, acquisition accuracy and
display settings.
Image set resolution: as an example, for a field-of-view (DFOV) of about 25cm, the size
of a single pixel in an image acquired with a 256x256 matrix will be 1x1mm. With a
512x512 matrix the pixel size will be about 0.5x0.5mm. The measurement accuracy can
obviously not be better than this.
Display resolution: each view equals 512x512 screen pixels, hence for a display field of
view (DFOV) of 25cm a screen pixel is equivalent to 0.5x0.5mm. Since you cannot place
a measurement point with a precision better than this, this may become the limiting factor
with high-resolution images.
Image set and display resolution set a lower bound on the geometrical accuracy:
− For an area measurement using a region of interest graphic, the geometrical accuracy
of the displayed area value is equal to +/- the circumference of the region of interest
multiplied by (pixel dimension)2 / 2.
− Mean and standard deviation values for the intensity of the pixels in the region are also
affected by this accuracy.
− Region of interest statistics are based on the pixels INSIDE the graphic defining the
region.
Note: Regardless of the zoom factor being used to view images, region of interest statis-
tics are calculated based on the pixels from the original, UNZOOMED image data
as they were in the Image Work.
Acquisition accuracy: any errors in the original image set resulting from the acquisition
process (calibration, patient motion, artefacts) will be added to the same extent to the
measurement error.
Display settings: since anatomical features are rarely of a uniform density, the apparent
dimension of an anatomical feature can change when you change the display settings
(window width and level). This may affect the accuracy with which an ROI can be fitted to
an anatomical feature on the screen.

You should be aware of the factors limiting measurement

accuracy when using the measurement information.
CAUTION

108
 CT Perfusion

Single-pixel ROI
As long as no other ROI is selected, the image intensity curve corresponding to the cur-
sor ROI is shown on the graph view and updated with each cursor movement
By holding down the <Shift> key and clicking left you can deposit an ROI on the view at
the cursor position, and effectively “freeze” the image intensity curve on the graph view.
The size of the ROI will be a single pixel, even if the cursor has been set to a larger size
Note: Once a single-pixel ROI is deselected (purple color), you may find it difficult to
reselect it on the image views by clicking on it, because of its small size.
However you can also reselect it by simply clicking on the corresponding curve in the
graph view, or by means of the ROI number active annotation at bottom right on the
graph view.
Or, hold the <Ctrl> key, then click and drag left to draw a small box around the ROI.
Since this selects all ROIs within the box, make sure only the desired ROI is selected.

109
CT Perfusion

Ellipse and Rectangle ROIs

You can deposit an ellipse or rectangle graphic on a view to define an elliptic or rectangu-
lar region of interest.

• Click on the (ellipse) or (rectangle) button.

An ellipse or rectangle ROI is deposited on the center of the views. You can now:

− Move the ROI by dragging either on the small + symbol

located in the center, or anywhere on the ROI graphic
except its four corners and rotation points.
− Resize the ROI by dragging on any of the corners (small
squares),
− Rotate the ROI by dragging on the rotation points (tick
marks on the four sides of the ROI graphic).

The mouse cursor changes shape to indicate the change in function.

The curve on the graph view and the statistics corresponding to the ROI are displayed
and updated as you adjust the ROI.
When you have positioned and sized the ROI satisfactorily, deselect it by clicking any-
where outside it.
Note: You can save the current size and shape of an ellipse or rectangle ROI, to be used
as the default size and shape whenever you deposit an ellipse or rectangle ROI.
To do so, select and size an ellipse or rectangle ROI as required, then press the
<R> key on the keyboard.

110
 CT Perfusion

Spline/ Polygon ROI

To create a region of interest with an irregular shape on the views, you can mark a con-
tour by depositing successive points on the views that the software connects either with
straight-line segments (polygon) or curved segments (spline).

A polygon is a ROI with three or more straight sides. A spline is the same polygon ROI
with curved sides.
To create a spline or polygon ROI on the views

• Click on the (spline) or (polygon) button.

• Click left on one of the views to deposit the successive
points.

When you have placed all points composing the ROI:

− To deselect the ROI without quitting spline/polygon mode:
• Click on the right mouse button.
− To quit spline/polygon mode and deselect the ROI:
• Click again on the (spline) or (polygon) button.

− To quit spline/polygon mode without deselecting the ROI:

• Click on the (selection) button.

111
CT Perfusion

You can also create a spline ROI in Selection mode:

• Hold down the <Shift> key on the keyboard, place the cursor where the ROI drawing is
to begin, and click left to deposit the first point. While continuing to hold down the
<Shift> key, move the cursor to the next point and click left again. When you have
placed all points composing the ROI, release the <Shift> key.

While you are marking the region of interest, the software automatically closes the con-
tour by adding a segment between the last point you have marked and the starting point.
Once you have defined the contour, you can:
− Switch the ROI contour between straight-line segments (polygon) and curved seg-
ments (splines):

• Select the ROI, then display and click on the (convert

polygon/spline) button to toggle between straight-line
segments and splines.

− Edit the ROI contour:

• In Selection mode, select the ROI.
• Place the mouse pointer on one of the definition points of the ROI (small squares),
click and hold left and drag the point to its new location.
− Move the ROI:
• Select the ROI, then drag anywhere on the ROI contour except its definition
points.
The curve on the graph view and the statistics corresponding to the ROI contour are dis-
played and updated as you adjust the ROI.

112
 CT Perfusion

Freehand ROI
To outline an irregularly-shaped region of interest more accurately than with a polygon or
spline ROI, you can mark a freehand trace on the views.

To creat a freehand ROI on the views:

• Click on the (freehand) button to activate freehand mode.

• Click and hold down the left mouse button and move the cursor
around to draw the ROI outline.
• Release the mouse button when you have completed the trace.
The ROI remains selected.

When you have drawn the ROI:

− To quit freehand mode and deselect the ROI:
• Click again on the (freehand) button.

− To quit freehand mode without deselecting the ROI:

• Click on the (selection) button.

You can also create a freehand ROI in Selection mode:

• Hold down the <Shift> key on the keyboard, place the cursor where the ROI drawing is
to begin, and hold down the left mouse button to start drawing. While continuing to

113
CT Perfusion

hold down the <Shift> key and the left mouse button, move the cursor around to trace
the ROI outline.
While you are marking the region of interest, the software automatically closes the outline
by adding a segment between the last point of the freehand trace and the starting point.
You will note that the procedures are very similar to defining a polygon/spline ROI, except
that you hold down the left mouse button continuously, rather than clicking to deposit indi-
vidual points.
Once you have defined the ROI, you can move it by dragging left anywhere on the ROI
outline.
However, you cannot edit its shape. If the outline is not satisfactory, delete the ROI
(using the (cut) button, the <Cut> special function key or the <Ctrl-X> key combination on
the keyboard), then redefine it.
The curve on the graph view and the statistics corresponding to the ROI outline are dis-
played and updated as you define the ROI.

114
 CT Perfusion

Density mask ROI

You can create a “density mask” region of interest on the views that is defined by a range
of pixel values.
This allows you to define an ROI that encompasses a discrete anatomical feature (e.g.,
vasculature) that can be defined by a range of pixel values, independently of where the
feature is located on the view.

To do this, you select a feature of interest either on the series view or on a function view,
then highlight the pixels with the same or similar values to define the density mask ROI.

To creat a density mask ROI on the views:

• Display the (density mask) button, and click on it to acti-

vate density mask mode.

• Position the cursor on the feature of interest in the view you want to use. If necessary,
move through the images to select the view that shows the feature of interest most
clearly.
• Hold down the middle mouse button.
This determines the starting value of the pixel value range of the density mask.
• While continuing to hold down the the middle mouse button, move the mouse to the
right to increase the range of pixel values, or to the left to decrease it.
The resulting density mask is displayed as a green hatched area on the view.

115
CT Perfusion

− To deselect the ROI without quitting density mask mode:

• Click on the right mouse button.
− To quit density mask mode and deselect the ROI:
• Click again on the (density mask) button.

− To quit density mask mode without deselecting the ROI:

• Click on the (selection) button.

You can also create a density mask ROI in Selection mode:

• First position the cursor on the feature of interest. If necessary, move through the
images to select the reference view that shows the feature of interest most clearly.
• Hold down either the <Alt> or the <> (“meta”) key on the keyboard, then the middle
mouse button. This determines the starting value of the pixel value range of the density
mask.
• While continuing to hold down the <Alt> or <> key and the middle mouse button, move
the mouse to the right to increase the range of pixel values, or to the left to decrease it.
The curve on the graph view and the statistics corresponding to the ROI are displayed
and updated as you adjust the ROI.
Additional on-view annotation appear at bottom right (above the W/L annotation) on the
view when you define a density mask:
− Va indicates the pixel value at the starting point on the feature of interest, I.e., the cen-
ter value of the density mask pixel value range.
− De indicates the deviation, i.e., the width of the pixel value range above and below the
center value. This is an active annotation, so you can also modify it directly (see Chap-
ter 6)
− The connected / disconnected annotation allows you to toggle between two modes:
- In connected mode only the pixels that are connected to the pixel used as a
starting point will be selected.
- In disconnected mode, all pixels in the defined density range are selected,
irrespective of their location on the view
To modify an existing density mask ROI:
• Reselect the ROI (click either on the hatched area or on the corresponding curve in the
graph view).
• Reselect Density mask mode, or hold down the <Alt> or <> key in Selection mode.
• Hold down the middle mouse button and move the mouse as before.

116
 CT Perfusion

Pixel List ROI

You can create a “pixel list” region of interest that consists of individual pixels placed on
the views, e.g., to create an ROI corresponding to a small irregular feature.

To create a pixel list ROI:

• Display the (pixel list ROI) button, and click on it to acti-

vate pixel list mode. The mouse cursor changes to a pen-
cil shape.
• Click with the left mouse button to place the individual
pixels. At each click the corresponding pixel is added to
the ROI.

When you have placed all pixels composing the ROI:

− To deselect the ROI without quitting pixel list mode:
• Click on the right mouse button.
− To quit pixel list mode and deselect the ROI:
• Click again on the (pixel list ROI) button.

− To quit pixel list mode without deselecting the ROI:

• Click on the (selection) button.

Note: The individual pixels of a pixel list ROI do not need to be connected.

117
CT Perfusion

Mirror ROI
You can create an axis of symmetry on the views, then create ROIs that are mirrored rel-
ative to this axis.
As an example, on a brain scan image this allows you to create two identical ROIs
located at the same locations left and right of the plane of symmetry of the brain, and
compare the resulting curves and ROI statistics.
The function can be used with irregularly shaped ROIs, because both location and shape
of the ROIs are mirrored relative to the axis of symmetry.

To create an axis of symmetry on the views:

• Click on the (axis of symmetry) button to display an axis of sym-

metry graphic on the views.
• Click and drag on the small squares at the two ends to position the
axis of symmetry on the views as required.
• Use the associated text box to name the axis if required. See sec-
tion 3, page 170 for details on how to enter a text annotation.

Note: You can create more than one axis of symmetry on the views (e.g., left/right and
front/back).
To create one or more mirror ROIs:

• Select the axis of symmetry you want to use,

• Select the ROI(s) that you want to create a mirror copy of.
To select more than one object (axis of symmetry and ROIs)
at the same time, hold down the <Ctrl> key on the keyboard,
then click on each of the objects.
• Display and click on the (create mirror ROIs) button.

Attention: Once created, you can select, move and resize a mirror ROI like any other
ROI. However, you should be aware that moving or resizing such an ROI will
destroy the symmetry.

118
 CT Perfusion

Split/Merge ROI
You can split an ROI on the views into a set of individual single-pixel ROIs, or merge sev-
eral single-pixel ROIs into a pixel list ROI.

To split an ROI into a set of single-pixel ROIs:

• Select the ROI, then click on the (split ROI) button.

The separate ROIs are shown on the image views and the corresponding curves are dis-
played on the graph view.
Each pixel-sized ROI can be selected and moved individually if required.
Note: If the selected ROI is too large to be split into individual single-pixel ROIs (the
maximum is 256 pixels), this is indicated by a message.
To merge several single®Cpixel ROIs into a pixel list ROI:
• Select the ROIs to be merged.
To select several ROIs at the same time, hold down the <Ctrl> key on the keyboard
and either click left on each ROI in turn or draw a frame around the ROIs holding the
left mouse button.

• Display and click on the (merge ROIs) button.

The resulting ROI is shown on the image views and the
corresponding curve is displayed on the graph view.

Note: These functions have no particular application for blood perfusion analysis, but
they may be used during general studies.

119
CT Perfusion

Group/Ungroup ROI
You can group several ROIs on the views into a single ROI, or ungroup such an ROI to
restore the individual ROIs.
To group several ROIs into a single ROI:
• Select the ROIs to be grouped.
To select several ROIs at the same time, hold down the <Ctrl> key on the keyboard
and either click left on each ROI in turn or draw a frame around the ROIs holding the
left mouse button.

• Click on the (group ROIs) button.

The resulting ROI is shown on the image views and the
corresponding curve is displayed on the graph view.

To ungroup an ROI created earlier by grouping several ROIs as described above:

• Select the ROI by clicking on it.

• Display and click on the (ungroup ROI) button.

The individual ROIs are restored on the image views and the
corresponding curves are displayed on the graph view.

120
 CT Perfusion

3 IMAGE ANNOTATIONS
You can add text annotations on the images, e.g., to draw attention to a specific area of
interest.

To create a text annotation on a series or function view:

• Click on the (aA) button in the main control panel.

A green box and arrow graphic appears on all image

views (not on the graph view). The box graphic pro-
vides the field for entering the text, the arrow allows
you to point out a specific structure.
An image view annotation is automatically displayed
on all image views (series view and function views),
with the same content and at the same position.

121
CT Perfusion

• Position the box and arrow and type the text. As

typing occurs, the box expands to accommodate
the text. Start a new line of text by pressing the
this is <Enter> key.
The mouse pointer MUST stay on the view for the
duration of annotation text entry.

this is sample text • To move the box and/or arrow after entering the
this is the 2nd line annotation:
Hold left on the annotation or arrow and drag it to
the new location, then release the mouse button.

• For a freestanding text annotation without an arrow,

press and hold left on the arrow’s point and drag it
into the box. When the mouse button is released,
the arrow is no longer visible.
• When you have finished adding the annotation and
positioning it, deselect it by pointing outside it and
clicking left. The box disappears and the annotation
remains.
• To move the annotation, re-click left on the annota-
tion. The box reappears and the annotation is reac-
tivated for a move.

Note: When you create an axis of symmetry on the views it is also accompanied by a
text annotation box (without an arrow). You enter the text and move it in the same
manner as described above.

122
 CT Perfusion

4 GRAPH VIEW ANNOTATIONS

To creat a text annotation on a graph:

• Click right on the graph view to open the graph view

pop-up menu, and select [Creat Annotation].
A green box and arrow graphic appears on the
graph view. The box graphic provides the field for
entering the text, the arrow allows you to point out
a specific feature.
• You can now type and position the annotation in the
same manner as described above in section 3 for
the image annotations.
•

Note: If you re-scale the graph view after adding an annotation, the position of the anno-
tation on the view will change also, and you may have to adjust it so that it points
again where you want it.
As noted in Chapter 6 “On-view Controls”,as soon as you have defined more than
one ROI, you can display the corresponding curves either superimposed in a single
graph or as multiple reduced-size graphs.
If you have added an annotation on the graph view in single-graph display mode, it
will not appear on the view when you switch to multiple-graphs mode (where its
meaning and position would be ambiguous), but it will reappear when you return to
the single-graph mode. The inverse applies to an annotation added on the graph
view in multiple-graphs mode.

123
CT Perfusion

5 MANAGING USER ANNOTATIONS

You manage both types of user annotation (ROIs and text) in the same manner.
To move, edit, copy or delete an annotation it must be selected. When selected, an
annotation is displayed in green.
• To select an annotation, click on it.
• To deselect an annotation, move the cursor away from it and click.
• To select more than one annotation at the same time, hold the <Ctrl> key on the key-
board and either click left on each annotation in turn or draw a frame around the ROIs
holding the left mouse button.
• To delete an annotation, select it by clicking on it, then click on the (Cut) button on the
control panel or press the <Cut> key (or <Control-X>) on the keyboard.

Multiple annotations can be added to an image and then later deleted using the (Copy),
(Paste) and (Cut) buttons on the control panel, the <Copy>, <Paste> and <Cut> special
function keys or the <Control-C> (copy), <Control-V> (paste) and <Control-X> (cut) key
combinations on the keyboard.
You can either show or hide user annotations (ROIs and text) on the views.

On the series view and function views:

• Use this button to toggle between showing and hiding the user
annotations. This button does not affect the annotations on the
graph view.

On the graph view:

• Click right on the view and select [Show Annotations] or [Hide Annotations] in the
pop up menu.
Note: These controls affect only the user annotations. To show or hide the system anno-
tations, press the <A> key on the keyboard.
When filming images or saving them as screen saves, user annotations will only
appear on the images if they are shown at the moment they are filmed or saved.
User annotations do not appear on images saved as processed images.

124
 CT Perfusion

CHAPTER 8 - RECORDING

To document the results of an analysis, you can:

− Film or print the images.
− Save the contents of any of the views (images or graphs) on the workstation disk in
screen save format.
− Save images or image sets as processed images, that can be re-loaded for further pro-
cessing either with CT Perfusion 2 or with other AW applications, and in particular with
3D image processing applications such as Reformat and 3D Analysis.
− Save lists of numerical data from the graph view in text format, or save view contents in
TIFF format, for export to other applications.

To film and/or save images, you use the functions from the Film/Save menu. Certain
functions can also be accessed using the on-view menus or the keyboard.

Filmed or saved images to be used for diagnostic purposes should always be correctly
annotated (system and user annotations). See section 7 in this chapter.

125
CT Perfusion

1 OVERVIEW

To document the results of an analysis, you can:

− Film or print views to obtain a hardcopy record.
− Save views (including annotations) in screen save format on the workstation disk.
− Save images or image sets as processed images, that can be re-loaded for further pro-
cessing.
− Export images and lists of data to other applications.

Controls
You use the the controls in the Film/Save menu on the main control panel to film and/or
save data. Some functions can also be accessed using the on-view menus and the key-
board.
See section 2 “Controls”.

Hardcopy
You can send images from the application to a laser camera for film output, or to a
DICOM color printer or black-and-white printer for paper output, depending on the output
device(s) linked to the workstation.
All system and user annotations (ROIs, text) are recorded on the filmed or printed images
exactly as displayed on the screen.
See section 3 “Filming and Printing”.

126
 CT Perfusion

Saving Views
You can record the contents of any of the views (images, graphs) on the workstation disk
in screen save format, then review and film or print these at a later date using a viewing
application.
All system and user annotations (ROIs, text) are recorded on the saved images exactly
as displayed on the screen.
To save images as screen saves you can use the controls in the Film/Save menu and on-
Cview menus, or the keyboard. You can also use the Filmer or the Scrapbook application
(if installed) to assemble a set of screen saves in a “scrapbook” that will be saved as a
single series.
See section 4 “Saving Views”.

Saving Processed Images

You can save the series data and the functional maps generated by CT Perfusion 2 in the
form of processed images, that can be re-Cloaded for further processing either with CT
Perfusion 2 or with other applications. In particular, if a saved image set contains data for
more than one scan location, it can be used with 3D image processing applications such
as Reformat and 3D Analysis.
The series data can be saved as “difference images” or as a new set of reformatted
images. Functional maps can be saved in several formats.
System information and image parameters are saved together with the images (not
recorded directly on the images as with screen saves). User annotations are not
recorded.
To save images or image sets as processed images, you use the controls in the Film/
Save menu, or the keyboard.
See section 5 “Saving Processed Images”.

Export
You can save view contents in TIFF format and ROI lists (numerical data) in text format,
either on a diskette or on the workstation hard disk. This allows you to export image and
text data to other computer systems (such as a PC or AW).
See section 6 “Export”.

127
CT Perfusion

Annotations
Filmed or saved images to be used for diagnostic purposes should always be correctly
annotated (system and user annotations).
See section 7 “Annotations”.

128
 CT Perfusion

2 CONTROLS

This section lists and illustrates the controls used to film and/or save data that are avail-
able in CT Perfusion 2. Their use is described in more detail in sections 3 to 6.
You access the filming and saving controls:
− In the Film/Save menu on the main control panel. See below.
− In the on-view menus, displayed by clicking right on a view.
− By means of the keyboard.

Film/Save Menu

• Click on (Film/Save) in the main control panel to open

the Film/Save menu.
You can now select:
• [Series Data], to display the panel used to save data
from the currently loaded series.
• [Graph Data], to display the panel used to save data
from the graph view (graphs, histogram or ROI lists).
• [Functional Maps], to display the panels used to
save one or more functional maps.

129
CT Perfusion

Film Composer
You use the Film Composer application to film selected images.
To display the Film Composer control panel:
• Select the Open [Film Composer] menu item in the Film/Save menu, To load images
into the Film Composer, you can:
• Use the Send to Film Composer controls in the Film/Save panels (see further),
or
• Place the mouse cursor on an image, then use the <F1> keyboard key to place it in the
next available slot,
or
• Use the <F2> keyboard key to place the entire page in the next available slots,
or
• Display the Film Composer control panel, then use “drag-and-drop” to place the
images in the slots.
To film, print or save the images you have assembled, use the controls in the Film Com-
poser control panels.
The system and user annotations (ROIs and text) on the filmed or saved images will be
the same as the annotations shown on the view(s) at the moment that you film or save
the image(s). See section 7 of this chapter.

130
 CT Perfusion

Series Data
You use this panel to save data from the currently loaded series. Select (Film/Save) >
[Series Data] to open the panel.

• Select the desired function:

- Use Send to Filmer or Send to
Scrapbook or Send to Film
Composer to load the current image in
the series view into the next available
slot in the Filmer, Scrapbook, or Film
Composer.
- Use Save as ScreenSave image to
save the current contents of the series
view as a screen save.
- Use Subtraction images to save the
series as a set of “difference” images.
- Use Reformatted images to save the
series as a set of reformatted images.
• Select (Save) to confirm the selection, then
(Close) the panel.

131
CT Perfusion

Graph Data
You use this panel to save graph data. Select (Film/Save) > [Graph Data] to open the
panel.

• Select the type of data to be saved:

− Graph, Histogram or ROI List.
Note that the type you select does not have to be
displayed in the graph view.
• Select the desired function:
− Use Save as ScreenSave image to save the
current contents of the series view as a screen
save.
− Use Send to Film Composer to load the data
into the next available slot in the Film Composer.
− Use Save as ScreenSave image to save the
data as a screen save.
• Select (Save) to confirm the selection, then
(Close) the panel.

Note: The ROI list may contain more data than can be displayed in a single view. To film
or save the ROI list it is preferable to first display the list in the graph view and
check its contents. It may be necessary to film or save an ROI list in several parts,
by scrolling the list using the image/rank number and ROI number active anno-
tations as applicable.

132
 CT Perfusion

Functional Map(s)
You use these panels to save functional maps. Select (Film/Save) > [Functional Maps]
to open the first panel, then use the (Back|Next) buttons to move between the panels.
First Panel

• Select the functional maps to be saved in the

first panel:
− To select from the currently displayed maps,
select Visible maps and Left, Right or All.
− To select from all available maps computed by
the protocol (even if not displayed), select
Available maps... , then select |Next) to dis-
play the corresponding panel (see page 185).
• .Select the location(s) to be saved
− Use Single location to save only the map(s)
for the current scan location.
− Use Multi location to compute and save the
map(s) for all scan location in the currently
loaded series.
• Select |Next) to move to the final panel (you
may have to click twice, if the Available maps
panel is active).

Note: A “visible” map may be hidden under the Film/Save panel. You can move the
panel by clicking and dragging on its title bar.
Only visible maps can be loaded into the Filmer, Scrapbook, or Film Composer.
For other selections, this function is not available.

133
CT Perfusion

Sub-Panel for Selection of Available Maps

If you have selected Available maps... and |Next) in the first panel, a separate panel
with a list of all available maps is displayed.

• To select the maps to be saved, click on the name

of each map to highlight it. To cancel a selection,
click on the name a second time.
• Select |Next) to move to the final panel.

Final Panel
You use the final panel to select the format to be used to save the selected functional
map(s).

Select the desired function:

− Use Send to Film Composer to load the
selected images into the next available slots in
the Film Composer.
Note: this function is only available if one or
more visible maps have been selected.
− Use Save as ScreenSave images to save the
selected images as screen saves.
− Use Save as processed images to save the
selected images as a set of processed images.
See page 196.
• Select (Save) to confirm the selection, then
(Close) the panel.

134
 CT Perfusion

Saving Split Function Views

If a function view has been split to display four reduced-size images, the following applies
when you select such a view for filming or saving:
− When you send such a view to the Film Composer, a standard size image is loaded for
each map, but with reduced annotations and low resolution (256x256).
− When you use “ Save as ScreenSave images” (see above), standard size and resolu-
tion screen save images are saved for each map.
− When you use “save view” in the on-view menu or the <S> key, only the map that is
currently pointed to by the cursor is saved as a screen save of standard size but with
reduced annotations and low resolution (256x256).
− When you use “ Save as processed images” (see above), standard processed images
are saved for each map.
− When you use the <F> key, only the map that is currently pointed to by the cursor is
saved as a standard processed image.

On-View Menus
All on-view menus contain a “save view” menu item that allows you to save the current
view in screensave format.
• Hold the right mouse button on the view to open the on-view menu, then select [Save
view]. The view is saved as a new series in the exam, type SCPT.

135
CT Perfusion

3 FILMING AND PRINTING

The AW workstation allows the user to send the images to a laser camera for film output
or to a printer (color and/or black-and-white) for paper output. To do so, the workstation
must be linked to an appropriate output device. See your sales representative and ser-
vice engineer for details.
The system and user annotations (ROIs and text) on the filmed images will be the same
as the annotations shown on the view(s) at the moment that you load the image(s) into
the application. See section 7 of this chapter.

136
 CT Perfusion

4 SAVING VIEWS
Instead of filming or printing the views, you can also document the results of an analysis
by saving the views in screensave format on the workstation hard disk.
You can save the contents of any of the views (image or graph) as a screen save, using
the controls in the Film/Save menu and on-view menus, or the keyboard. Each view is
saved as a new series of the exam, with type “SCPT” (secondary capture).
To save the contents of a view as a screen save:
• Use the Save as ScreenSave image menu item in the Film/Save menu panels (see
section 2),
or
• Move the cursor onto the view that you want to save,
• Hold down the right mouse button, and select [Save view] in the on-view menu,
or
• Press the <S> key on the keyboard.

Annotations
Any annotations on the view will be saved with it and will be shown when the view is
accessed at a later date. However, the annotations become part of the image: they can
no longer be edited or deleted (see section 7 of this chapter).

137
CT Perfusion

Color
If your workstation and installed software support the saving and review of images in
DICOM color format, CT Perfusion 2 will by default save screensave images in color
using this format.
You may at times want to transfer images from CT Perfusion 2 to other workstations and
systems, that do not support the DICOM color format. In that case:
• Select (Pref./Settings) in the main control panel, then [Saving].
• Turn off the Save ScreenSave images in color check box.
• (Close) the panel.
Screen save images will now be recorded using gray levels instead of color.

138
 CT Perfusion

5 SAVING PROCESSED IMAGES

You can save the series data and the functional maps generated by CT Perfusion 2 in the
form of processed images, that can be re-loaded for further processing either with CT
Perfusion 2 or with other applications. In particular, if a saved image set contains data for
more than one scan location, it can be used with 3D image processing applications such
as Reformat and 3D Analysis.
The series data can be saved as “difference images” or as a new set of reformatted
images. Functional maps can be saved in several formats.
System information and image parameters are saved together with the images (not
recorded directly on the images as with screen saves). User annotations are not
recorded.
The color ramp on the function views is not saved with the images. However it is auto-
matically restored when functional maps that were saved as processed images are re-
loaded into CT Perfusion 2 (using the “Import” protocol).

139
CT Perfusion

Series Data Subtraction Images

An image set loaded by CT Perfusion 2 can be saved in the form of “difference images”.
For each image in the set at a given scan location, the first image at the same scan loca-
tion is subtracted from the image, then the image is saved. If the set contains images for
more than one scan location, the process is repeated for each scan location in the entire
set.
The image set is saved as a new image set in a new series of the exam. The new series
appears in the Browser/Patient List with description “Nth-1st” and with the same modality
as the original image set.
To change the save type (before saving):
• Select (Pref./Settings) in the main control panel, then [Saving]. Set the Series Data
Save Type as required. (Close) the panel.
To save the currently loaded image set as a set of “difference images”:
• Select (Film/Save) in the main control panel, then [Series Data]. In the panel that is
displayed, select Subtraction images and (Save).
Note: An image set saved in this manner can be loaded, displayed and analyzed again
with CT Perfusion 2, but due care should be taken when interpreting the analysis
results from such data sets, because the pixel values no longer represent absolute
HU values, but difference values.

Series Data Reformatted Images

An image set loaded and processed by CT Perfusion 2 can be re-saved in the form of
processed images.
You can use this function to re-save the complete image set (all locations) after you have
performed registration to remove the effects of patient motion.
The image set is saved as a new image set in a new series of the exam. The new series
appears in the Browser/Patient List with description “Sorted” and with the same modality
as the original image set. The type can be REFORMAT, PROCESS or SCPT. The
default is REFORMAT.
To change the save type (before saving):
• Select (Pref./Settings) in the main control panel, then [Saving]. Set the Series Data
Save Type as required. (Close) the panel.
To save the currently loaded image set as a set of processed images:
• Select (Film/Save) in the main control panel, then [Series Data]. In the panel that is
displayed, select Reformatted images and (Save).

140
 CT Perfusion

Functional Maps
The functional maps generated by CT Perfusion 2 can be saved in the form of processed
images. These can be re-loaded by CT Perfusion 2 for further processing (using the
“Import” protocol). They can also be processed with other AW applications.
The (Film/Save) menu lets you select exactly what to save: one or more maps (selected
from the visible maps or from a list of all available maps), either for the current scan loca-
tion only or for all locations.
The functional maps are saved as new series of the exam, one for each map. The new
series appears in the Browser/Patient List with the name of the map (e.g., “Blood Vol-
ume”) in the description column and with the same modality as the original image set.
To save functional maps as processed images:
• Select (Film/Save) in the main control panel, then [Functional Maps].
• In the first panel that is displayed, select the functional map(s) to be saved, then
(Next).
• In the second panel, select Save as processed images and (Save),
To save a single functional map (that is displayed on the screen) as a processed
image, you can also:
• Move the cursor onto the function view with map that you want to save,
• Press the <F> key on the keyboard.
Note: :If you have split a function view, to display four reduced-size functional maps:
- If you use the (Film/Save) menu to select and save the view, all functional
maps that are present in the view are saved as full-size processed images.
- If you move the mouse pointer on one of the reduced-size maps and press
the <F> key, only that map is saved as a full-size processed image.
A function view can display a composite image that consists of the overlay of a functional
map on a reference image. When saving the contents of such a view as a processed
image, only the functional map is saved.

141
CT Perfusion

Pixel Values in Saved Images

When exploring saved CT Perfusion 2 images with other viewing applications (such as
the Basic Display Viewer), the pixel values in the saved images may be different from the
original values in the functional map or composite image.
Pixel values are stored on the image disk as integers.
To avoid loss of information when a functional map is saved, a scale factor is applied
whenever the values returned by the function (algorithm) are too small to be meaningful
when rounded off to integer values.
See Appendix 1 “Functions”, section 5 for a detailed description and a list of scale factors.

UNDER NO CIRCUMSTANCES SHOULD THE PIXEL VALUES FROM

SAVED FUNCTIONAL MAPS BE USED BY ANY SOFTWARE APPLI-
WARNING CATIONS THAT RELY ON HOUNSFIELD VALUES.
THIS APPLIES IN PARTICULAR TO DOSE COMPUTATION SOFT-
WARE APPLICATIONS.

142
 CT Perfusion

6 EXPORT
You can save view contents in TIFF format and ROI lists (numerical data) in text format,
either on a diskette or on the workstation hard disk.
Note: Before using this function to save data on the hard disk.

Save View in TIFF Format

TIFF (Tagged Image File Format) is a general-purpose image file format used to export
screen captures to other (AW) applications and systems, e.g., to be used as illustrations
for reports, publications or documentation.

143
CT Perfusion

7 ANNOTATIONS

Show/Hide Annotations

The user annotations (text and ROIs) on the image views can be
shown or hidden by means of the (show/hide annotations) button in
the main control panel.
• User text annotations on the graph view can be shown or hidden by
means of the [Show annotations] and [Hide annotations] menu
items in the on-view menu of the graph view.
• The system annotations on the views can be toggled on or off using
the <A> key on the keyboard.

Saving Annotations
When a view is filmed (or printed), any annotations on the view will also appear on the
film.
When a view is saved on the image disk as a screen save, any annotations on the view
will be saved with it and will be shown when the view is accessed at a later date.
At this stage, the annotations become part of the image: they can no longer be edited or
deleted.
If any annotations on the screen have been turned off they will not be present on either
filmed or screen save images.
When saving processed images, the information from the system annotations is saved
automatically with the view, even if those are hidden at the moment that you save the
image.
User annotations (text and ROIs) are not saved with processed images.

Although images without annotation may be suitable for

teaching purposes, diagnosis should not be performed with
CAUTION such images.

144
 CT Perfusion

Patient Name
While working on an exam, you can hide the patient name on the views for increased
confidentiality. If you have done so, make sure to show the patient name again on the
views BEFORE filming or saving images for diagnostic purposes.

When filming or saving images for diagnostic purposes,

always make sure the patient name is displayed on all views.
CAUTION

Cursor Annotation
Whenever you move the cursor on one of the image views (series, function or composite
view), a cursor annotation is displayed on the view (near top right of the view). This
annotation shows the RAS coordinates and the current pixel value at the cursor location.
This cursor annotation is NOT recorded, neither on film, nor on any images saved to disk.

145
CT Perfusion

Blank page

146
 CT Perfusion

APPENDIX- FUNCTIONS
The functional maps displayed in the function views are the result of a calculation pro-
cess defined by a function that generates a single value at each pixel location of the orig-
inal data set.
Within CT Perfusion 2, a function can be either one of the basic algorithms, or it can have
been derived from one of these algorithms.
The application is supplied with a set of pre-defined protocols, that contain the necessary
information and controls to set up the input parameters, and then invoke the appropriate
algorithm(s).
CT Perfusion 2 comprises a CT perfusion algorithm designed specifically for the analysis
of blood perfusion in CT brain and body scan data, and a set of other generic algorithms.
This Appendix details the basic theoretical background of the algorithms and derived
functions and their input parameters.

Pixel values in saved images may be different from the original values in the functional
maps. This Appendix describes the differences and lists the scale factors used.

147
CT Perfusion

1 OVERVIEW
General
A functional map displayed in a function view is a single image calculated from a set of
images acquired at a single location but at different times (time course images). The
image intensity values in the functional map represent the result of a calculation process
that generates a single value at each pixel location of the original data set.
The calculation process is defined by an algorithm. Each algorithm normally requires one
or more input parameters. You use the algorithms provided with CT Perfusion 2 by
means of the protocols.
A protocol contains the necessary information and controls to set up the input parame-
ters, and then invokes the appropriate algorithm(s).
CT Perfusion 2 provides you with a set of pre-defined protocols.
This Appendix describes the theoretical background of the algorithms and derived func-
tions.
− For the CT Perfusion algorithm, see section 2 .
− For the other algorithms and functions, see section 3 .
− For the input parameters used with these algorithms and functions, see section 4 .
− Pixel values in saved images may be different from the original values in the functional
maps. For a detailed description of the differences and the scale factors used, see
section 5.
Algorithms
The algorithms incorporated in the CT Perfusion 2 application are listed below.
The Brain Stroke, Brain Tumor and Body Tumor protocols use the CT Perfusion algo-
rithm, which computes:
− Blood Volume (BV),
− Mean Transit Time (MTT),
− Blood Flow (BF),
− Permeability Surface area product (PS, Brain Tumor and Body Tumor protocols only).
The CT Standard protocol uses the following algorithms:
− Positive Enhancement Integral,
− Time To Peak,
− Maximum Slope of Increase,
− Maximum Slope of Decrease.

148
 CT Perfusion

Input Parameters
When you select a protocol, the successive panels displayed by the protocol allow you to
set the input parameters needed for the algorithm(s) used by the protocol.
The input parameters used by the algorithms and functions in the CT Perfusion 2 proto-
cols are listed below, and described in section 4:
− Pre-enhancement image(s) and Post-enhancement image,
− Processing thresholds,
− Artery ROI,
− Vein ROI.
Some input parameters for the CT perfusion algorithm are set to default values, but can
be adjusted if necessary:
− Maximum blood flow value,
− Hematocrit ratio,
− Average tissue density,
− Algorithm resolution and Spatial smoothing factor.

149
CT Perfusion

2 CT PERFUSION ALGORITHM

Overview
The CT Perfusion algorithm allows you to characterize and quantify image intensity vari-
ations in CT image sets following the injection of a contrast agent.
The algorithm is used to compute absolute values for Blood Flow (BF), Mean Transit
Time (MTT), Blood Volume (BV) and Permeability Surface area product (PS).

The formal definitions of Blood Flow and Mean Transit Time assume that the injection of
the contrast agent is instantaneous, which of course can never be the case in practice.
This means that the algorithm must take into account the actual injection rate of the con-
trast agent to obtain quantitative results for BF and MTT.
To this purpose, the algorithm uses data from a reference ROI located in an artery to
deconvolve the time course data and compute an impulse residue function (IRF) for each
pixel location.
The impulse residue function is best described as the time curve that would have resulted
from an ideal injection of contrast agent with a duration of one unit of time (impulse). The
algorithm uses the real injection rate, as determined from the artery reference ROI, to
“translate” the time course data at each pixel location into the corresponding IRF.

Blood flow, mean transit time, blood volume and permeability surface area product are
derived from the IRF.
Where applicable, they are then normalized relative to the pixel value for 100% blood in a
large vessel (usually a vein) using a second reference ROI and relative to the average tis-
sue density to obtain final results expressed per unit mass.
For display purposes, the IRF is reconvolved with the arterial input curve and the result-
ing curve is shown on the graph view together with the original pixel data, either as a blue
curve for the cursor ROI or as a red curve for the currently selected user ROI. This
allows you to assess the goodness of fit of the computed data (reconvolved time curve) to
the original time curve.
The accuracy of the results from the algorithm is critically dependent on the time resolu-
tion of the image data set. A time resolution of 1 second is adequate, 0.5 seconds is
ideal. Accuracy will degrade rapidly for larger time resolution values. The algorithm
imposes a minimum MTT equal to the time resolution.
This applies in particular to the first minute of the data. For longer acquisitions (up to sev-
eral minutes, as used in tumor studies) the time interval can be larger towards the end of
the data set.

150
 CT Perfusion

Computing time is also very dependent on time resolution. If N is the number of images
in a given slice, computing time varies roughly as N3, i.e., computation with a 0.5s time
resolution is approx. 8 times as slow as with a 1.0s time resolution.

Input parameters
The input parameters used by the algorithm are described in section 4.

Computed Results

Blood Flow
Blood Flow (BF) is computed and displayed in ml per 100g of wet tissue per minute.
The blood flow is derived from the initial value of the impulse residue function:
BF = IRF(t0)

Mean Transit Time

Mean Transit Time (MTT) is computed and displayed in seconds
Mean transit time characterizes the temporal position of the transient increase in the time
course data. Transient increases that occur later have a larger transit time (higher value
in the function view, i.e., more towards the red when using the “rainbow” color map).
In mathematical terms, the mean transit time is computed as the first moment of the
impulse residue function.
The first moment of a curve y=f(t) is defined as the mean value of t where each t is
weighted by the value of f(t), i.e., it is the sum of all t*f(t) values, divided by the sum of all
f(t) values:
t=0(t*f(t))
t=0f(t)
For the computation of MTT, f(t) is the impulse residue function.

151
CT Perfusion

Blood Volume
Blood Volume (BV) is computed and displayed in ml per 100g of wet tissue.
The blood volume is the product of the blood flow and the mean transit time:
BV = BF*MTT

Permeability Surface Area Product

The Permeability Surface area product (PS) is computed and displayed in ml per 100g of
wet tissue per minute.
The permeability surface area product characterizes the diffusion of some of the contrast
agent from the blood vessels into the interstitial space due to brain-blood-barrier (BBB)
leakage or tumors.
It is computed from the impulse residue function. Contrast agent diffusion appears in the
IRF as a residual enhancement that occurs after the initial impulse response and that
decreases exponentially with time.
Note that PS uses the same dimension and units as BF, because it also quantifies flow (in
this case the flow of contrast agent into the interstitial tissue).
The permeability surface area product functional map is used in tumor examinations,
hence the corresponding function is included only in the Brain Tumor and Body Tumor
protocols.

152
 CT Perfusion

3 OTHER ALGORITHMS
The algorithms below are used in the CT Standard protocol.
− Positive enhancement integral,
− Time to peak,
− Maximum slope of increase and Maximum slope of decrease.

Positive Enhancement Integral

Time course data acquired during the injection of a contrast agent may have image inten-
sity variations, which result in positive enhancement.
The time course pixel intensity si is expressed as:
si = s0 f(t),
A parameter that is used to characterize the time intensity changes is the integral of the
area I under the enhancement curve.

The function returns the integral over the image range of the difference between the
value at each time point and the pre-enhancement value.
Input parameters (see section 4): pre-enhancement and post-enhancement image num-
bers.

Time To Peak
Time-to-peak (TTP) is the time between the onset of the enhancement transient (last pre-
enhancement image) and the peak value of the time curve (image with the maximum
value before the first post-enhancement image).
Time-to-peak is computed and displayed in seconds, using the raw time curve data
directly.
Input parameters (see section 4): pre-enhancement and post-enhancement image num-
bers.

153
CT Perfusion

Maximum Slope
One way to characterize image intensity changes during a dynamic process is to calcu-
late the slope of the time course values at each time course index, i, given by:
slopei = si+1 - si.
The single valued parameter returned by the Maximum Slope of Increase algorithm is
simply the maximum value of the slope i function:
MAXi=0,N (slopei).
By analogy, the Maximum Slope of Decrease algorithm returns the minimum value of
the slope i function:
MINi=0,N (slopei).
Input parameters (see section 4): pre-enhancement and post-enhancement image num-
bers.
Note: When either of these algorithms has been selected, the location of the maxi-
mum slope of increase or decrease is also shown on the graph view, either
as a blue segment for the cursor ROI curve or as a red segment for the cur-
rently selected user ROI.

154
 CT Perfusion

4 INPUT PARAMETERS

General
This section describes the input parameters used by the standard CT Perfusion 2 func-
tions (algorithms and derived functions). You adjust these parameters by means of the
controls in the panels that are displayed when you select a protocol.
When the software proposes a default value for an input parameter, you can often leave
this parameter at the proposed value at least for the initial exploration of the data set.

155
CT Perfusion

Input Parameters
Pre-enhancement image(s) and Post-enhancement image
These input parameters are used to define the range of images that will be processed by
the selected algorithm(s).
The pre-enhancement image range is mostly used by the algorithms to determine a
baseline for the computations. It can be defined as the range of images before the onset
of the first intensity transient following the injection of the contrast agent.

In the CT Standard protocol you can define both a first and a last pre-enhancement
image.
The “first pre-enhancement image” parameter is usually set to the first image of the data
set, but it can be set to a later image to limit the pre-enhancement range to the image(s)
just before the onset of the first transient, and eliminate any anomalies that may be
apparent in the first few images.
The “last pre-enhancement image” parameter should be set to the last image prior to the
onset of the transient.
These input parameters are used to define the range of images that show the intensity
transient following the injection of the contrast agent.

In the Blood Perfusion protocols you define only the last pre-enhancement image (the
first pre-enhancement image is the first image of the data set).
Set the “last pre-enhancement image” parameter to the last image prior to the onset of
the transient.

The post-enhancement image range settings are used primarily to limit processing time.
This may be necessary in the case of long-duration acquisitions (several minutes) as
used in tumor studies.
As noted earlier, computing time varies roughly as N3 where N is the number of images
in a given slice. For twice the number of images computation will be approx. 8 times
slower.

In the Brain Stroke protocol, the post-enhancement image range is determined by the
“first post-enhancement image” parameter. This parameter should usually be left at the
default value (the last image). Reduce its value only if processing time appears exces-
sive. For correct results the parameter should always be set at least to an image corre-
sponding to a time point after the end of the first intensity transient following the injection
of the contrast agent as identified on the graph view.

156
 CT Perfusion

In the Brain Tumor and Body Tumor protocols (tumor studies), the post-enhancement
image range is determined by the “last post-enhancement image” parameter. For the
most accurate results, this parameter should be left at the default value (the last image).
It should be reduced only if the duration of the acquisition is too long to be processed
within an acceptable time.

For the algorithms in the CT Standard protocol, the “first post-enhancement image”
parameter can usually be left at the default value (the last image). You can reduce the
value of this parameter if processing time appears excessive (only with long acquisi-
tions), or if you specifically want to limit the range of images used for analysis to those
that show the first transient, and exclude those that show the effect of blood recirculation
(secondary transients).

Processing Thresholds
All pixels in the series view with HU values outside the processing thresholds will be
masked out, i.e., they will NOT be computed and will appear black in the function views.

This has a dual purpose:

− You can eliminate irrelevant image data outside the patient from the computation. This
can amount to as much as one half of the data, and the computing time is reduced
accordingly. While this may make little difference for simple algorithms, the gain in
time can be significant for complex algorithms such as the CT Perfusion algorithm in
the blood perfusion protocols.
− By eliminating extraneous data you can “clean up” the function views (again, in particu-
lar outside the patient) for easier interpretation of the resulting image.

Artery ROI
The algorithm uses a reference ROI located in an artery to account for the actual injection
rate of the contrast agent.
The artery ROI can be defined by the user or automatically by the software.
In automatic mode the artery is defined by selecting among all pixels with a sufficient
enhancement (all vessels) the region of pixels with the smallest peak time, i.e., the point
where the contrast agent arrives first.

157
CT Perfusion

Vein ROI
The algorithm uses a reference ROI inside a large vessel (vein) to normalize the result
relative to the HU value for 100% blood.
As with the artery ROI, the vein ROI can be defined by the user or automatically by the
software.
In automatic mode the region of pixels with the largest blood volume (which is usually
located inside a vein) is used as a reference ROI.
Note: Since the largest vessel is usually a vein, we refer to a “vein ROI”. However, any
ROI located in a vessel large enough to be fully representative of the HU value for
blood (100%), and not affected by partial volume effects, is acceptable (even if it
happens to be located in an artery).

Advanced Settings
The protocols use default settings for certain parameters, such as maximum blood flow
value, hematocrit ratio, tissue density and the resolution used by the algorithm. For spe-
cific cases, you can adjust one or more of these parameters (see page 75).
Maximum Blood Flow Value
The algorithm limits the computed blood flow value to the value defined by this parame-
ter.
Normally, you should leave this parameter at the “normal” (default) value of 1000, to fully
take into account the occurrence of localized high blood flow values.
However, the earlier versions of this software (CT Perfusion 1) used a lower default value
of 134 for this parameter. If you have a requirement to compare results obtained using CT
Perfusion 2 with results from an earlier version, set this parameter to the value used with
that version.
Hematocrit Ratio
For adults the hematocrit ratio can usually be considered as a constant (default value
0.70). Adjust this setting where required, such as for exams of small infants where the
hematocrit ratio is usually higher (typical value 0.85).
Average Tissue Density
The perfusion algorithm initially computes blood volume, blood flow and permeability sur-
face area product per unit volume, then uses the average tissue density to normalize the
results to the values per unit mass.
The default value is 1.05. Adjust the setting if the average tissue density in the anatomi-
cal region being examined differs significantly from this value.

158
 CT Perfusion

Algorithm Resolution and Spatial Smoothing Factor

The Algorithm Resolution determines how many pixels from the reference image are
used for computing the functional maps. As an example, a resolution setting of 2 will
compute only every second line and every second column (one pixel in four). The result
is interpolated before the image is displayed.
The purpose of using a lower resolution is to speed up the computation. Relative to a
resolution setting of 1, a setting of 2 will speed up computation by a factor 4, a setting of 3
will speed it up by a factor 9.

The image data are smoothed before computation by means of the algorithm. This
reduces noise in the original image data, and the resulting functional map will be corre-
spondingly smoother.
The Spatial Smoothing factor determines the strength (size) of the gaussian filter func-
tion, or spatial averaging, applied to the image data. A factor of 1 means no smoothing,
while a factor of 10 means maximum smoothing, and corresponds to a 10x10 pixel spa-
tial averaging filter function. A larger spatial smoothing factor filters more of the noise,
but also reduces the visibility of small details. You can usually leave this parameter at the
default setting. The spatial smoothing factor does not affect the speed of the computa-
tion.

The two settings should not be confused. To summarize:

• Set the algorithm resolution for the trade-off between function view resolution and com-
putation speed that best meets your requirements,
• Adjust the spatial smoothing factor as a function of the amount of noise present in the
data.
The images in the series and function views can be displayed either smoothed (using a
bi-linear interpolation) or unsmoothed (pixel replication). This is a display function only
and has no effect on the computation or the resolution of the functional maps.

159
CT Perfusion

5 PIXEL VALUES IN SAVED IMAGES

When exploring saved CT Perfusion 2 images with other viewing applications (such as
the Basic Display Viewer or Volume Viewer), the pixel values in the saved images may be
different from the original values in the functional map.
UNDER NO CIRCUMSTANCES SHOULD THE PIXEL VALUES
FROM SAVED FUNCTIONAL MAPS BE USED BY ANY SOFT-
WARE APPLICATIONS THAT RELY ON HOUNSFIELD VAL-
WARNING! UES. THIS APPLIES IN PARTICULAR TO DOSE
COMPUTATION SOFTWARE APPLICATIONS.

Functional Maps
For saved functional maps that have been saved as processed images, the explanation
is the following:
When images are saved, the pixel values are stored on the image disk as integers.
To avoid loss of information when a functional map is saved, a scale factor is applied
whenever the values returned by the algorithm used are too small to be meaningful when
converted to integer values.
As an example, assume the computation of a ratio by an algorithm that returns a value
between 0.0 and 1.0. Storing the resulting functional map as an image with pixel values
of either 0 or 1 would obviously be of no use. Therefore the ratio value is multiplied by
100, resulting in a pixel range of 0 to 100.
When viewing such a saved functional map with another viewing application, you multiply
the displayed pixel value by a scale factor of 0.01 to obtain the ratio.

A scale factor is applied in the following cases:

Value returned by algorithm Multiply pixel value in To obtain:

saved image by:
Ratio (division) 0.01 Ratio
Blood Volume 0.1 BV in ml/100g tissue
Mean Time of Transit MTT in seconds
0.01
(MTT)
Blood Flow 0.25 BF in ml/100g/min
Permeability Surface Area PS in ml/100g/min
0.025
Product

160
 CT Perfusion

GLOSSARY

Note: Words in italics refer to terms defined elsewhere in the Glossary.

Absolute - The absolute value refers to the actual numerical value of the signal intensity.
Ratio refers to the absolute value divided by the average pre-enhancement value.
Active Annotation - A system annotation on a view that can be modified by the user to
control certain viewing parameters (e.g., window width and level), either by adjusting a
numerical value, or by selecting an item from a drop-down menu. Active annotations are
displayed in red.
Algorithm - A step-by-step process used to solve a problem. In CT Perfusion 2 this
refers to the mathematical function used to convert the acquired data into images.
Annotation - Generally, workstation-supplied text which accompanies an image when it
is displayed on-screen, describing when and how that image was acquired, with what
parameters. Also, text and ROI graphics added on a view by the user.
Area - In CT Perfusion 2, statistical measurement that represents the size of an ROI in
square millimeters.
Browser - The panel used in the application to select available images for display and
manipulation.
Cine Loop - An image display mode used for viewing a series of images at up to 30
images per second.
Color Ramp - A strip of boxes appearing to the left of a function view, showing the
shades of color (or shades of gray) used to represent the pixel values of the displayed
functional map.
Composite View - The overlay of a functional map onto a reference image to create a
composite image.
CT (Computed Tomography) - Process of deriving anatomical information by computer
synthesis of X-ray data, acquired by means of a CT scanner in the form of parallel
“slices”.
Cursor (1) - A graphic element on the monitor screen, which moves in unison with the
mouse’s movements as the mouse is moved on the mouse pad. The cursor provides the
user interface with monitor screen functions such as option selection, on/off control (“tog-
gle”), and cursor placement in type-in fields.

161
CT Perfusion

Cursor (2) - A vertical line appearing in a type-in field on the screen. This line represents
the location of keyboard characters (alphanumeric) to be typed in at the cursor location.
Placing the mouse cursor in a type-in field, then clicking the left mouse button, places the
cursor in the desired field. Characters typed in appear to the left of the cursor.
Cursor ROI - A cursor ROI is an ROI consisting of a specified number of pixels. The ROI
is considered to cover the same area as the cursor.
Default image - The image which is displayed when you select a new exam or a new
series. The default will be the first image in the center location of that exam or series.
Default value - A value supplied by the computer or pre-entered by the operator. This
value is retained if the operator answers a prompt by pressing <Enter>. The default can
be overridden by entering a different value.
Density Mask ROI - A technique to define an ROI by highlighting the desired range of
pixel values within the displayed image.
DFOV (Display Field Of View) - The real dimensions of a view (width and height) with
reference to the RAS coordinates.
DICOM - Abbreviation for Digital Imaging and Communications in Medicine. Standard for
the formatting and exchange of medical images and associated information.
Display matrix - The number of pixels in a displayed image, expressed in terms of num-
ber per axis °™ e.g., 512 x 512.
Display Normal - Function which resets a view to normal orientation and zoom factor.
Dynamic Data Set - Time ordered acquisition with either a sequential change in scan
time for each scan image (i.e., in contrast take up studies) or with a trigger delay (multi-
phase scan).
Dynamic Time Curve - A cursor ROI that plots the mean value of the pixels to the graph
view each time the cursor is moved.
Field of View (Acquisition FOV) - The area of the anatomy being imaged, usually
expressed in centimeters. FOV image size is a function of the acquisition matrix times the
pixel size.
Film Composer - Window used to lay out and film selected images.
Filmer - Function used to lay out and film or save selected images.
First Pre-Enhancement Image No. - Typically one of the initial images of the data set.
First Post-Enhancement Image No. - Typically seen as the first time point after the end
of a positive enhancement region.
Free-hand ROI - An ROI defined by “drawing” a free-hand trace on a view.
Function View - The views used to display the functional maps created with one of the
algorithms.

162
 CT Perfusion

Functional Map - A single image that is calculated from a set of time course images at a
single location. The image intensity values in the functional map represents the result of
a calculation process that generates a single valued parameter at each pixel location in
the original dynamic data set.
Graphics - User-defined ROIs.
Graph View - The upper right hand view, used to display the time curves, histogram and
ROI lists.
Histogram - An image analysis function that lets you create a bar graph representation of
the pixel value distribution in an ROI.
HU (Hounsfield Unit) - Scale unit denoting the density within a voxel in a CT data set.
Image - In this document the term “image” is used to designate the part of the exam data
being processed and displayed on the workstation screen. Depending on the display set-
tings, a view (q.v.) can display an entire image, or part of it (zoom).
Image Display Area - During use of a viewing application, the portion of the screen
where images are displayed.
Last Pre-Enhancement Image No. - Typically the point just prior to a positive increase
of the plotted time intensity curve.
Last Post-Enhancement Image No. - Typically one of the last points of the data.
Mean - Statistical measurement that represents the average pixel value in an ROI.
Maximum Slope of Increase or Decrease - For time course data that exhibits a tran-
sient increase, the Maximum Slope of Increase algorithm will return a value that charac-
terizes the maximum slope of increase. Transient increases that occur rapidly will have
larger Maximum Slope of Increase values and those occur gradually will have smaller
Maximum Slope of Increase values.
Similarly, the Maximum Slope of Decrease algorithm will return a value that characterizes
the maximum slope of decrease.
On-view menu - Menu displayed either on a view or on a particular feature of a view
such as a user annotation by pressing the right mouse button.
Patient List - The panel used in the AW 4.0 or later Basic Display application to select
available images for display and manipulation.
Pixel - Abbreviation for “picture element”, the smallest distinguishable component of a
digital image display.
Polygon ROI - An ROI defined by a set of points that are connected by the software with
straight-line segments. A spline ROI is the same ROI with curved sides.
Positive Enhancement Integral - Time course data acquired during the injection of a
contrast agent may have image intensity variations, which result in positive enhance-

163
CT Perfusion

ment. The Positive Enhancement Integral function characterizes such time intensity
changes by means of the integral of the area under the enhancement curve.
Rank - The rank number is the number assigned to the image when the images are
sorted by increasing time and is displayed in the lower left hand corner of the image in the
series view.
RAS - Abbreviation for Right/Anterior/Superior. Designation for the patient-linked coordi-
nate system used in CT data sets.
Ratio - An absolute value refers to the actual numerical value of the signal intensity. Ratio
refers to the absolute value divided by the average pre-enhancement value.
Rectangle - In CT Perfusion 2, on-view tool used to create and size a rectangle-shaped
region of interest (ROI).
Review Controller - On-screen tool to rapidly move through the images of an exam
using sliders and arrow buttons, or display them in cine mode.
Roam - Another term for “Scroll”.
ROI (Region Of Interest) - User-defined area of an image to be analyzed.
ROI List - Numerical list of pixel values in an ROI, displayed in the upper right hand view.
ROI lists can be saved as screen saves or in the form of a text file.
Scroll - An on-view control, used to view a particular part of an image by moving the
image around within a view, when the image has been enlarged (“zoomed”) and hence
the complete image no longer fits in the view.
Select - The act of placing the pointer on an option appearing on the screen, then clicking
the left finalize the option selection.
Series - Specific type of images within a study; a subset of an exam.
Series View - Upper left hand view, used to display the images from the current exam
series.
Spline ROI - An ROI defined by a set of points that are connected by the software with a
smooth curve (spline). A polygon ROI is the same ROI with straight sides.
Standard Deviation - In CT Perfusion 2, statistical measurement that provides a mea-
sure of variability of pixel values within an ROI.
System Annotation - An annotation on a view added by the system software, containing
data concerning the displayed image. Certain system annotations can be active, i.e.,
they can be modified by the user.
Text Annotation - A user annotation on a view containing text. Text annotations can be
used to add comments, or to add a legend to an anatomical feature on a view.
Threshold - A threshold number used to mask out background information.

164
 CT Perfusion

TIFF (Tagged Image File Format) - Image file format used to export screen captures to
other (non-AW) applications, e.g., to be used as illustrations for reports, publications or
documentation.
Title Bar - A bar at the top of a window that provides information about that window, and
also allows you to move the window to a different position on the screen by clicking and
dragging on it with the mouse.
Toggle - The act of switching a function from on to off, or off to on, with a single mouse
click of the pointer on the function’s button. Toggle buttons usually appear within windows
and other monitor screen areas, but some keys on the keyboard may also provide toggle
functions.
Type-in field - A specific area in a window on the screen which allows the user to type
commands or information into the system. Type-in fields are accessed using the mouse
pointer.
User Annotation - An annotation on a view added by the user, containing either text or
the result of a measurement.
View - Part of the workstation screen, used to display image data. The view area of the
CT Perfusion 2 screen contains four views. A view can display an entire image, or part
of it (zoom).
View Area - During use of a viewing application, the portion of the screen(s) where
images are displayed. The view area normally contains four views, but a single view can
be enlarged so as to take up the entire view area.
Voxel - Abbreviation for “volume element,” the basic element in a CT data set. A three-
dimensional region of an imaged object represented in two dimensions by a pixel.
Window (1) - Describes the range of pixel values that are assigned a shade of gray or a
specific color in the CT Perfusion 2 function images. Narrow windows offer greater reso-
lution and contrast of anatomy having similar densities. It also helps you find the values
for anatomy in which you are interested. See window width and level.
Window (2) - “Window” is also the term used for an on-screen graphical tool used to dis-
play information.
Window Width and Level (W/L) - In this context, “window” refers to the range of pixel
values within the image data, that is assigned a shade of gray for display. “Level” refers to
the center value. “Width” refers to the range of pixel values displayed around this central
level (the value corresponds to twice the number of intensities above and below the cur-
rently set level).
The adjustment is marginally similar to adjusting brightness and contrast controls: a nar-
row window (low width) translates to a high contrast of the display, and similarly a low
level translates to a high value of brightness.
Zoom - Function that allows you to shrink or enlarge a displayed image.

165
 Table of Contents
Navigator (optional)

CHAPTER 1 - INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1 OVERVIEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
2 CONVENTIONS FOR THIS MANUAL . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

CHAPTER 2 - START NAVIGATOR SOFTWARE . . . . . . . . . . . . . . . . . . . . 3

1 INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2 START NAVIGATOR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
3 CONTROLS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
3-1 Active annotations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
3-2 On-view menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
3-3 Protocol and control panels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
4 SET UP THE IMAGE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
5 EXPLORE WITH THE "NAVIGATOR" . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
5-1 Navigator markers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
5-2 The navigator symbol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
5-2-1 View direction controls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
5-2-2 Movement controls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
5-2-3 Keyboard controls (summary) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
5-2-4 "Flying" the navigator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
5-2-5 Rotate/Translate controls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
5-2-6 Navigator markers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
6 ADJUSTING THE NAVIGATOR VIEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
6-1 Rendering mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
6-2 Cut mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
6-3 Field of view . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
6-4 Reference image . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
Navigator (optional)

7 SYNCHRONIZED NAVIGATION MODE . . . . . . . . . . . . . . . . . . . . . . . . . . 16

7-1 Staring/stopping Synchronized Navigator. . . . . . . . . . . . . . . . . . . . . . . . . . 16
7-2 Synchro-Axial View . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
7-3 Synchro-Sagittal View. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
7-4 Synchro-Coronal View . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
8 DEFINE THE FLYTHROUGH PATH . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
8-1 Insert Step. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
8-2 Insert&Seek Step (recommended) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
8-3 Other controls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
9 SET MOVIE PARAMETERS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
9-1 Sampling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
9-2 Display speed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
9-3 multiview mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
9-4 Round trip . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
10 VIEW SEQUENCE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
10-1 View Sequence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
10-2 Viewing controls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
10-2-1 Pause/Restart and Step (<&>) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
10-2-2 Image Index Slider . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
10-2-3 Stop . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 23
10-2-4 Distance annotation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
11 SAVE. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
12 USING NAVIGATOR. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
13 ERROR MESSAGES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
 Navigator

CHAPTER 1 - INTRODUCTION

1 OVERVIEW
The Navigator software package is an optional extension of the CT application software.
The Navigator application allows you to move the viewpoint to locations inside the 3D vol-
ume. You can display the inside of “hollow” structures, providing real-time luminal views.
You can also view the outside of “solid” structures inside the 3D volume, in close-up and
in perspective. (With 3D imaging, you construct a 3D object that you can examine from
all directions, but the viewpoint is always located outside the object.)
To define what part of the 3D model will be displayed as “solid”, and what part will be dis-
played as “hollow”, you select a Navigator threshold mode (“Black in white”, “White in
black” or “Within borders”) and set the corresponding Navigator threshold(s).
By means of the record-and-playback feature, you can define a path through the anat-
omy, and view and record the resulting “fly-through” sequences. You can vary the view
aperture (from narrow-angle to wide-angle views).
Note: With the right settings, you can use Navigator to explore a 3D image set in a man-
ner very similar to endoscopy. However, Navigator is NOT a replacement for any
endoscopic or angiographic procedures, and is NOT approved as a screening
device.

WHEN USING NAVIGATOR SOFTWARE, AN INCORRECT

THRESHOLD SETTING OR INCORRECT THRESHOLD MODE
WARNING! CAN RESULT IN PATHOLOGIES OR OTHER ESSENTIAL
ANATOMY NOT BEING VISIBLE ON THE NAVIGATOR VIEWS.
ALWAYS CORRELATE THE NAVIGATOR VIEW WITH THE
ORIGINAL DATA (ACQUISITION VIEW, BASELINE VIEWS).

THE NAVIGATOR SOFTWARE USES A CONICAL PROJEC-

TION (PERSPECTIVE). FOR THIS REASON, THE ASSESS-
WARNING! MENT OF DIMENSIONS AND DISTANCES ON NAVIGATOR
VIEWS TENDS TO BE SUBJECTIVE.
ALWAYS CORRELATE APPARENT DIMENSIONS AND DIS-
TANCES ON THE NAVIGATOR VIEW WITH THE ORIGINAL
DATA (ACQUISITION VIEW, BASELINE VIEWS).

The Navigator software is not approved as a screening device.

CAUTION

1
Navigator

The Navigator software procedures have not been demonstrated

CAUTION to be a replacement for any endoscopic or angiographic proce-
dures.

2 CONVENTIONS FOR THIS MANUAL

Note: The text in this manual uses certain type styles and symbols to differentiate
between one tool or graphic and another. This manual uses the following conven-
tions:
• Menu titles appear in bold face (example: Application Menu)
• Menu options appear in bold face, within brackets (example: [exit])
• On-screen tools appear in bold face, within braces (example: {Scroll Bar})
• Graphical buttons appear in bold face, within parentheses (example: (View))
• On-screen prompts and messages appear in italics (example: Login:)
• User typed-in responses appear in bold face italics (example: root)
• Keyboard hardkeys and mouse buttons are underlined (example: Return or left)
• This manual calls the 3-dimensional, computer generated luminal image, the "Naviga-
tor model" or "Navigator image."
• This manual calls the [Navg] or [Navg Smooth] view type, the "Navigator" view.
• This manual calls the graphic representation of the Navigator in each view type the
"Navigator indicator."

2
 Navigator

CHAPTER 2 - START NAVIGATOR SOFTWARE

1 INTRODUCTION
The Navigator application generates 3D images of anatomical structures from a 3D
image set. It uses a conical view, as if a camera was placed at a viewpoint inside the
structure.
View layout and view management is the same as for other applications, with a “Naviga-
tor” view at top left and baseline and oblique images in the other three views.
To move and point the “camera” you use the “navigator” symbol on the Navigator view.
Various controls allow you to “steer” the navigator up/down and left/right, and move it for-
ward or back.
To initially move the viewpoint to any position in the image set and roughly set the view
direction, you use small simplified navigator symbols (the navigator markers) on the
baseline and oblique views.
When using the Navigator application, you will normally go through the following steps:
• Select the exam/series in the Image Works Browser, click the button of “VA” then start
the Volume Analysis 2 application, and select Navg Guide in the protocol list (see
paragraph 2),
• Set the threshold mode and threshold(s) to obtain a satisfactory image (see paragraph
4),
• Explore the anatomical structure of interest with the navigator (see paragraph 5),
• Adjust the image settings as necessary (see paragraph 6),
• Define a FlyThrough path (see paragraph 8),
• Set the movie parameters (see paragraph 9),
• View the sequence (see paragraph 10),
• Save the sequence (see paragraph 11).
Note: In this chapter, “Navigator” (with a capital) refers to the application, while “naviga-
tor” refers to the navigator symbol on the Navigator view.

3
Navigator

2 START NAVIGATOR
To start using the Navigator application:
• Select the exam/series in the Image Works Browser.
The image set should meet the same requirements as for the other Volume Analysis 2
applications. See Chapter 4.
• Start the Volume Analysis 2 application, then select [Navg Guide] in the protocol list.
− The Navg Guide protocol uses low resolution and is the preferred choice when explor-
ing larger anatomical structures and when setting up “fly-through” sequences: pro-
cessing is faster and small irregularities and “noise” tends to be smoothed out.
Alternatively, you can create a Navigator view using the (Display Tools) > (View Layout)
controls.

3 CONTROLS
Apart from the “navigator” symbols mentioned in paragraph 1 and further described in
paragraph 5, the Navigator application uses the same types of controls as the other Vol-
ume Analysis 2 applications: active annotations, on-view menus, protocol panels and
control panels.

3-1 Active annotations

The active annotations on the Navg. view that are specific to the Navigator application
are illustrated below. They are displayed in red as in the other Volume Analysis 2 appli-
cations.
Illustration 1 - Active Annotations on Nave. View

Navg. P 155 set aperture

Ex: 1234 (30.0 to 120.0 in standard mode)
Se: 3 (120.0 to 360.0 in fisheye mode)
set rendering mode Smooth -524 ^ 90.0
(Fast, Smooth, Integral or VR)

adjust threshold
Cut off
Cut mode set threshold mode
^
(Off or On) ( ^ or or 0 )

4
 Navigator

A Navigator view contains the following active annotations:

• Smoothing mode: select Fast, Smooth, Integral or VR (see Section 6, paragraph Ren-
dering mode below),
• Threshold(s): increment or decrement current value (see paragraph 4 “Set-up the
Images”),
• Threshold mode: select [Black in White], [White in Black] or [Within Borders] from
the drop-down menu (see paragraph 4 “Set-up the Images”),
• Aperture size: Increment or decrement current value,
• Cut mode: select Off or On. If the Cut mode is On an additional active annotation is dis-
played at the end of the line, representing the Distance between the Navigator point of
view and the Cut.
• Patient Name (top right on the view, not shown in Illustration 1): show or hide the
patient name field.
You use the Navigator active annotations in the same way as those for other view types.
See Chapter 6 “Display and Controls”.
Briefly:
• To change a numerical active annotation:
- Point on it, hold the middle mouse button down and drag left or right, or
- Point on it and click with the left or right mouse button to change it in steps, or
- Point on it, type a new value on the keyboard and validate with <Enter>
(remember to enter a - for negative values).
• To change a text or symbol active annotation, hold the left or right mouse button down,
and select the desired option from the drop-down menu.
The various controls are described further in this chapter as applicable.

5
Navigator

3-2 On-view menu

Additional Navigator functions can be accessed through the Navigator on-view menu.
Click and hold right on the Navigator view (away from the navigator symbol or any active
annotation) to display the menu, then select the desired function.
[Save image] saves the current view in a separate series in the current exam.
[Light/contrast] opens the Navigator Shading panel, that allows you to change the
brightness and contrast of the Navigator view (Smooth mode only).
[Turnaround] turns the viewing orientation 180 degrees around the vertical axis of the
view (reverses direction of the Navigator).
[Create trace] and the associated [Clear last point], [Clear trace] and [Lock cursor to
trace] menu items can be used to create and edit traces on the Navigator view, as
described in Chapter 6 “Display and Controls” paragraph 5.
If you take into account that on a Navigator view the 3D cursor is placed on the “wall” of
the vessel or other anatomical feature being displayed, you can, in theory, create and
edit traces, e.g., to perform measurements. In practice, you should use this feature
with care, because on the Navigator view you have no indication how “deep” into the
3D object the 3D cursor is located, unless you continuously correlate the Navigator
view with the baseline views. The result may not be at all what you wanted or
expected.
Also refer to Chapter 6 “Display and Controls” paragraph “3D cursor on 3D views”.
[Enlarge] increases the Navg. view to full screen (4x) size. The menu option changes to
[Reset size] which decreases the Navg. view to the original size.
[Lock orientation] prevents rotation and limits Navigator control to forward/backward,
up/down/left/right motion. The menu option changes to [Unlock Orientation] which
restores rotation control to the Navigator.
[Center on cursor] centers the view on the current position of the 3D cursor.
[Reset pointer] returns the 3D cursor, and the navigator with it, to its initial display posi-
tion in the center of the image set.

3-3 Protocol and control panels

If you start to use Navigator via the Navg. Guide protocol, the protocol panels for proce-
dures such as setting up and viewing a FlyThrough sequence are displayed automati-
cally.
You can remove them from the screen by means of the (Close) button at the bottom of
each panel, then return them to the screen by clicking on the (Navigator) protocol button
in the main control panel on the left of the screen.
Alternatively, select the function you require from the (Navg. Tools) menu.
The controls in the Navigator protocol and control panels are described in the following
paragraphs.

6
 Navigator

4 SET UP THE IMAGE

At startup the Navigator screen consists of the Navigator view (top left), and the three
baseline views, axial, sagittal and coronal. You can change any of the baseline views to
an oblique view if required.
On the baseline and oblique views the navigator viewpoint (which coincides with the 3D
cursor) is indicated by a yellow dot, and the view direction by a short yellow line.
The Navigator view shows a three-dimensional image. However this image may need to
be set up in threshold and/or threshold mode.
Go through the following steps to set up the Navigator view:
• On the baseline and oblique views move the 3D cursor inside the structure of interest
(the navigator viewpoint follows the 3D cursor). Point the view direction marker
approximately the right way (click and drag on the end of the short yellow line to rotate
it around the viewpoint).

• Open the threshold mode drop-down menu using the active

Black In White annotation on the Navigator view (see paragraph 3), and select
White In Black
the threshold mode:

Within Borders

- [Black in White] “Black in white” mode ( annotation) if on the baseline and

oblique views the lumen (the structure that is to be displayed as hollow) is
darker than the surrounding tissue. An example would be an airway.
- [White in Black] mode ( annotation) if the lumen (the structure that is to be
displayed as hollow) is lighter than the surrounding tissue (e.g. a contrast-
enhanced vessel).
- [Within Borders] mode (0 annotation) if the lumen (the structure that is to be
displayed as hollow) is defined by a specific range of voxel values, or, in other
words, is surrounded by both lighter and darker structures that should be
displayed as solid.
An example is a contrast-enhanced vessel with calcifications where both the
surrounding soft tissue (darker) and the calcifications (brighter) should be
displayed as solid.
• Now adjust the threshold (or thresholds in the case of the “Within borders” mode) using
the active annotation(s) on the Navigator view, until you obtain a satisfactory visualiza-
tion of the feature of interest.

7
Navigator

The easiest way to adjust a threshold annotation on the Navigator view is to click and
drag to the left and right on it (also see paragraph 6).
If you use this method, the range of voxels that will be displayed as hollow is indicated
on the baseline and oblique views by green hatching for as long as you keep the
mouse button down.
Note: The software attempts to choose a reasonable threshold when the threshold mode
is changed.
Once you have obtained an initial image in the Navigator view, you still must adjust the
threshold(s) to more accurately match the voxel values of the structures you want dis-
played as hollow and solid, respectively (also see paragraph 12).
You can now start “navigating” in the structure, as described in the following paragraphs.
Note: In terms of voxel values:
- The [Black in White] mode ( annotation) displays all voxels with values
above the threshold as solid,
- The [White in Black] mode ( annotation) displays all voxels with values
below the threshold as solid,
- The [Within Borders] mode (0 annotation) displays all voxels with values
outside the threshold range as solid.
On the view, voxels below the lower threshold are differentiated from those
above the upper threshold by the use of two different colors for the surface
shading. You can use the (Display Tools) > (View Colors) controls to set
the color to be used for all objects above the upper threshold.
Note: Selection of threshold mode and settings depends on various factors, such as
type of exam, type of anatomical structure, acquisition parameters, etc. Some
suggested settings for CT exams are shown below.

Type of Structure Threshold Mode Threshold Value

Lungs, ENT Black in White -500
Colon Black in White -800
Skin Black in White -500
CTA Vessels White Borders Variable
Bone Black in White 160

WHEN NAVIGATOR SOFTWARE, AN INCORRECT THRESH-

OLD SETTING OR INCORRECT THRESHOLD MODE CAN
WARNING! RESULT IN PATHOLOGIES OR OTHER ESSENTIAL ANAT-
OMY NOT BEING VISIBLE ON THE NAVIGATOR VIEWS.
ALWAYS CORRELATE THE NAVIGATOR VIEW WITH THE
ORIGINAL DATA (ACQUISITION VIEW, BASELINE VIEWS).

8
 Navigator

5 EXPLORE WITH THE “NAVIGATOR”

Once you have set up the Navigator, you can explore the anatomical structures.
To move around inside the anatomical structures, you use the “navigator” markers on the
baseline and oblique views, and the “3D cube” on the Navigator view.
The navigator markers are the yellow dot-and-line markers on the baseline and oblique
views, that indicate the position of the current view point, and the projection of the view
direction on the view. See Figure 4 below.
The 3D cube is the red symbol in the shape of large cross-hairs with attached handles on
the Navigator view that you can use to “fly” through the structures, just in the same way
as in any 3D view. The 3D cube is displayed whenever you move the mouse pointer onto
the Navg. view.
The functions of the navigator markers and the navigator are complementary:
- You can move the navigator markers on the baseline and oblique views
rapidly to any location in the image set, but it is not easy to move the
viewpoint marker accurately over a small distance, or point the view direction
marker precisely in a given direction.
- On the other hand, you use the navigator to point and advance accurately in
the anatomical structure of interest, but you cannot use it to move rapidly to a
completely different location in the image set.

9
Navigator

5-1 Navigator markers

You use the navigator markers to move rapidly to any location in the image set, and point
the view direction approximately the right way.

Fig. 4: Navigator markers on axial, sagittal and coronal views

• Move the navigator viewpoint marker in the same manner as you move the 3D cursor:
- Click and drag on the 3D cursor and yellow dot of the navigation marker, or
- Point with the mouse to the new position and press the <Shift> key.
• To point the view direction marker:
- Click and drag on the end of the short yellow line to rotate it around the
viewpoint.

10
 Navigator

5-2 The navigator symbol

You use the “navigator symbol” on the Navigator view to move and point the “camera”
inside the anatomical structure of interest.
The red navigator symbol is displayed whenever you move the mouse pointer onto the
Navg. view.
With the various navigator controls you can move the view direction (line of sight) up,
down, left or right, and move the viewpoint (the position of the camera) forward or back
along the line of sight.
Navg. P 155 Hospital Name
Ex: 1234 PatientName
Se: 3 Sex Age
Smooth-524 ^ 45.0 Date
Navigator symbol

L L
1 1
0 2
1 3D cursor 1

Handles

P 135

5-2-1 View direction controls

To point the navigator in a new direction:
• Click and drag on any one of the handles of the navigator symbol (see Figure 5 on pre-
vious page).
If you want to rotate the view:
• Use the (Rotate/Translate) > (By deg) controls.
When you are changing the view direction, releasing the mouse button for a moment re-
aligns the navigator symbol with the new view direction.
To align the view with the axis of the object:
• Press the <L> key on the keyboard.
This points the current view toward the most remote portion of the image, which aligns
the view with the main axis of the object.
You can use this function to realign the view direction after moving the navigator manu-
ally, and to negotiate sharp turns in the sequence.

11
Navigator

5-2-2 Movement controls

To move the viewpoint forward in steps:
• Click and drag (with the left mouse button) on any one of the handles of the navigator
symbol as described above to set the view direction, then click on the middle mouse
button while still holding down the left mouse button.
Alternatively, to move the viewpoint:
• Press the <F> key on the keyboard to move forward in steps.
• Press the <B> key on the keyboard to move backwards in steps.
In particular, if you inadvertently move the navigator “inside object” (hit the “wall”), the
<B> key on the keyboard allows you to step back along the current view direction, until
you are back in the "hollow" part of the structure.
To align the view with the axis of the object, then move the viewpoint forward:
• Press the <P> key on the keyboard.
Note: The step size is determined by the setting (in mm) in the (Rotate/Translate) > (By
mm) control panel.
You may want to make sure that the step size is not set to zero: the navigator for-
ward and backward movement will appear to be blocked in that case.
Note: For more information on this function, refer to Chapter 6 “Display and Controls”,
Section 6 “Smart Cursor”.

5-2-3 Keyboard controls (summary)

As already noted in the previous paragraphs, you can use keyboard shortcuts to move
the navigator, and align it with the main axis of the object. When using any of the key-
board shortcuts, make sure that the mouse pointer is on the Navigator view.
• <F> (“Forward”) moves the navigator forward by one step,
• <B> (“Back”) moves the navigator backwards by one step,
• <L> (“aLign”) aligns the view direction with the axis of the object,
• <P> (“Path”) aligns the view direction with the axis of the object, then moves the navi-
gator forward by one step.
Note: The <S> key is used to save the current view to the image disk. See paragraph 11
“Save”.

12
 Navigator

5-2-4 “Flying” the navigator

You can combine the use of the controls as described above to “fly” the navigator through
the structure of interest.
You “steer” the navigator by holding and dragging with the left mouse button on one of the
handles while at the same time each click on the middle mouse button steps you forward.
Alternatively you can “steer” the navigator with the mouse and use the <F> and <B> keys
on the keyboard to move forward and backward.
With some practice you can “fly through” and explore a structure easily and accurately.
Note: Rather than clicking each time on the middle mouse button to step forward, you
can even “fly” the navigator through the anatomy in a near-continuous manner by
holding down both left and the middle mouse button.
Releasing the mouse buttons stops the navigator motion.

5-2-5 Rotate/Translate controls

As already described in Chapter 8 “Reformatting” and Chapter 9 “3D Imaging”, you can
use the controls in the Rotate/Translate control panels to change view angle and move
the 3D cursor.
Since the navigator is closely linked to the 3D cursor, you can also use these controls to
point and move the navigator.
You can use the (By deg) controls to change the view direction in steps, and the (By
mm) controls to move the view point in steps.
To return the view direction to one of the six standard orientations (Superior, Inferior,
Anterior, Posterior, Left or Right).
• Use the (S) (I) (A) (P) (L) (R) buttons on the main control panel.

5-2-6 Navigator markers

As already noted, moving the navigator markers on a baseline or oblique view moves the
viewpoint and view direction on the Navg view. When you move the mouse pointer back
on the Navg view the position and orientation of the navigator symbol will reflect the new
viewpoint and view direction.

13
Navigator

6 ADJUSTING THE NAVIGATOR VIEW

6-1 Rendering mode

Three-dimensional image processing tends to produce complex images, and Navigator is
no exception. Because of this, Navigator uses four different types of image rendering
modes:
• two surfacic rendering modes:
- a Fast mode to explore the image set and quickly move to an area of interest
but without displaying a high level of detail. It is mainly useful to explore very
small objects.
- a Smooth mode to produce high-quality rendering of the structures in the area
of interest; it is mainly useful to explore bigger objects, such as airways or the
colon.
• two volumic rendering modes:
- an Integral mode, which produces a gross and quick volumic rendering of the
studied area. It displays the pixels with a certain penetration thickness. The
thickness appears on the view at the end of the line of the Integral
annotation. The value is an active annotation and can therefore be modified.
- a standard VR mode.
You use the rendering mode active annotation (see illustration in paragraph 3) to switch
between Fast, Smooth Integral or VR modes.
Note: The Smooth rendering mode should not be confused with the Navg. Smooth pro-
tocol at startup (see paragraph 2). Both have a smoothing effect on the display,
but for entirely different reasons. The Navg. Smooth protocol uses a lower resolu-
tion to display the view, while the Smooth rendering mode acts by filtering out
jagged edges and contour effects from the image.
Note: If the amount of image data in the field of view exceeds certain limits the software
automatically switches to a coarse low-resolution display mode to accelerate the
updating of the Navg view while you are modifying the view point or view direction.
Once you have made the changes, a “Hit spacebar to display high definition view”
message is displayed. Press <Space>on the keyboard to return to the higher res-
olution view.

14
 Navigator

6-2 Cut mode

By default the Cut mode is off.
The Cut on visualization mode allows you to cut a cylinder within the 3D view with a
given depth and a given diameter. It allows you to see what lies behind a superfluous
structure, for example to see the lung from inside the bronchis, or to study what is at the
bottom of the cylinder, which is displayed as a 2D view. Window level is available on this
2D view.

6-3 Field of view

The Navigator view is a conical view. By default the aperture of the view cone (view
angle) is 90o .
You can modify the field of view by adjusting the aperture active annotation (see illustra-
tion in paragraph 3) from 30.0o (maximum “zoom” effect) to 120.0o (wide-angle view) and
from 120.0o to 360.0o (Fish-eye mode).

THE NAVIGATOR SOFTWARE USES A CONICAL PROJEC-

TION (PERSPECTIVE) THAT CAN INTRODUCE DISTOR-
WARNING! SIONS, ESPECIALLY FOR FISH EYE VIEWS. SUCH VIEWS
SHOULD ALWAYS BE USED IN CORRELATION WITH 2D
BASELINE VIEWS.

As with the DFOV (Display Field of View) function, when you decrease the aperture
value, you narrow the field of view and enlarge anatomy.

As soon as you set the aperture value above 120.0o , you automatically toggle into Fish-
eye mode. A red circle appears on the view,visualizing the value of the aperture at this
spot on the image. At first it represents the 120.0o aperture, then another circle appears
when you reach the 180.0o aperture, next the aperture shifts while the image grows
smaller. The fish-eye effect allows you to see what is behind your point of view, as if you
were progressively turning a glove inside out.

6-4 Reference image

The small square reference image on the Navigator view (normally in the bottom right
corner) shows the projection of the pyramid-shaped view cone on one of the baseline
views (by default the axial view). The base of the pyramid represents the field of view,
the tip of the cone corresponds to the current viewing location.
A pop-up menu (click and hold right on the reference image) allows you to:

15
Navigator

• Select which baseline view is used as a reference (axial, sagittal or coronal),

• Move the reference image to another corner of the Navigator view,
• Adjust how much of the baseline view is shown in the reference image (magnify/
unmagnify).

7 SYNCHRONIZED NAVIGATION MODE

At startup the Navigator screen consists of the Navigator view (top left), and the three
baseline views, axial, sagittal and coronal. These baseline views are calculated accord-
ing to the acquisition plane of the original images.
Activating the Synchronized Navigation mode means that the baseline views represent
instead views calculated in respect with the direction of the Navigator.
Note: When in Synchronized Navigation mode, the navigator markers can no longer be
used to change the viewpoint, as the views are centered on them. To do that, you
must act on the navigator symbol itself.

7-1 Starting/stopping Synchronized Navigation

You start the Synchronized Navigation mode by activating the [Start Synchronized Nav-
igation] function of the Navigator on-view menu. The menu option changes to [Stop
Synchronized Navigation] which allows the Navigator to return to standard baseline
views.

7-2 Synchro-Axial View

When the Synchronized Navigation mode is selected, the first 2D view toggles to the
oblique view transversal to the direction of the Navigator.

7-3 Synchro-Sagittal View

When the Synchronized Navigation mode is selected, the second 2D view toggles to the
oblique view parallel to the direction of the Navigator, in th evertical plan.

7-4 Synchro-Coronal View

When the Synchronized Navigation mode is selected, the third 2D view toggles to the
oblique view parallel to the direction of the Navigator, in the horizontal plan.

16
 Navigator

8 DEFINE THE FLYTHROUGH PATH

Paragraph 5 described how to view and move through the anatomical features of interest
step by step using the navigator.
You can also define a path through the structure, and have the software calculate a “Fly-
Through” movie sequence. You can then view this sequence repeatedly, pause and
restart it, modify the display speed, and/or save the sequence as a separate series.
You do not need to define each view in the sequence, but only the points (steps) on the
trajectory of the “movie camera” where the view direction changes. The Navigator soft-
ware interpolates frames between the designated locations to smooth out the FlyThrough
sequence. The software accepts up to 256 defined steps per FlyThrough sequence.
To prescribe a path through the structure, first activate the FlyThrough sequence controls
(via (Navg Tools) > [Navg Path] or from the Navigator protocol). You can now insert
steps in the FlyThrough sequence prescription, placing the steps either manually with the
(Insert Step) command or semi-automatically with the (Insert & Seek Step) command.
Note: See paragraph 14 for a list of possible error messages.

THE NAVIGATOR SOFTWARE USES A CONICAL PROJEC-

TION (PERSPECTIVE). FOR THIS REASON, THE ASSESS-
WARNING! MENT OF DIMENSIONS AND DISTANCES ON NAVIGATOR
VIEWS TENDS TO BE SUBJECTIVE.
ALWAYS CORRELATE APPARENT DIMENSIONS AND DIS-
TANCES ON THE NAVIGATOR VIEW WITH THE ORIGINAL
DATA (ACQUISITION VIEW, BASELINE VIEWS).

8-1 Insert Step

(Insert Step) adds the current location of the navigator (view point and view direction) to
the path prescription. You use this command to insert the starting point for the sequence,
then to define steps in the path prescription manually.
• You can add locations to a prescription in progress. The software automatically inserts
the step into the correct place in the sequence.
• The software will warn you if you try to insert a new step too close to an existing step in
the current sequence.

17
Navigator

8-2 Insert & Seek Step (recommended)

(Insert & Seek Step) adds the current location of the navigator to the path prescription,
moves 1 cm to the next location in the current viewing direction, and then adjusts the
view to align it to the axis of the structure. You use this command to quickly prescribe a
path through the anatomy of interest.
Note: If you insert a step that flies the Navigator too close to an object, the software
automatically selects a smaller step, to avoid collision.

8-3 Other controls

(Delete Step) removes the current step from the prescription.
(Backtrack) allows you to move back to the last prescribed step of the path.
(Align to Axis) rotates the current view toward the most remote portion of the image,
which aligns the view with the main axis of the object. You can use this function to realign
the view direction after moving the navigator manually, and to negotiate sharp turns in the
sequence.
Note: You can also use the <L> key on the keyboard.
(Show/Hide Path) controls whether the prescribed path is displayed on the non-Navg.
views.
The {Step slider} indicates the current location in the FlyThrough sequence. As soon as
you have defined more than one point, you can use this slider to rapidly move to a partic-
ular step in the sequence, e.g. to insert an intermediate step or to delete a step. The dis-
play updates accordingly.

18
 Navigator

9 SET MOVIE PARAMETERS

Before actually viewing a “FlyThrough” movie sequence, you must set up the movie
parameters:
• Sampling distance,
• Display speed,
• Multiview mode on/off,
• Round trip on/off.

9-1 Sampling
Use the {Sampling} slider to select the default distance, in millimeters, between each
view in the FlyThrough sequence, which helps determine the number of images that the
software interpolates between each step.
You can adjust the default distance between 0.5mm and 20.0mm. The software will auto-
matically adjust the actual distance between views where necessary to optimize the Fly-
Through path.
You must set this control before using (Navg Options) > [Navg Cine] > [Fly through] >
(View Sequence) to compute and view the FlyThrough sequence.
Note: The setting of the {Sampling} slider determines approximately how many images
there will be in the sequence (number of images = length of path divided by sam-
pling distance).
You should make sure the {Sampling} slider setting corresponds to your require-
ments. Too small a sampling distance increases the time required to compute and
display the sequence, without necessarily contributing additional information.
As an example, using the minimum sampling distance of 0.5mm over a 10cm path
will produce 200 images. Increasing the sampling distance to 1mm reduces this
to 100 images, and halves both the computing and the display time of the
sequence.
Note: If the sequence is too large for the available workstation memory this will be indi-
cated by a warning message. When this happens you can:
• Increase the sampling distance, to reduce the number of images in the sequence,
and/or
• If you have used the Navg Guide protocol to build the 3D model (see paragraph
2), or
• Accept the sequence “as is”: the sequence will be displayed, but much more
slowly, because the system will need to swap images between memory and image
disk.

19
Navigator

9-2 Display speed

Use the {Display speed} slider to set the speed in images/sec at which the view
sequence will be displayed. The default setting is 10 images/sec, but you can adjust it
between 1 to 30 images/sec.
You can adjust this control while the FlyThrough sequence is being displayed.

9-3 Multiview mode

Multiview mode simultaneously updates the images in up to four views during the FlyTh-
rough sequence, screen saving or filming. This feature changes the Navg. view to a four-
on-one Multiview format that displays reduced-size versions of the Navg., Sagittal, Coro-
nal, Axial, or selected 2-dimensional views. During the FlyThrough sequence,
the reduced-size 2-dimensional views update in sync with the 3-dimensional Navigator
image. The normal-sized Axial, Sagittal and Coronal planes do not update until you
pause or stop the FlyThrough sequence.
• When you pause the FlyThrough sequence for more that a few seconds, the normal-
sized Navg. image replaces the four-on-one Multiview image.
• When you restart the sequence, the four-on-one Multiview image replaces the normal-
sized Navg. image.
• The Multiview image returns for a moment when you step forward and pause, or step
backward and pause. The view changes back to the Navigator image, and the other
three views also update, each time you step and pause.
You must select whether to use Multiview mode BEFORE using (Navg Tools) > [Navg
Cine] > [Fly through] > (View Sequence) to compute and view the FlyThrough
sequence (the system ignores the state of the (Multiview mode) control afterwards).
You select or cancel Multiview mode from the [Navg Tools] > [Navg Cine] panel.
Note: The system requires significantly more time to compute a Multiview mode
sequence. The software displays and updates messages (“xx images of xx to
compute” and “xx min xx sec remaining”) to keep you informed of its progress.

9-4 Round trip

• To display the view sequence from first to last image, then from last to first (forward/
backward), set (Round Trip) to off.
• To display the view sequence from first to last image, then start again with the first
image (i.e., as a loop), set (Round Trip) to on.
You must set this control BEFORE using (View Sequence) to compute and view the Fly-
Through sequence.

20
 Navigator

10 VIEW SEQUENCE
Once you have defined the path (see paragraph 8) and set the movie parameters (see
paragraph 9), you can compute, view and save the FlyThrough sequence.
You should note that you cannot modify a FlyThrough sequence. However, you can at all
times stop the FlyThrough sequence display, and modify the path definition, or change
the movie parameters, then recompute the sequence.

10-1 View Sequence

Select (Navg Tools) > [Navg Cine] > [Fly through] > (View Sequence) to start the com-
putation of the images that compose the sequence.
During the computation a small pop-up window indicates progress. The (Stop) button in
this window allows you to interrupt the computation if necessary.
As soon as all images have been computed, the FlyThrough movie sequence starts auto-
matically.
Note: If you have selected Multiview mode (see paragraph above) the system requires
significantly more time to compute the sequence. In this case, the software dis-
plays and updates messages indicating “xx images of xx to compute” and “(xx min
xx sec remaining)” to keep you informed of its progress.
Note: During the computation of the sequence a warning message may indicate that the
sequence will "hit the object" (run into the 3D surface) if it follows your current Fly-
Through sequence prescription.
You can accept this: in that case the software will fly into the 3D surface, per your
request, and "Inside object" will be displayed in the view sequence for the section
of the path inside the 3D object.
Otherwise, return to the Navg path > Define Path controls (see paragraph 8),
delete one or more steps as necessary and insert one or more steps close to the
current position to prevent the collision, then select (Navg Tools) > [Navg Cine] >
[Fly through] > (View Sequence) again to restart the computation of the
sequence.

21
Navigator

10-2 Viewing controls

Once the FlyThrough view sequence is computed, you can use the viewing controls:
• (Pause/Restart) button,
• {Image Index} slider,
• (Stop) button.

10-2-1 Pause/Restart
• (Pause) simply pauses the FlyThrough movie display. The button label changes to
(Restart), and a second click on the button resumes the FlyThrough movie display,

10-2-2 Image Index Slider

Grab the {Image Index} slider to pause the FlyThrough movie display (or use the
(Pause) button).
Click and hold on the slider, then drag it to the left or right to to display the corresponding
image in the sequence.
Note: When using the (Pause/Restart) and step buttons, or the {Image Index} slider:
- After you step forward or step backward and pause, the software updates all
the non-FlyThrough views about 2 seconds later.
- In Multiview mode, the Multiview image returns for a moment when you step
forward and pause, or step backward and pause. The view changes back to
the Navigator image, and the other three views also update, each time you
step and pause.
- You can pause the FlyThrough movie display, and manipulate the navigator
on the Navg. view. When you restart the sequence, the software
automatically returns to the original pause point, and resumes the FlyThrough
movie display.

22
 Navigator

10-2-3 Stop
Once the view sequence is computed, it can no longer be modified. The label on the
(View Sequence) button changes to (Stop).
You use this control to stop the FlyThrough movie display.
This allows you to return to the Define Path controls (paragraph 8) and modify the path,
or to modify the movie parameter settings (paragraph 9), then recompute the sequence.
You also must stop the FlyThrough movie display if you want to save the FlyThrough
sequence on the image disk (see paragraph 11).
• Click on (Stop) to stop the FlyThrough movie display. Once you stop the display, you
can no longer use (Restart) to resume the display.
The label of the (Stop) button changes back to (View Sequence). When you reselect
this button, the view sequence is recomputed entirely.

10-2-4 Distance annotation

Once the view sequence is computed, a distance annotation is added on the view (near
top left) which indicates the distance from the starting point along the fly-through path.
This is particularly useful while using (Pause) and (Step) to accurately locate a particular
view along the path.

23
Navigator

11 SAVE
After you have prescribed a FlyThrough sequence and used (View Sequence) to com-
pute the views in the sequence, you can save individual images from the sequence, or
save the entire current FlyThrough sequence as a new series of screen save images.
• To save the entire FlyThrough sequence, (Stop) the display. You can now use (Save
Sequence) to save the current FlyThrough sequence as a new series of screen save
images.
Note: To save Navigator movie sequences on the image disk, make sure beforehand
that you have adequate free disk space available. A FlyThrough sequence may
contain several hundred images, hence it can take up a significant amount of disk
space.
Note: Once you (Stop) the display, you can no longer (Restart) it.

24
 Navigator

12 USING NAVIGATOR
With the appropriate settings, you can use Navigator to explore a 3D image set in a man-
ner very similar to endoscopy. However, as noted, Navigator is NOT a replacement for
any endoscopic or angiographic procedure, and is NOT approved as a screening device.
When we use the terms “hollow” and “solid” you should keep in mind that these are artifi-
cial. What appears in the Navigator view as “solid” and what appears as “hollow” is
totally determined by the threshold mode and threshold setting.
As an example, take a contrast-enhanced vessel that appears bright among the soft tis-
sue that surrounds it.
With “white in black” mode and the right threshold setting, the “inside” of the vessel will be
displayed as hollow and the Navigator allows you to explore this “inside”.
The wall of the vessel, and the vessel contents, are not of uniform density, so when you
increase the threshold the vessel will appear narrower. Increase the threshold even more
and the vessel will appear to start filling with floating lumps. Physically these lumps are
real: they indicate the statistical variations (noise) of the density of the fluid filling the ves-
sel, but anatomically they have no significance.
On the other hand, lower the threshold too far, and “holes” will start appearing in the
“walls” of the vessel because the software starts “seeing” part of the vessel walls and the
surrounding soft tissue as “hollow”. Again, the holes you see are not an anatomical real-
ity.
In the same data set, switch to “black in white” mode, move the Navigator viewpoint out-
side the vessel, and adjust the threshold as necessary. Now the soft tissue is treated as
“hollow”, the vessel as “solid”, and you’re looking at the “outside” of the vessel! Again,
adjusting the threshold determines what exactly you will see.
In other words, the threshold mode and threshold setting give you complete control over
what is displayed as “solid” and what is displayed as “hollow” in the Navigator view. You
could even display solid bone as a hollow space, if that was what you wanted!
Because it gives you such complete control over the viewpoint and over what is displayed
(through the threshold mode and threshold setting), Navigator can be a powerful tool for
exploring details in a 3D data set, in a manner not available from other applications.
BUT, you should use it with common sense. You should remain aware that what is being
displayed is totally determined by the display parameters you have chosen.
Navigator allows you to “look” at a 3D data set in a manner unlike other applications.
This can give you valuable added insights during a diagnosis. Nevertheless you should
always correlate what you see on the Navigator view with the original data (acquisition,
baseline and oblique views).
Because of the use of perspective and varying view angles, the assessment of dimen-
sions and distances on Navigator views tends to be subjective. To perform measure-

25
Navigator

ments, use baseline or oblique views. Navigator views are 3D views and should NOT be
used for measurements without continuous correlation with the baseline views.

WHEN USING THE NAVIGATOR SOFTWARE, AN INCOR-

RECT THRESHOLD SETTING OR INCORRECT THRESHOLD
WARNING! MODE CAN RESULT IN ESSENTIAL ANATOMY, OR PATHOL-
OGIES, NOT BEING VISIBLE ON THE NAVIGATOR VIEWS.
ALWAYS CORRELATE THE NAVIGATOR VIEW WITH THE
ORIGINAL DATA (ACQUISITION VIEW, BASELINE VIEWS).

THE NAVIGATOR SOFTWARE USES A CONICAL PROJEC-

TION (PERSPECTIVE). FOR THIS REASON, THE ASSESS-
WARNING! MENT OF DIMENSIONS AND DISTANCES ON NAVIGATOR
VIEWS TENDS TO BE SUBJECTIVE.
ALWAYS CORRELATE APPARENT DIMENSIONS AND DIS-
TANCES ON THE NAVIGATOR VIEW WITH THE ORIGINAL
DATA (ACQUISITION VIEW, BASELINE VIEWS).

26
 Navigator

13 ERROR MESSAGES
This paragraph lists the error and warning messages that may be displayed while you are
defining or displaying a FlyThrough sequence, with suggestions for corrective action.
Create a Navigator view: You tried to create a FlyThrough sequence while no Navg.
view is displayed. Create a Navg. view.
Select one Navigator view or Select only one Navigator view: You tried to create a
FlyThrough sequence with two or more Navg. views on display. Display a single Navg.
view.
Only 256 steps can be defined: A FlyThrough sequence can contain at most 256 steps;
you tried to define a 257th step. Use (Delete Step) or (Backtrack) to remove steps.
The new point is too close to an existing point: You tried to prescribe a step less than
one voxel away from a previously defined step. Use (Delete Step) or (Backtrack) to
remove the step and try again.
Use the (Insert & Seek Step) function to prevent this problem.
Try to define points more evenly: While you prescribe steps in a FlyThrough
sequence, the system monitors the average distance between steps. It warns you when
you try to prescribe a step less than one third the average distance, or larger than three
times the average distance.
Use the (Insert & Seek Step) function to prevent this problem.
Path will hit the object: The navigator will run into the 3D surface if it follows your cur-
rent FlyThrough sequence prescription. The software repositions the cursor and prompts
you to accept the current position, or to insert a step close to the current position, to pre-
vent a collision. However, if you insist, the software will fly into the 3D surface, per your
request, and “Inside object” will be displayed in the view sequence for the section of the
path inside the 3D object.
Could not insert this point: avoid sharp turns: You asked the impossible. The navi-
gator should fly smoothly through the object. If, for example, you prescribe a new step
between two previously prescribed steps, you are asking the navigator to rapidly change
directions, and the software cannot find a satisfactory insertion point in the sequence.
Use (Delete Step) or (Backtrack) to remove the step and try again. Use the (Insert &
Seek Step) function to prevent this problem.
Cine display of movie is going to be slow (%d images)
use low-resolution mode to build the object or use larger sampling step
( )=%f mm ):
The memory requirements of the FlyThrough Sequence prescription exceeds currently
available memory space.
The software prompts you to accept the slow version of the movie, or cancel the current
movie and change one or more parameters.

27
Navigator

Accept the slow movie mode, or cancel the movie and change the 3D model build proto-
col to Navg Guide, or set the {Sampling} slider to a larger sampling distance.
Cine display of movie is going to be slow (%d images)
use larger sampling step ( ) =%f mm ):
The memory requirements of the FlyThrough Sequence prescription exceeds currently
available memory space, and you already set the 3D model build protocol to Navg
Smooth. Accept the slow movie mode, or cancel the current movie and change the
{Sampling} slider to a larger sampling distance.

28
 A BUILD MODEL menu
FlyThrough Sequence
Active Annotation FlyThrough Sequence, illustration
Adjust Threshold Initial
Location on illustration Initial, illustration
Set Aperture Navigator, illustration
Set Smoothing mode Threshold/VOI
Set Threshold Mode
Show/Hide Patient Name
Threshold and Threshold Mode Annotation
D
Adjust Threshold Delete Step

Align to Axis Display RAS coordinates

Display Speed Slider

B
Backtrack
E
Build [Navg Smooth] model Exit Navigator Package
Quick Steps

Build [Navg] model F

Quick Steps
FILE menu，[Quit]
BUILD MODEL menu
[Angio] FlyThrough Sequence
[CT Bone] Align to Axis
[CT Lungs] Backtrack
[CT Soft] Close
[Custom] Command window
[MPVR] Delete Step
[Navg Smooth] Display Speed Slider
[Navg] Image Index Slider
[Reformat] Insert & Seek Step
Build Navigator Model Insert Step
Navigator Main menu
Pause and Step
C Pause/Restart sequence
Sampling Slider
Cancel Multiview mode Save Sequence
Set Multiview mode
Close，Exit FlyThrough Sequence
Show/Clear Path
Command Window Step Slider
 View Sequence Navigator inside object

Navigator Main Command Window

I Batch button
Color buttons
Image Index Slider DISPLAY MODES menu
FILE menu
Image Measurement Accuracy GRAPHICS menu
Image selection Icon area
Illustration
Image Series Requirements Modify Model button
Movement/Rotation Increment buttons
Initial Command Window，Illustration
Oblique Mode button
Insert & Seek Step Select Object button
Tilt/Rotate Mode button
Insert Step
Tumble button
View Planes buttons
K View Type buttons
Window Parameter Preset buttons
Keyboard Shortcuts Navigator Main menu
[Cancel Multiview mode]
[Enlarge] to full screen
M [FlyThrough Sequence]
Manual Conventions [Light / Contrast]
[Lock Orientation]
Manual Overview [Reset Pointer]
[Reset] to MID
Measure Angle
[Screen Save]
Measure Distance [Set Multiview mode]
[Turnaround 180]
Measure ROI
[Unlock Orientation]
Modify ROI
Navigator outside object
Multiview mode

N O
Overview
Navigator
Manual
Description Manual Conventions
Illustration Navigator Control
Navigator Main menu Quick Steps
Realtime control Software
Start Navigator Package
 P Start Navigator Package
(Navigator) button
Pause and Step Build New Model
Quick Steps
Pause/Restart FlyThrough Sequence Reload Saved Model
Select Series/Images
Q Step Slider

Quick Steps System Requirements

Display Navigator Image
On-View keyboard shortcuts
T
Quit Navigator Package
Threshold
Adjust Threshold
R Set Threshold Mode
Suggested modes and values
Reload saved model
Threshold and Threshold Mode Annotation
Report Cursor Location

Requirements
V
Hardware and Software
Navigator Image Set View FlyThrough Sequence
ROI，Measure Region of Interest

S
Sampling Slider

Save FlyThrough Sequence

Series selection

Set Aperture

Set Multiview mode

Set Smoothing mode

Set Threshold Mode

Show/Clear Path

Show/Hide Patient Name

Software Overview
 Table of Contents Denta
Scan(Optional)

Chapter 1 - Introduction
• Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-1
• Hardware . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-1
• Conventions for This Manual . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-1
Chapter 2 - Scanning Protocols
• Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-1
Mandible . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-1
Maxilla . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-1
• Scan parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-2
• Technique . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-2
Chapter 3 - Starting DentaScan
• Exam Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-1
• DentaScan Main Window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-1
• Patient Name Annotation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-1
• DentaScan Panorex Curve Prescription Command Window . . . . 3-2
• Curve Prescription Command Menus . . . . . . . . . . . . . . . . . . . . . 3-2
FILE menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-2
EDIT menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-3
VIEW menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-3
SET menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-3
HELP menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-3
• Curve Prescription Command Buttons . . . . . . . . . . . . . . . . . . . . . 3-4
(Panorex) button . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-4
(Next) and (Prior) buttons . . . . . . . . . . . . . . . . . . . . . . . . . . 3-4
(Goto) button . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-4
Chapter 4 - Panorex and Oblique Generation
• Preparation for Drawing the Panorex Curve . . . . . . . . . . . . . . . . 4-1
Denta Scan(Optional)

• Drawing the Panorex Curve . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-2

• Recommended Panorex Curve Entry Procedure . . . . . . . . . . . . . 4-4
• Generating the Panorex View . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-5
• DentaScan Interactive Reformation Command Window . . . . . . . . 4-8
• DentaScan Interactive Reformation Command Menus . . . . . . . . . 4-8
FILE menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-8
EDIT menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-8
VIEW menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-8
SET menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-9
HELP menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-9
• DentaScan Interactive Reformation Command Buttons . . . . . . . . 4-10
(Film Sequence) button . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-10
(New Object) button . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-10
(+) (0) and (-) buttons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-10
Moving the Panorex view . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-10
• Moving the "3D Cursor" . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-11
Moving the "3D Cursor" On the Axial Slice . . . . . . . . . . . . . 4-11
Moving the "3D Cursor" On the Panorex View . . . . . . . . . . . 4-12
Moving the "3D Cursor" On the Oblique Slice . . . . . . . . . . . 4-13
• Changing Window Level and Width of Displayed Images . . . . . . . 4-14
Chapter 5 - Filming
• Setting Film Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-1
• Generating Film Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-3
Chapter 6 - Processing a New Series
CHAPTER 1 DentaScan

Introduction
Overview
DentaScan is an optional software module designed to be used with CT system.
It provides an interactive window-oriented user interface. It performs real time
image reformation specific to CT dental imaging: oblique and Panorex reforma-
tions. These reformations can be cross-referenced in real time, saved on the
image data base and printed on film.

The equipment on which this application runs includes one or

CAUTION more hard disk drives which may hold medical data related to
patients. Such equipment may in some countries be subject to
regulations concerning the processing of personal data and the
free circulation of such data.
It is strongly recommended that access to patient files be pro-
tected from all persons not in medical attendance.

Hardware
This software module runs on a standard CT system platform. It uses the plat-
form's hard copy system for film output. It accepts image data acquired on any
GE CT system.

Conventions for This Manual

Note: Throughout the text in this manual, we have used certain type styles and
symbols to differentiate between one tool or graphic and another. The
conventions we have used are as follows:
❏ Menu titles appear in bold face (e.g., Application Menu),
❏ Menu options appear in bold face, within brackets (e.g., [Exit]),
❏ On-screen tools appear in bold face, within braces (e.g., {Scroll Bar}),
❏ Graphical buttons appear in bold face, within parentheses (e.g., (View)),
❏ On-screen prompts and messages appear in italics (e.g., Login:),
❏ User typed-in responses appear in bold face italics (e.g., root), and
❏ Keyboard hardkeys and mouse buttons are underlined (e.g., Enter or left).

1-1
DentaScan

Blank page.

1-2
CHAPTER 2 DentaScan

Scanning Protocols
Overview
When scanning both a mandible and maxilla, two separate series must be used.
The gantry can be tilted at any angle. However image quality is higher if the tilt
angle is 0.
Both scanning directions are allowed (superior to inferior and inferior to supe-
rior).
Immobilization is critical to the success of this examination and it is recom-
mended that a bite stick or similar object be placed in the mouth to separate the
teeth and allow the jaw to remain stationary and relaxed. Immobilize the head
with surgical tape.

Mandible
The mandible should be scanned from the base of the mandible up through the
lower teeth.

Maxilla
The maxilla should be scanned from inferior to superior beginning at the surface
of the upper teeth to the base of the maxillary sinus. Include some sinus mucosa
to extend above and out of the hard palate.

2-1
DentaScan

Scan parameters
THICKNESS: 1.0 mm
SPACING: 1.00 mm
DISPLAY FOV: (See below)*
SCAN FOV: Head
ALGORITHM: Bone
MATRIX: 512
*DFOV (Display Field Of View) must be chosen for life-size display on a 4
on 1 format. This value is determined upon installation of DentaScan by
your GE Service Engineer according to how your hard copy output device
(typically a laser imager) is configured. Consult your application specialist
should your hard copy output device ever be reconfigured.
Best image quality is obtained if the gantry is not tilted, but the Dentascan soft-
ware corrects for the tilt automatically.
Table locations should stay in a range of 30 mm if you want the standard five
Panorex views to fit on a single view. For example, use 30 slices in the case of
a 1 mm spacing. If the number of slices is higher, the views that do not fit on the
first film will be deferred to another film.

Accuracy in the inter-axial slice directions is limited to either

CAUTION the THICKNESS or twice the SPACING value, whichever is
GREATER.

Technique
The technique depends on the kV and mA settings available on the CT system
being used. A low mA value is recommended because the region of interest con-
sists only of bone (consult your CT application specialist).

2-2
CHAPTER 3 DentaScan

Starting DentaScan
Exam Selection
Before starting DentaScan, you must first select the series or set of images (that
meets the requirements given in chapter 2 "Scanning Protocols") on the Image
Works Browser.
Note: If only one image is selected on the Browser, the entire series is used by
DentaScan.
See the Image Works 2.0 Basic Display Operator Manual for further details on
series selection and general image management.
Once you have selected the desired series or set of images, click left on the
(Denta) button in the Browser.

DentaScan Main Window

A display window pops up, displaying one of the acquisition slices. Its position in
the stack is a default parameter which can be adjusted (see chapter 5).

Patient Name Annotation

The patient name annotation is located in the upper right corner of the acquisi-
tion slice.
You can show or hide the patient name annotation as follows:
❏ Press and hold right on the annotation,
❏ Select [Show] or [Hide] from the menu that pops up,
❏ Release.

When hiding a patient name, it will not appear on film output. If

CAUTION patient name is needed on film output, don't forget to cancel
hiding of the "patient name" annotation before filming opera-
tion.

3-1
DentaScan

DentaScan Panorex Curve Prescription Command Win-

dow
A small curve prescription command window is displayed in the top right corner
of the main window (see figure below). This command window contains the nec-
essary command menus and buttons for using the DentaScan Panorex curve
editor. The top row of this window is a menu bar which contains several menu
headings: FILE, EDIT, VIEW, SET, and HELP. Below the menu bar there are
several command buttons: (Panorex) (Next) (Prior) and (Goto).These menus
and buttons are accessed by clicking left in the desired location.

Dentascan

FILE EDIT VIEW SET HELP

Panorex

Next Prior Goto

Curve Prescription

Dentascan Curve Prescription Command Window

Curve Prescription Command Menus

FILE menu
FILE When you click left on the FILE menu title, the following options are
displayed:
New Object
Quit [New Object], allows you to return to the Image Works Browser to
select a new image series.
[Quit], allows you to exit the DentaScan program.

3-2
 DentaScan

EDIT menu
EDIT When you click left on the EDIT menu title, the following options are
displayed:
Clear last point
Delete curve [Clear last point], allows you to delete the last Panorex point
entered.
[Delete curve], allows you to delete the entire Panorex curve.

VIEW menu
VIEW When you click left on the VIEW menu title, the following option is
displayed:
Panorex
[Panorex], allows you to generate the Panorex view once the Pan-
orex curve is correctly defined.

SET menu
SET When you click left on the SET menu title, the following option is dis-
played:
Film parameters
[Film Parameters], allows you to enter parameters for filming (see
chapter 5).

HELP menu
HELP When you click left on the HELP menu title, the following options are
displayed:
On Context
About Filming [On Context], allows you to pop up a window displaying the avail-
able commands. The Help window is automatically updated accord-
ing to the context. Click on the (OK) button to close this window.
[About Filming], allows you to pop up a window displaying informa-
tion about the FOV setting required to obtain a life-size image on
film.

3-3
DentaScan

Curve Prescription Command Buttons

(Panorex) button
Panorex
Panorex
When you click left on the (Panorex) button, the Panorex view and
oblique slices, based on the Panorex curve that you have defined,
are generated (see chapter 4).

(Next) and (Prior) buttons

Next When you click left on the (Next) or (Prior) button, the next or prior
Prior axial slice, respectively, in the stack of slices is displayed.

Note: The use of these buttons has no effect on the axial slice that will appear
on film output (see chapter 5).

(Goto) button
Goto When you click left on the (Goto) button, you can manually enter the
image number of the axial slice that you wish to see displayed. Move
the mouse pointer into the "Image number" field that appears, enter
the desired number (use the Back Space key to delete characters)
and then click left on (OK) to validate your choice or (Cancel) to can-
cel your choice.
Note: The use of these buttons has no effect on the axial slice that will appear
on film output (see chapter 5).

3-4
CHAPTER 4 DentaScan

Panorex and Oblique Generation

Preparation for Drawing the Panorex Curve
The Panorex curve will be drawn on the displayed axial slice. If the displayed
slice is not the desired one, use one of the following three ways to move through
the stack of slices:

Next Click left on (Next) or (Prior) buttons in the command window.

Prior

OR
Point to the slice's annotation number "Im: xx" located at its upper
left corner. Click left to move up one slice or right to move down one
slice.
OR

Point to the slice's annotation number "Im: xx" located at its upper
left corner. Hold down the middle mouse button and drag leftward to
move continuously downward in the stack or drag rightward to move
continuously upward in the stack.

4-1
DentaScan

Drawing the Panorex Curve

Once the proper axial slice is displayed, draw the Panorex curve as follows:
Holding the Shift key down, press left, drag and release mouse but-
ton at the first desired point.

Shift

Still holding the Shift key down, press left, drag and release mouse
button at the next desired point. During dragging, a rubber band
appears between the last point entered and the current mouse posi-
tion.

Shift

Still holding the Shift key down, continue pressing left, dragging and
releasing the mouse button at all the desired points. Release the
Shift key after entry of the last point.

Shift

The actual drawn curve will be a smooth curve fitted on the points that you
defined. You can use as many points as you want, but it is recommended that
you use a small number of evenly spaced points (typically 5 to 7). See "Recom-
mended Panorex Curve Entry Procedure" below.
If during the process of defining the curve points you want to change the dis-
played axial, you can still use the (Prior) and (Next) buttons in the command
window, or page through the axial slices by clicking on slice annotations.

EDIT If you want to delete the last defined point, click left on [Clear last
point] in the EDIT menu.
Clear last point
Delete curve It is also possible to delete the whole curve by choosing [Delete
curve] in the EDIT menu. The Panorex curve can be deleted at any
stage of the session, even if you have already generated Panorex
views and oblique slices as described in the following sections.

4-2
 DentaScan

To change the position of the defined points directly on the axial view, proceed
as follows:.
Move the mouse pointer to the axial view and press the Meta ( )
key (next to the Space bar). The original curve points are now dis-
played as red squares and new points are displayed as green dia-
monds.

Still holding the Meta key down, press left on the defined point
(marked as a red square) to be moved and drag to the new location.

Still holding the Meta key down, release left mouse button to enter
the new point location.

Release the Meta ( ) key to quit the point editing mode.

4-3
DentaScan

Recommended Panorex Curve Entry Procedure

Panorex curves with sharp bends can result in loops on lingual
CAUTION or bucal Panorex images. The following procedure is therefore
highly recommended for avoiding this problem when entering
Panorex curve points:

❏ Begin by placing three points as shown below:

❏ If the curve is not close enough to the desired shape, add new points
using the Meta key and moving the new points (green diamonds) to the
desired locations.
❏ Release the Meta key after point editing. Repeat the above step if neces-
sary.
Note: A satisfactory result should be obtained with a maximum of 5 or 7 points.
Note: When the beginning or end point has to be slightly changed, it is prefera-
ble to edit it with the Meta key, as opposed to adding a new point close to
it.

4-4
 DentaScan

Generating the Panorex View

Before generating the Panorex view, be sure to have the desired axial slice FOR
FILM OUTPUT on the display. (This has no effect on the curve that you have just
drawn).
This is done by either using the (Prior) and (Next) buttons or clicking on slice
annotations.
Note: Once the Panorex view is generated, the axial slice for film output
remains fixed, even if you change the viewed axial slice afterwards. The
only way to change the axial slice for film output once the Panorex view
has been generated is to modify the "Reference axial" value as described
in chapter 5 below.

Panorex Once the proper axial slice is displayed for film output, click left on
the (Panorex) button in the command window, or select [Panorex]
from the VIEW menu.
Two new views are generated:
❏ The Panorex view in the lower right corner of the display window.
❏ An oblique slice is reformatted perpendicularly to the Panorex view and
displayed in the upper left corner of the display window.
The trace of the reformatted oblique plane is visible on the axial slice.

4-5
DentaScan

After generating the Panorex curve, by selecting the (Panorex)

CAUTION button, check the oblique traces for possible intersection. If
ANY of the oblique traces on the axial slice intersect each other,
use the (+), (0) and (-) buttons (see "DentaScan Interactive Ref-
ormation Command Buttons" below) to determine if any of the
defined Panorex curves will lie outside the intersections. All
Panorex curves occurring outside the oblique intersections are
NOT USABLE. If this happens, try editing the original curve by
selecting [Edit curve] in the EDIT menu: move or remove points
to make the sharp bends smoother.

Oblique

Innermost and outermost

Panorex lines are not
usable since they occur
outside a point of
intersection.

The figure on the next page shows what the DentaScan display looks like after
Panorex generation.

4-6
 DentaScan

Dentascan

FILE EDIT VIEW SET HELP

Film Sequence Number of Copies : 1

New Object

Shift curve ( mm ) : 0.0

+ 0 -

Interractive Reformation

The DentaScan display following Panorex

generation

4-7
DentaScan

DentaScan Interactive Reformation Command Window

Once the Panorex view is generated, the command window changes (see insert
on the previous page). The new menus and buttons are described below.

DentaScan Interactive Reformation Command Menus

FILE menu
FILE When you click left on the FILE menu title, the following option is dis-
played:
Quit
[Quit], allows you to exit the DentaScan program.

EDIT menu
EDIT When you click left on the EDIT menu title, the following options are
displayed:
Edit curve
Delete curve [Edit curve], allows you to return to the Curve Prescription function
for curve editing.
[Delete curve], allows you to return to the Curve Prescription func-
tion ready for entry of a new Panorex curve (the existing Panorex
curve and views are deleted).

VIEW menu
VIEW When you click left on the VIEW menu title, the following option is
displayed:
Cancel preview
[Cancel preview], allows you to delete, from the screen, markings
that will appear on the film output. When the VIEW menu is again
selected, the option [Preview film] appears, which allows you to
restore markings that will appear on the film output to the screen.

4-8
 DentaScan

SET menu
SET When you click left on the SET menu title, the following options are
displayed:
Film Parameters
First Oblique [Film Parameters], allows you to enter parameters for filming (see
Last Oblique chapter 5).
Whole Curve [First Oblique], allows you to reduce the range of obliques to the
area of interest. The currently selected oblique becomes the first one
in the reduced range when this option is selected.
[Last Oblique], allows you to reduce the range of obliques to the area of inter-
est. The currently selected oblique becomes the last one in the reduced range
when this option is selected.
[Whole Curve], allows you to restore the full range of obliques following reduc-
tion of the range via the [First Oblique] and/or [Last Oblique] options.

HELP menu
HELP When you click left on the HELP menu title, the following options are
displayed:
On Context
About Filming [On Context], allows you to pop up a window displaying the avail-
able commands. The Help window is automatically updated accord-
ing to the context. Click on the (OK) button to close this window.
[About Filming], allows you to pop up a window displaying informa-
tion about the FOV setting required to obtain a life-size image on
film.

4-9
DentaScan

DentaScan Interactive Reformation Command Buttons

(Film Sequence) button
Film Sequence When you click left on the (Film Sequence) button, a film hard copy
is generated (see chapter 5).

(New Object) button

New Object When you click left on the (New Object) button, you can return to
the Advantage Windows browser and select a new image series
(see chapter 6).

(+) (0) and (-) buttons

+ When you click left on the (+) or (-) button, you translate the Panorex
0 curve towards the bucal or lingual side of the original Panorex curve,
respectively (see below). When you click left on the (0) button, you
- return the Panorex curve to its original position.

Moving the Panorex view

The Panorex curve can be translated towards the bucal or lingual side. You do
this in one of the following two ways:
❏ Click left on the (+) or (-) button in the command window, which performs
a translation towards the bucal or lingual side of the original Panorex curve,
respectively. The amount of the translation is equal to the value in the "Dis-
tance between Panorex views" field in the "Defaults" command window (see
chapter 5 "Filming").
OR
❏ Move the mouse pointer into the "Shift curve" numerical field and enter a
numerical value in mm (use the Back Space key to erase characters) and
press the Enter key to make the new value effective.

Shift curve ( mm ) : 0.0

+ 0 -

Note: To revert to the original curve, click left on the (0) button in the command
window.

4-10
 DentaScan

Moving the "3D Cursor"

A small red dot is displayed simultaneously on the Panorex view, the oblique
slice and the axial slice at the intersection of the Panorex curve with the oblique
plane trace.
These three red dots define a unique 3D point cross-referenced between the
three views.
You can move this "3D cursor" directly on any one of the three views as
described below.
By combining these three movements, you can precisely investigate the anat-
omy of the patient at very specific locations.

Moving the "3D Cursor" On the Axial Slice

Move the mouse to the axial slice.
Hold down the Shift key.
Move the mouse to the desired position.

Shift

Note: The displayed oblique is updated and the red dot on the Panorex view will
move to reflect the new 3D cursor position.
Release the Shift key to freeze the state of the displayed images.
Note: An alternate method is to move the mouse arrow to the 3D cursor, press
left mouse button, drag to the desired 3D cursor location and release.

4-11
DentaScan

Moving the "3D Cursor" On the Panorex View

Move the mouse to the Panorex view.
Hold down the Shift key.
Move the mouse to the desired position.

Shift

Note: The oblique slice is re-computed and its trace is updated on the axial
view.The displayed oblique will be updated and the red dot on the Pan-
orex view will move to reflect the new 3D cursor position.
Note: Horizontal mouse movements change the oblique slice location, but verti-
cal mouse movements will change the displayed axial slice, since such
movements correspond to moving the 3D cursor up and down in the
stack of axial slices.
Release the Shift key to freeze the state of the displayed images.
Note: An alternate method is to move the mouse arrow to the 3D cursor, press
left mouse button, drag to the desired 3D cursor location and release.

4-12
 DentaScan

Moving the "3D Cursor" On the Oblique Slice

Move the mouse to the oblique slice view.
Hold down the Shift key.
Move the mouse to the desired position.

Shift

Note: The small red dot corresponding to the 3D cursor position is updated on
the axial view and the small red dot on the oblique slice moves to reflect
the new 3D cursor position. Also, if the new 3D cursor position falls in the
Panorex view, the corresponding red dot will appear there.
Note: As in the case of the Panorex view, vertical mouse movements will
change the displayed axial slice, since such movements correspond to
moving the 3D cursor up and down in the stack of axial slices.
Release the Shift key to freeze the state of the displayed images.
Note: An alternate method is to move the mouse arrow to the 3D cursor, press
left mouse button, drag to the desired 3D cursor location and release.

4-13
DentaScan

Changing Window Level and Width of Displayed Images

Window level and width parameters are also used for film out-
CAUTION put (see chapter 5).

The window level and width settings are visible in the bottom left corner of the
Axial view (L stands for Level and W stands for Width). You can alter these set-
tings as follows:
Move the mouse to the Panorex or axial view and press the middle
mouse button.

Drag the mouse up or down to make the image brighter (window

level reduced) or darker (window level increased), respectively.

Drag the mouse left or right to make the image sharper (window
width reduced) or smoother (window width increased), respectively.

Note: A small red square appears at the location where you pressed the mouse
button. The further you move from this square, the more sensitive mouse
movements are.
Note: The window level and width parameters can be preset to default values
via the "Defaults" command window (see chapter 5).

4-14
CHAPTER 5 DentaScan

Filming
Setting Film Parameters
DentaScan allows you to hard copy a set of images on film according to a pre-
defined format. This format defines:
❏ The axial slice to be printed on the film,
❏ The spacing between oblique slices (must be between 1 and 5 mm),
❏ The width of the oblique slices (must be less than DFOV/2),
❏ The width of the oblique markers on the film,
❏ The number of Panorex views (maximum 9) and the spacing between
them displayed on the film (must be between 1 and 5 mm).
❏ It also defines the window level parameters used for image generation. To
view the state of these parameters, proceed as follows:

SET Click left on the SET menu in the command window, then click left on
[Film Parameters].
Film Parameters
First Oblique A window entitled "Defaults" pops up. (See figure below).
Last Oblique Move the mouse pointer into the numerical field corresponding to the
Whole Curve parameter to be modified and enter a numerical value (use the Back
Space key to erase characters) and press the Enter key to make the
new value effective.

Reset Click left on (Reset) to revert to original values.

Save Click left on (Save) to store the current values as default values for
future DentaScan sessions.

Update Click left on (Update) to use the current settings from the reference
axial as the default settings.

Done Click left on (Done) to close the Defaults window.

Click left on (Yes) for Auto Filming if you wish to obtain a hard copy. Click left on
(No) if you do not wish to obtain a hard copy.
Click left on (Yes) for Archive Images if you wish to save the images generated
by DentaScan. If you do so, a unique series is created in the exam from which
the images were generated. Click left on (No) if you do not wish to save images.

5-1
DentaScan

Defaults
Done Reset Save Update

Auto Filming Yes No

Archive images Yes No

Window Level : 600

Window Width : 3200

Reference axial : 5

Spaching between obliques (mm.) : 3.0

Width of obliques (mm.) : 30.0

Width of oblique markers (mm.) : 5.0

Number of Panorex views : 5

Distance between Panorex views (mm.) : 2.0

5-2
 DentaScan

Generating Film Output

Film Sequence To generate film output, click left on (Film Sequence) in the com-
mand window.
The following pages show a typical film output with cross-reference marks.

Before filming, if the patient name is required, check that it has

CAUTION not been hidden on the axial slice on the screen.

Check that a cross-reference mark exists for each axial slice

CAUTION used for Panorex and Oblique construction.

5-3
DentaScan

Slice
cross-reference marks

5-4
 DentaScan

Oblique
cross-reference marks

5-5
DentaScan

Panorex
cross-reference marks

Millimeter rulers

5-6
CHAPTER 6 DentaScan

Processing a New Series

New Object
To process a new series, click left on (New Object) in the command
window once you have defined the Panorex views.

Yes Click left on the (Yes) button in the Question window that appears.
Doing this will close the DentaScan display window and let you
select another series on the Advantage Windows Browser.

Load
Once you have selected your new series on the Browser, click left on
(Load) in the DentaScan *command window .

6-1
DentaScan

Blank page.

6-2
 Table of Contents Volume
Rendering (Optional)

CHAPTER 1 - GENERAL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1 OVERVIEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
2 HOW TO USE THIS DOCUMENT. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
3 CONVENTIONS FOR THIS MANUAL . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
CHAPTER 2 - INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1 OVERVIEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1-1 Rendering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1-2 Rendering Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1-3 Volume Rendering. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1-4 Illustration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2 VOLUME RENDERING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
3 OTHER RENDERING TECHNIQUES . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
3-1 Surface Rendering. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
3-2 MIP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
4 PARTIAL VOLUME FILTER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
CHAPTER 3 - STARTING VOLUME RENDERING . . . . . . . . . . . . . . . . . . . 11
1 OVERVIEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2 VOLUME RENDERING ON 3D VIEWS . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2-1 At Startup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2-2 Main screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
3 VOLUME RENDERING ON MPVR OBLIQUE VIEWS . . . . . . . . . . . . . . . 12
CHAPTER 4 - CONTROLS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
1 GENERAL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
1-1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
1-2 VR Presets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
1-2-1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
1-2-2 Factory-Defined presets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
1-2-3 VR presets controls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
Volume Rendering (Optional)

1-3 VR Opacity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
1-3-1 VR opacity control panel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
1-3-2 Low and high control points . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
1-3-3 Max. opacity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
2 OPTIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2-1 Color or Black-and white shading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2-1-1 Enhanced Resolution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2-1-2 Enhance Contrast . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2-1-3 Partial Volume Filter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2-2 Save Preset . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
3 VR COLORS, COLORS SHADING MODE. . . . . . . . . . . . . . . . . . . . . . . . . 22
3-1 Shading Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
3-2 Color Slider . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
3-3 Brightness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
3-4 Color Style . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
3-5 Creating Color Sliders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
3-6 Deleting Color Sliders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
4 VR COLOR, DEFINE OBJECT MODE . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
4-1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
4-2 Defining an object . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
4-3 Adjusting the Range of visible Voxels . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
4-4 Adjusting the opacity of an object . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
4-5 Detaching an object . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
GLOSSARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
 Volume Rendering

CHAPTER 1 - GENERAL

1 OVERVIEW
Volume Rendering is an optional software package to be used with Volume Analysis 2 .
In general terms, “rendering” refers to the techniques used to represent a 3D (three-
dimensional) object on a two-dimensional surface.
The basic Volume Analysis 2 application uses two main rendering techniques: surface
shading, which treats the 3D object as a solid and fully opaque object, and projection
shading, which treats the 3D object as if it were translucent.
With the Volume Rendering option, volume rendering is added to these rendering tech-
niques.
Volume rendering uses the notion of opacity, which can be defined as the degree to
which light cannot penetrate an object. Different “opacity” values are assigned to differ-
ent voxel values which can represent different tissue properties such as density.
The effect is to make highly opaque objects more clearly visible among less opaque
objects which appear transparent to a greater or lesser degree. The result is the ability to
see many layers of tissue, instead of only the first layer as with surface shading.
Volume rendering can be used in three different modes: black-and-white shading, color
shading, and multiple object color coding.

WHEN USING THE VOLUME RENDERING SOFTWARE

OPTION, AN INCORRECT THRESHOLD SETTING, INCOR-
WARNING! RECT CURVE TYPE (THRESHOLD MODE), OR INCORRECT
OPACITY SETTING, CAN RESULT IN ESSENTIAL ANAT-
OMY, OR PATHOLOGIES, NOT BEING VISIBLE ON THE
VIEWS DISPLAYED IN VOLUME RENDERING MODE.
ALWAYS CORRELATE VIEWS THAT USE VOLUME REN-
DERING WITH THE ORIGINAL DATA (ACQUISITION VIEW,
BASELINE VIEWS).

1
Volume Rendering

2 HOW TO USE THIS DOCUMENT

For an introduction to the underlying principles of volume rendering, read paragraph 2
“Introduction”.
To start using volume rendering on 3D or MPVR oblique views, refer to paragraph 3
“Using Volume Rendering”.
For the controls and functions used by the Volume Rendering software, refer to para-
graph 4 “Controls”.
For a list of the terms that are specific to the application and to volume rendering tech-
niques, refer to the Glossary at the back of the manual.
Note: In this chapter, “Volume Rendering” (with capitals) refers to the software option,
while “volume rendering” refers to the volume rendering techniques as used on
the 3D and MPVR oblique views.

3 CONVENTIONS FOR THIS MANUAL

Throughout the text in this manual, we have used certain type styles and symbols to dif-
ferentiate between one tool or graphic and another. The conventions we have used are
as follows:
• Menu titles appear in bold face: Application menu.
• Menu options appear in bold face, within brackets: [Exit].
• Graphical buttons appear in bold face, within parentheses: (View).
• On-screen tools appear in bold face, within braces:{Scroll Bar}.
• On-screen prompts and messages appear in italics: Login:.
• User typed-in responses appear in bold face italics: sdc.
• Keyboard hardkeys and mouse buttons are underlined: Enter or left)

Whenever this manual refers to “clicking”, “selecting”, “pressing a button”, etc. with the
mouse, this always means using the left mouse button, unless the middle or right mouse
button is specifically mentioned.

2
 Volume Rendering

CHAPTER 2 - INTRODUCTION

1 OVERVIEW

1-1 Rendering
In general terms, “rendering” refer to the techniques used to represent a 3D (three-
dimensional) object on a two-dimensional surface.

As a example, if a sphere is projected on a flat surface,

the result is a circle. Any indication of it being a 3D
object is lost.
The perception of depth can be restored to the projec-
tion with rendering techniques.
In the simple illustration at the left, only a few levels of
shading are required to restore the impression of a
sphere. Here, the shading provides the depth cues cor-
responding to the amount of light that would be
reflected from a source above and to the left of the
sphere.

1-2 Rendering Techniques

In medical imaging, it will be obvious that the complexities of shape, spatial orientation
and tissue properties require more sophisticated display and rendering techniques than
in the simple illustration above.
The aim is to extract as much information as possible from a three-dimensional CT data
set displayed on a two-dimensional workstation screen.
To this purpose, Volume Analysis 2 combines various types of rendering with other tech-
niques such as volume segmentation (thresholding, scalpel) to define a 3D object, rota-
tion of the 3D object on the screen to view it from different angles, or perspective with the
Navigator option.
The basic Volume Analysis 2 application uses two main rendering techniques:
• Surface shading, which treats the 3D object as a solid, fully opaque object.
• Projection shading, which treats the 3D object as if it were translucent, using different
projection algorithms, such as Maximum Intensity Projection (MIP), Minimum Intensity
Projection (MinIP) and RaySum.
With the Volume Rendering option, volume rendering is added to these rendering tech-
niques.

3
Volume Rendering

Volume rendering uses the notion of opacity, which can be defined as the degree to
which light cannot penetrate an object. In the context of medical imaging, different opac-
ity values are assigned to different voxel values which can represent different tissue prop-
erties such as density.
The effect is to make highly opaque objects more clearly visible among less opaque
objects which appear transparent to a greater or lesser degree. The result is the ability to
see many layers of tissue, instead of only the first layer as with surface shading.
Volume rendering can be used in three different modes:
• Black-and-white shading Color shading
• Color shading
• Multiple object color coding

1-3 Volume Rendering

Volume rendering uses the notion of opacity for 3D visualization.
Opacity can be defined as the degree to which light cannot penetrate an object. In the
context of medical imaging, different opacity values are assigned to different tissue prop-
erties such as density (voxel value). The effect is to make highly opaque objects more
clearly visible among less opaque objects which appear transparent to a greater or lesser
degree.
The result is the ability to see many layers of tissue, instead of only the first layer as in
surface rendering. Because the transition between what can and cannot be seen is now a
gradual one, and not abrupt as in surface rendering, the user should find it faster to
define suitable views.
A brief overview of the rendering process may help in understanding this technique.
Digital medical image data are manipulated in a matrix of volume elements called “vox-
els”. An image is constructed by analyzing each voxel and projecting the result on a two-
dimensional surface subdivided in picture elements called “pixels”. Given the tissue prop-
erties and information about the perspective of the viewer, the resulting image shows the
result of how light behaves as it passes through the volume of voxels.
With a model defined in terms of opacity, rays of light will penetrate the tissue in varying
degrees. At each point, a certain amount of light is reflected and the rest is transmitted to
the next layer where the same process occurs. In the end, the viewer sees an image
which is the sum of reflections from each layer. This differs from a surface rendered
model where the light rays are entirely reflected by the first surface they encounter. The
simplified example below will help illustrate this difference.

4
 Volume Rendering

1-4 Illustration
Volume rendering allows the simultaneous display of object with different properties. As
an example, consider a CT contrast study where the user seeks to display a contrast-
enhanced blood vessel among the surrounding soft tissue.

A Illustration (A) depicts a vessel embedded in soft tissue of a lower

density.

Using thresholding techniques, the vessel and surrounding soft tis-

sue cannot be displayed clearly at the same time.

If the threshold is set low enough to show the tissue, then the vessel
B
is hidden except at the cut plan (B).

If the threshold is set high enough to clearly show the vessel, then
C
the soft tissue is no longer visible (C).

Rendering modes such as Maximum Intensity Projection (MIP) can

D
be used to display structures of different density. However, the
structures are displayed as projections on the surface of the object
and all depth cues are lost (D). In addition, since MIP displays the
brightest, or most dense, object in the ray path, “bright” structures
such as bone tend to obscure any other structure present.

Volume rendering can be used to show both objects at the same

E
time. By assigning a low opacity to the soft tissue and a high opacity
to the dense vessel, the vessel becomes clearly visible among its
semi-transparent surroundings (E).

The example is simplified because it treats the two objects as though they have uniform
and distinctly different densities. In reality, any single object is likely to have a non-uni-
form density, and the difference in density between multiple objects is often not enough to
make them clearly distinguishable.

5
Volume Rendering

Therefore, the process of segmenting tissue based on voxel value threshold can be diffi-
cult. If the threshold is set too low, some of the unwanted surrounding tissue will appear
on the image and mask the vessel behind it. If the threshold is set too high, part of the
vessel will not be displayed.
The volume rendering technique is more tolerant to variation in tissue properties. Where
thresholding either entirely shows or hides tissue of a given density, volume rendering
allows the tissue to “fade” from visible to invisible.
The following sections provide a more detailed explanation of how this works.

6
 Volume Rendering

2 VOLUME RENDERING
Volume rendering is a 3D visualization technique which uses the data frim the entire vol-
ume to creat an image. Information from every point along a ray path is used to deter-
mine the resulting reflection along that path. As light travels through the volume, the
opacity at each point determines the amount of light which is reflected from that point.
The remaining light travels to the next layer where the process is repeated.
In medical imaging, each element of volume is represented by a voxel. The reflection
which creates the final image is the result of the light reaching a voxel times the opacity of
the voxel, or, expressed as an equation:
reflection=(amount of light) X (opacity).
The figure below illustrates this.
The row of shaded squares represents voxels within tissue of increasing density.
The graphs show the opacity value, the light that reaches each voxel, and the resulting
reflection.

The opacity of the voxel defines how

visible it is.
Light
− An opacity of 0 (zero) is assigned to
light reaching each voxel completely transparent voxels, that
will not be seen on the image.
− An opacity of 1 (one) is assigned to
opacity completely opaque voxels that do
not transmit light but completely
reflect it, and hence appear solid.
− Voxels with intermediate opacity
values will appear semi-transpar-
reflection
ent.
1 2 3 4 5 6 7 8 9 The amount of light reaching each
depth voxel decreases as it passes deeper
into the tissue. Each voxel reflects
some of the light, and only the remain-
ing light reaches the next layer.
The reflected light is used to generate
the final image which the user sees.

7
Volume Rendering

The final result now depends on the shading technique:

• Black-and-white shading(voxel
value averaging)
The shading value for a voxel is sim-
ply defined by its value. The final
result will be the weighted average of
voxel value along each ray.
In practice, the most significant con-
tribution will come from voxels
located close to a surface where the
tissue property is the same. Edges
will be visible as dark lines because
a ray will strike a larger number of
partial volume voxels.

This type of shading is useful for vascular or bone display from CT acquisitions.

• Color shading
The shading value for a voxel is
defined by the local orientation of the
surface at the location of the voxel.
The color is based on the voxel value
in the case of a single object, or on the
color assigned to the different object
in the case where multiple object are
defined.

• 3D object color coding

As with color shading (see above), the
shading value for a voxel is defined by
its value, i.e., its opacity, and the local
orientation of the surface at the loca-
tion of the voxel.
The color is based on the color
assigned to the attached object, or
objects in the case where multiple
objects are defined.

8
 Volume Rendering

3 OTHER RENDERING TECHNIQUES

3-1 Surface Rendering

Surface rendering is a 3D visualization technique where all objects are fully opaque. As a
consequence, surface rendering is more sensitive to threshold or classification choices.

The choice of a low threshold may hide

object by including more tissue than
desired.
If the threshold is too high (under-esti-
mation of object), part of the object of
interest may be lost.

Volume rendering will only show faint traces of tissue if there is a distinguishable differ-
ence in tissue property.

3-2 MIP
The MIP (Maximum Intensity Projection) rendering mode displays only the maximum
pixel intensity along a ray path.

This mode may be used to render vas-

cular structures. No depth cues are
provided, making it difficult to display
complex strcuctures.
In CT Angiography (CTA), bright values
from bone will hide vessels in front or
behind them.

Volume rendering is well suited to the display of vascular anatomy, because it is less sen-
sitive to threshold choice than surface rendering. For Angiography, volume rendering can
display thinner vessels in front of larger ones and will provide the correct depth cues with-
out surface shading.

9
Volume Rendering

4 PARTIAL VOLUME FILTER

The “partial volume effect” refers to

the appearance of voxels with inter-
Fat/bone interface
mediate values at the interface of
Voxel size two dissimilar tissue types.
A common example occurs during
Bone CT Angiography exams, where the
Fat (0 HU)
Fat Fat typical value for vasculature (in the
illustration 300 HU is assumed) is
Vessel or fat/bone intermediate between fat (0 HU
interface (300 HU)
assumed) and bone (600 HU
assumed). As a result, artifical
Bone (600 HU) “vessels” may appear at the inter-
face between fat and bone, where
"Vessel" artifact
the average of these two tissues is
in the range of vasculature.
In many cases this makes it difficult
to select a single class of tissue,
because partial volume artifacts
exist in the desired range of values.
Note: The HU values assumed for
the different tissue types in
the illustration are to be
regarded as examples only.
Real values may differ sig-
nificantly.

The Partial Volume Filter attenuates the partial volume by associateing partial volume
voxels with adjacent voxels which represent the more opaque object. In the illustration
above, the value of the nearby bone voxels will be used for the fat/bone interface.
The filter has two efffects:
• It improves the visibility of structures that would otherwise be hidden by partial volume
voxels.
• It improves the color coding of voxels in color mode.
See Chapter 4, paragraph 2-3-3 on how to activate the Partial Volume Filter.

10
 Volume Rendering

CHAPTER 3 - STARTING VOLUME RENDERING

1 Overview
Within the Volume Analysis 2 application you can use volume rendering on 3D views
and on MPVR oblique views.
• To use volume rendering on 3D views, you can select the “VR” (Volume Rendering)
protocol in the “General” category after startup of Volume Analysis 2.
• To use volume rendering on an MPVR oblique view (“thick” slice), you must first set up
the view, then switch the rendering mode to “volume rendering”.
Note: Any build protocol can be used with volume rendering, although the use of the
“VR”, “Reformat”, “Angio” or the optional “Navigator” and “Navigator Smooth” pro-
tocols is recommended.
Setting the model mode in a "Custom" protocol to “Volume”, or using protocols
such as “CT Bone” that use the “Volume” model mode , may at times result in very
large volumes of data and poor system performance. Memory requirements may
be above system limits.
Note, that the “Volume” model mode adds data in the 3D model that are used for
surface shading, and that Volume Rendering does not require these data.
See Chapter 5 “Building the 3D Model” for more details.

11
Volume Rendering

2 Volume Rendering on 3D Views

2-1 At startup
To obtain a 3D view in volume rendering mode at startup,
• Start Volume Analysis 2 from the Image Works Browser as usual,
• Select (General) as the anatomical region category,
• Select and apply the [VR] protocol.

2-2 Main screen

On the main Volume Analysis 2 screen (control panel and views), to switch a 3D view to
volume rendering mode:

• On the view, click and hold left or right on the rendering mode active
annotation (near top left) and select [Volume Rendering] in the drop-down
menu,
OR
• On the view, click and hold left or right on the view type active annotation
(top left) and select [VR] in the drop-down menu. The view type will still
indicate “3D” but the rendering mode switches to “Volume Rendering”,
OR
• In the control panel, select (Display Tools) > [Viewport Layout]) and set
1 2
3 4
the view type to [VR] for the desired view.

3 Volume Rendering on MPVR Oblique Views

You can use volume rendering on MPVR oblique views. To do so you must first have set
up the MPVR oblique view as detailed in Chapter 7 “Reformatting”.
To switch an MPVR oblique view to volume rendering mode:
• On the view, click and hold left or right on the MPVR rendering mode active annota-
tion (top annotation at lower left, next to the slice thickness annotation) and select [VR]
in the drop-down menu.

12
 Volume Rendering

CHAPTER 4 - CONTROLS

1 GENERAL

1-1 Overview
When you activate the volume rendering option on 3D or MPVR oblique views (see previ-
ous paragraph), an additional (VR Tools) tool menu button is displayed on the main Vol-
ume Analysis 2 control panel.
The VR Tools menu gives you access to the three VR (Volume Rendering) control pan-
els:
- VR Presets,
- VR Opacity,
- VR Colors.
You can switch between the three control panels either by means of the (Back) and
(Next) buttons (at bottom right on the panels) or by means of the VR Tools menu.
You can remove a Volume Rendering control panel temporarily from view by means of its
(Close) button, without affecting settings of the volume rendering mode on the current 3D
or MPVR oblique view. To bring the panel back into view, reselect the corresponding
menu item in the VR Tools menu.
To stop using volume rendering on a view and remove the (VR Tools) tool menu button,
simply select a different rendering mode or view type.

13
Volume Rendering

1-2 VR Presets
Overview
When using volume rendering on a view, you must set up several parameters: opacity
curve shape and values, type of shading, window width and level, etc.
You can do this manually, using the controls on the view and on the VR Opacity and VR
Colors panels (see further), or you can use the pre-defined settings available on the VR
Presets panel.
Volume Rendering is supplied with a set of factory-defined presets.
You can also store settings you have defined yourself as new presets, and retrieve them
at a later time. You can add your own comments to each new preset. See paragraph
“Save Preset”.
You can delete any presets that are either not relevant to your work, or that are unsatis-
factory for any other reason. See paragraph “VR Presets Controls”.

Factory-Defined Presets
The factory-defined presets are intended as examples, and contain detailed comments
that explain the rationale for each setting.
• Use them as an aid to explore various settings for visualizing different anatomical
structures, when you are familiarizing yourself with the software.
• Use them as as a starting point for the display of a particular structure, then use the
manual controls to optimize the display.

The factory-defined presets are provided as examples. You should

CAUTION
be aware that actual results will vary for individual patients and
acquisition techniques.

14
 Volume Rendering

VR Presets controls
To apply a preset to the current image:
• Select the anatomical category from the menu associated with the (Anatomy) button,
to display the corresponding list or presets (panel of icons).
To display the comments associated with a preset icon, leave the mouse pointer
steady on the icon for a few moments.
If more presets are defined than can be shown in the window, use the small {up} and
{down} buttons below the window to move through the available presets.
• Click on the preset icon (a border round the icon shows that it is selected). The preset
is applied automatically.
To delete a preset:
• Click on the preset icon to select it, then click on (Delete Preset).
Refer to paragraph 5 below for the functions of the (Color) and (Enhanced Resolution)
buttons.

15
Volume Rendering

1-3 VR OPACITY
VR Opacity control panel
You use the controls on this panel to define and adjust the opacity curve, to set certain
options, and to save presets.
Define opacity curve
Three parameters define the opacity curve:
- Curve type.
- The low and high control points on the voxel value scale.
- Maximum opacity value.
Curve Type
To select the opacity curve shape, click on the corresponding (Curve Type) button. See
paragraph above for the definition of “Max. Opacity”.

This is the most usual choice. It may be used to diaplay “bright” structures,
such as bone or vessels in CT data sets.
− Voxels with values below the setting of the left control point slider will be
transparent.
− Voxels with values above the setting of the right control point slider will
be at “Max. Opacity”.
− The opacity of voxels with values between the two sliders will increase
linearly from transparent at the value of the left slider to “Max. Opacity”
at the value of the right slider.
This is a step curve with only one control point. It allows you to obtain a
surface shading rendering on the VR views.
− Voxels with values below the setting of the control point will be transpar-
ent.
− Voxels with values above the setting of the control point will be at “Max.
Opacity”.

16
 Volume Rendering

This curve type is used to display structures with voxel values within a
defined range. Liver lesions in CT or CTA bolus data sets are examples of
such structures.
Partial volume effects (see Chapter 2) can also generate voxel value
within the range you have selected at the interface between dark and
bright regions. The Partial Volume Filter (see Chapter 2 and paragraph 2-
3-3 in this chapter) can be used to attenuate this effect.
− Voxels with values both below the setting of the left control point slider
and above the setting of the right control point slider will be transparent.
− The opacity of voxels with values between the two sliders will increase
linearly from transparent at the value of the left slider to “Max. Opacity”
at the midpoint between the two sliders, then decrease to transparent at
the value of the right slider.
This curve type can be used to display dark structures, such as airways, or
vessels in “black blood” MRA techniques. It can also used with a cut plane
to display the lumen of a vessel as “etched” in the plane.
− Voxels with values below the setting of the left control point slider will be
at “Max. Opacity”.
− Voxels with values above the setting of the right control point slider will
be transparent.
− The opacity of voxels with values between the two sliders will decrease
linearly from “Max. Opacity” at the value of the left slider to transparent
at the value of the right slider.
This curve type can be used with a cut plane to display the lumen of a ves-
sel as “etched” in the plane, similar to the curve type above.
− Voxels with values both below the setting of the left control point slider
and above the setting of the right control point slider will be at “Max.
Opacity”.
− The opacity of voxels with values between the two sliders will decrease
linearly from “Max. Opacity” at the value of the left slider to transparent
at the midpoint between the two sliders, then increase to “Max. Opacity”
at the value of the right slider.

Note: A range of pixel values may not exclusively define a single tissue or organ. As a
consequence, it may be difficult to visualize particular structures of interest since
they may be hidden by other objects with the same values.

17
Volume Rendering

Low and High Control Points

• To adjust the low or high control

20 240 point, press and drag the left or
right diamond-shaped slider to the
desired value.
HU HU
0 100 200 300 When you move a control point
slider, the scale below the sliders
is resized automatically. You can
also resize the scale manually by
pressing and dragging on the
scale markings.

The triangular mark on the scale continuously shows the voxel value at the current posi-
tion of the 3D cursor on the image window. This can be used as an aid to position the
control point sliders.
The selected curve type is displayed between the sliders (the illustration shows the “Low-
to-high” curve type).
Moving one of the control point sliders also highlights the voxels on the reformatted and
MIP views that have an opacity value greater than 10% of the maximum opacity value.
This will help you to select the range of voxel values corresponding to the anatomy that
you want to display.

18
 Volume Rendering

Max. Opacity
The maximum opacity value (“Max. Opacity”) is set by means of the {Max. Opacity}
thumbwheel.

• Press on the {Max. Opacity} thumbwheel and drag it

Max. Opacity : 80% up and down to the desired value, or click on the tri-
angles above and below the thumbwheel to increase
and decrease the value by steps.
This defines the opacity (in percent) of the point of
maximum opacity of the selected curve. A value of
100% corresponds to fully opaque, where a voxel
reflects 100% of all incident light. A value of 1%
means that a voxel is nearly transparent, and will
reflect only 1% of the incident light.

Setting the maximum opacity value to a low value may result in very dark images,
because only a fraction of the incident light is used to create the final image.
On black-and-white images, use the regular window width/level controls to achieve a cor-
rect shading level.
On a color image, the brightness is adjusted automatically whenever the maximum opac-
ity value is adjusted. Additional manual adjustment using the {Brightness} thumbwheel
on the Color Settings card may be necessary for optimum results on the structures of
interest.

19
Volume Rendering

2 Options

2-1 Color or Black-and-white Shading

For details concerning the use of color or black-and-white shading, see paragraph 2
“Introduction”. For details of the available color controls, see paragraph 6 “VR Colors”.
• To select color shading mode for the Volume Rendering images, press the (Color) but-
ton. Press the button again to return to black-and-white shading mode.

Enhanced Resolution
In enhanced resolution mode, both horizontal and vertical resolution are increased by a
factor of two. This results in a fourfold increase in image processing time, therefore it is
advised to perform initial set-up and adjustments in standard resolution mode, then
switch to enhanced resolution at the end for final assessment of the image.
• To select enhanced resolution mode, press the (Enhanced resolution) button. Press
the button again to return to standard resolution.
Note: On oblique views, enhanced resolution is selected automatically, regardless of the
setting of the enhanced resolution mode button.

Enhance Contrast
The Enhance contrat parameter allows you to automatically modify the W/L to intensify
contrast on black and white images. There are two levels of intensification, depending on
what the image requires: mild or strong.
• To modify the contrast on black and white images, click on the arrow of the (Enhance
Contrast) button. A pullright menu opens up. Click on the (Mild) or the (Strong) but-
ton, depending on the level of contrat your image needs. Click on the (None) button to
return to a standard visualization.

This type of setting will alter the linear relationship between the
CAUTION
original dataset and the VR volume (in particular, as far as W/L is
concerned).

Partial Volume Filter

The “partial volume” effect occurs at the interface between tissues with widely separated
densities, as described in paragraph 2 “Introduction”. The Partial Volume Filter allows you
to attenuate this effect.
• To activate the Partial Volume Filter, press the (Partial Vol. Filter) button. Press the
button again to turn the filter off.
This filter is not available on oblique views. Setting the Partial Volume Filter to “on” has no
effect on oblique views.

20
 Volume Rendering

2-2 Save Preset

To save the current Volume Rendering parameters as a new preset:
• Click on the (Save Preset) button to opens the Save Preset window.
• Select the anatomical category under which you want to save the preset from the
menu associated with the (Anatomy) button.
• Move the mouse pointer into the “Name” field and type the name for the new preset.
Use only alphanumeric characters or spaces and avoid long names.
• Move the mouse pointer into the text field and type any comments you may want to
add.
To move to a new line, use the Enter key. To delete text in front of the cursor, use the
BackSpace or Del key.
To move the text cursor (e.g. for corrections) move the mouse pointer to the new posi-
tion and click.
Do NOT use the characters / or \ or “ in the text.
• To cancel the save operation, press the (Cancel) button. The preset is not saved.
• To save the new preset, press the (Save) button.

21
Volume Rendering

3 VR COLORS, COLORS SHADING MODE

3-1 Shading Modes

As described in detail in paragraph 2 “Introduction”, two different shading techniques can
be used for Volume Rendering images:
• Black-and-white shading (voxel value averaging).
• Color shading (surface shading).
When the color shading mode is on (active), the image will be shaded based on the ori-
entation of the surfaces in the voxels that contribute to a given pixel, and colored accord-
ing to the value of these voxels. The colors are set up by means of the VR Colors panel:
see below.
When the color shading mode is off, the image will be shaded using the average of the
voxel values as weighed by their contribution to the pixel values.
To select the color shading mode:
• Click on the (Color) button either in the VR Presets or the VR Opacity panel. Click on
the button again to return to black-and-white shading.

3-2 Color sliders

• You set the colors to be used

for color shading by means of
20 130 240
the diamond-shaped “sliders”.
A color is assigned to each
HU HU slider. The arrows indicate the
0 100 200 300
currently selected slider.
• The “tick” mark on the lower
scale continuously shows the
voxel value at the current posi-
tion of the 3D cursor.

In the example above:

- All voxel values below 20 will be painted with the color assigned to the first
slider.
- Voxel values between 20 and 130 are painted with shades between the colors
assigned to the first and second slider.
- Voxel values between 130 and 240 are painted with shades between the
colors assigned to the second and third slider.
- All voxel values above 240 will be painted with the color assigned to the third
slider.

22
 Volume Rendering

• To change the setting of a slider, simply click on it and drag it left or right. The exact
setting of the slider is continuously displayed above it.

• To assign a color to a particular slider, first click on the

slider to select it, then in the color palette circle drag the
color marker (cross and arrows) to the desired color.

When you move a slider, the scale below the sliders is resized automatically. You can
also resize the scale manually by pressing and dragging on the scale markings.
While you move one of the color sliders, those voxels on the reformatted and MIP views
that are affected by that slider are highlighted by colored hatching. This will help you to
select the range of voxel values corresponding to the anatomy that you want to display.

3-3 Brightness
The use of a low opacity setting on the VR Opacity panel (see paragraph 5) may at times
result in a dark image.

• To modify the brightness of the image, press on the

{Brightness} thumbwheel and drag it up and down to
Brightness: 100%
the desired value, or click on the triangles above and
below the thumbwheel to increase and decrease the
value by steps.
On color images, the brightness value is automatically
adjusted when the opacity value is modified. However,
additional manual adjustment using the {Brightness}
thumbwheel may be necessary for optimum results on
the structures of interest.
With a high opacity setting, a high brightness value may
cause the brighter color values to saturate to white.

23
Volume Rendering

3-4 Color Style

The color style setting determines the type of transition between color shades, smooth or
sharp. Smooth color shade transitions are useful to show subtle voxel value differences.
A sharp transition between color shades makes multiple structures easier to see (with a
smooth transition, color coded borders between structures may appear blurry).
To select the appropriate color style:
• Click on the (Color Style:) button and select either [Ramp] or [Step] from the menu.
“Ramp” mode produces smooth color transitions between the values corresponding to
the slider settings.
In “Step” mode the color changes abruptly at the half-way point between two sliders.

Note: In Define Object mode (see paragraph 7), this feature is not available and the
option is displayed in grey.

3-5 Creating Color Sliders

You can at all times creat a new color slider:
• Click on the (Add Color) button to add a new color slider.
The new slider will be created at the position of the “tick” mark on the lower scale, i.e.
at the voxel value corresponding to the current position of the 3D cursor.
The color of the new slider is defined by the location of the marker on the Color Palette.

3-6 Deleting Color Sliders

You can delete a color slider if required, but only if there are more than three color sliders
on the panel.
To delete a color slider:
• Select the slider to be deleted, then click on the (Delete Color) button.
OR
• Click on the slider and drag it upwards, out of the color slider panel.

24
 Volume Rendering

4 VR COLOR, DEFINE OBJECTS MODE

4-1 Overview
The Define Objects mode allows you to define up to four separate “objects” on the view.
Each separate object is attached to a color slider. You define a separate opacity curve for
each object, using the controls on the VR Colors panel.
The contributions from the images created with each individual curve are blended to form
the final image.
In this mode you can:
• Adjust the range of voxel values that defines each object.
• Adjust the contribution of each object to the final image.
• Detach an object (remove it from the image).
Note: In this mode, the controls and settings on the VR Opacity panel (in particular the
opacity curve setting) have no effect.

4-2 Defining an Object

To define an object:
• Select the color slider to be used by clicking on it.
• Attach an object to it by clicking on the (Attach Object) button.
• Set the associated color by means of the color palette.
The opacity curve for the object will be displayed on the color editor. Color display mode
is automatically selected, but you may revert to black and white mode.

4-3 Adjusting the Range of Visible Voxels

Each object is defined by a range of

20 130 240
visible voxels. You can adjust this
range by dragging the slider to
HU HU which the object is attached, and
0 100 200 300
the two neighboring sliders.

For an object attached to a slider that is located between other sliders, the opacity curve
is defined so that it falls off halfway between sliders, with a degree of overlap. For an
object attached to the extreme left or right slider, the opacity curve extends to the begin-
ning or end of the range of voxel values, respectively (see figure).

25
Volume Rendering

4-4 Adjusting the Opacity of an Object

To adjust the contribution of an object to the final composite image:

• Select the corresponding slider by clicking on it.

Opacity: 100%
• Set the contribution of the object by means of the {Opacity}
thumbwheel. Press on the thumbwheel and drag it up and
down to the desired value, or click on the triangles above and
below the thumbwheel to increase and decrease the value by
steps.

An opacity value of 100 % will show the object fully, while a lower value will make it pro-
gressively less visible.
Note: The {Brightness} thumbwheel controls the brightness of the entire image, not of
the individual attached objects.

4-5 Detaching an Object

To detach an object (remove it from view):

• Select the slider to which the object is attached by clicking on it.
The legend on the (Attach/Detach Object) button will indicate (Detach Object).
• Detach the object by clicking on the (Detach Object) button.
The opacity curve associated with the object disappears from the VR Color panel, and
the object disappears from the view.
Note: As long as one or more objects are attached to the color sliders, the controls on
the VR Opacity control panel (curve type, opacity curve editor, opacity level) have
no effect. If you change the setting of any of these controls, a message is dis-
played, warning you that the changes will have no effect until you detach all
objects.

26
 Volume Rendering

Blank page.

27
Volume Rendering

GLOSSARY
Artifact- Feature in an image resulting either from the initial data acquisition or subse-
quent computer processing that does not correspond to a real feature in the original ana-
tomical structure. Also see Partial Volume Effect.
CT (Computed Tomography)- Process of deriving anatomic information by computer
synthesis of X-ray data.
CTA (CT Angiography)- The use of CT techniques optimized for angiography.
HU (Hounsfield Unit) - Scale unit denoting voxel density in a CT or MR data set.
MR, MRI (Magnetic Resonance Imaging) - Creation of images using the magnetic reso-
nance phenomenon. Current applications involve imaging the distribution of hydrogen
nuclei (protons) in the body.
MRA (MR Angiography) - The use of MR techniques optimized for angiography.
Partial Volume Effect - The appearance of voxels with intermediate values at the inter-
face (separating surface) between two tissue types with clearly distinct and different den-
sities.
Pixel - Abbreviation for “ picture element,” the smallest unit a computer screen can dis-
play.
Rendering - Techniques used to represent a three-dimensional object on a two-dimen-
sional surface.
SVOI (Spherical Volume of Interest) - On-screen tool using a sphere-shaped cursor to
define a volume of interest.
Voxel - Abbreviation for “volume element,” the basic element in a CT or MR data set.

28
 Table of Contents
Advanced Vessel Analysis
(optional)

CHAPTER 1 - GENERAL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-1

1 INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-2
2 HOW TO USE THIS DOCUMENT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-2
3 CONVENTIONS FOR THIS MANUAL . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-3

CHAPTER 2 - INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-1

1 OVERVIEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-2
2 PROTOCOLS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-3
3 VIEWS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-3

CHAPTER 3 - SUMMARY OF OPERATIONS . . . . . . . . . . . . . . . . . . . . . . . 3-1

1 OVERVIEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-2
2 SUMMARY OF OPERATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-2
2-1 Selecting and Loading Exam . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-2
2-2 Vessel Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-2
2-3 Editing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-3
2-4 Exploring the Exam . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-4
2-5 Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-4
2-6 Reports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-5
2-7 Customizing Protocols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-5
Advanced Vessel Analysis (optional)

CHAPTER 4 - PROCEDURES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-1

1 OVERVIEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-2
2 IMAGE REQUIREMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-2
3 USING PROTOCOLS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-4
4 PROCEDURES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-5
4-1 Select and Start . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-5
4-2 Step 1 - Define Section to Analyze . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-6
4-2-1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-6
4-2-2 Marking Points . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-6
4-2-3 Adding and Modifying Points . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-7
4-3 Step 2 - Identify and Edit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-8
4-3-1 Identify . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . .. . . . . . 4-8
4-3-2 Verify . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-9
4-3-3 Edit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . ... .. ... . 4-10
4-4 Step 3 - Select Section of Interest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-11
4-4-1 Overview . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . ... .. 4-11
4-4-2 Explore and Analyze . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . 4-12
4-4-3 Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-13
4-4-4 Adding and Modifying Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-15
4-5 Step 4 - Adding Quick Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-16
4-5-1 Overview . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-16
4-5-2 Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-16
4-5-3 Procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-17
4-6 Step 5 - Reports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-18

CHAPTER 5 - VIEWS AND CONTROLS . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-1

1 OVERVIEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-2
1-1 General . . . .. .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-2
1-2 Controls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-2
2 LUMAN VIEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-3
2-1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-4
2-2 Views . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-4
2-2-1 Geometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-5
2-2-2 Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-5
 Advanced Vessel Analysis (optional)

2-3 Graph . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-6

2-4 Cursor Line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-6
2-5 Active Annotations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-7
2-6 On-View Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-8
3 ACCESSORY VIEWS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-8
3-1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-8
3-2 Baseline Views . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-8
3-3 Oblique View . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-9
3-3-1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-9
3-3-2 Controls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-9
3-4 3D View . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-10
4 OTHER VIEWS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-11
4-1 Curved Views . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-11
4-2 Volume Rendering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-12
4-3 Navigator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-12

CHAPTER 6 - REPORTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-1

1 REPORTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-1
1-1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-1
1-2 Report Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-1
1-2-1 Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-2
1-2-2 Images . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-2
1-2-3 Screen Saves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-2
1-2-4 Annotations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-3
1-3 Procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-3
2 MEASUREMENTS ACCURACY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-4
2-1 Voxel Dimensions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-5
2-2 Geometrical Accuracy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-5
2-3 Quantification Algorithm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-6
2-4 Acquisition Accuracy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-7
2-5 Partical Volume Effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-7
2-6 Display Settings and Display Resolution . . . . . . . . . . . . . . . . . . . . . . . . . 6-7
Advanced Vessel Analysis (optional)

CHAPTER 7 - PROTOCOLS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-1

1 PRE-DEFINE PROTOCOLS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-2
1-1 Stenosis Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-2
1-2 Carotid Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-2
1-3 Aorta Analysis(Abdomen) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-2
2 CUSTOM PROTOCOLS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-3
2-1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-3
2-2 Configuring Vessel Definition (“Tracking”) Points . . . . . . . . . . . . . . . . . . . 7-3
2-3 Configuring Measurement Points and Related Measurements . . . . . . . . . 7-6
2-3-1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-6
2-3-2 Configuring Measurement Points . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-6
2-3-3 Configuring Related Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-9
2-4 Saving the Protocol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-12
2-5 Deleting a Protocol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-13

GLOSSARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
CHAPTER 1 Advanced Vessel Analysis

CHAPTER 1 - GENERAL

Advanced Vessel Analysis is an optional software extension of the Volume Anal-

ysis application for Image Work systems.
It takes the form of new “Vessel Analysis” protocols available within the Image
Works Volume Analysis application.
This manual describes the Advanced Vessel Analysis option. For information
concerning the Volume Analysis application, refer to the corresponding user
documentation.
This chapter lists the system requirements, notes on how to use the software, a
brief summary of the contents of the manual, and the typographical conventions
used.

The equipment on which this application runs includes

CAUTION one or more hard disk drives which may hold medical data
related to patients. Such equipment may in some coun-
tries be subject to regulations concerning the processing
of personal data and the free circulation of such data.
It is strongly recommended that access to patient files be
protected from all persons not in medical attendance.

Advanced Vessel Analysis is an optional software exten-

CAUTION sion of the Volume Rendering application and uses sev-
eral of the features of this application. When using
Advanced Vessel Analysis you should be fully familiar
with these features and the relevant safety information
from the Volume Rendering user documentation.

1-1
Advanced Vessel Analysis

1 INTRODUCTION
Advanced Vessel Analysis is an optional software extension of the Image Work Vol-
ume Analysis application on Image Work systems.
This manual is a supplement to the Volume Analysis user guide and
describes only the use of the Advanced Vessel Analysis option.
For all information concerning the use and functions of the Volume Analysis
application itself, refer to the Volume Analysis and Volume Rendering user
documentation.

2 HOW TO USE THIS DOCUMENT

• For an overview of the features and functions of the Advanced Vessel Analysis
software, read Chapter 2 “Introduction”,
• For a summary of the successive operations you perform when using Advanced
Vessel Analysis for the analysis of 3D angiography data, see Chapter 3 “Sum-
mary of Operations”,
• To familiarize yourself with the procedures to load an exam and start Advanced
Vessel Analysis from the Volume Analysis application, then process the exam,
refer to Chapter 4 “Procedures”,
• For a description of the “Lumen” (Advanced Vessel Analysis) view and available
controls, and a summary of the other Volume Analysis view types that you will be
using with Advanced Vessel Analysis, consult Chapter 5 “Views and Controls”,
• To use the information (measurements and images) in the filmed/saved
Advanced Vessel Analysis reports, and in particular to assess the measurement
accuracy, refer to Chapter 6 “Reports”,
• To modify existing Advanced Vessel Analysis protocols and save them as new
custom protocols, refer to Chapter 7 “Protocols”,
• For a list of the terms in this manual that are specific to the application, refer to
the Glossary at the back of the manual.

3 CONVENTIONS FOR THIS MANUAL

Throughout the text in this manual, certain type styles and symbols are used to dif-
ferentiate between one tool or graphic and another:
• Menu and control panel titles appear in bold face: Application menu.
• Menu options appear in bold face, within square brackets: [Exit].
• Graphical buttons appear in bold face, within parentheses: (View).

1-2
 Advanced Vessel Analysis

• On-screen tools appear in bold face, within braces: {Scroll Bar}.

• On-screen prompts and messages appear in italics: Login: .
• User typed-in responses appear in bold face italics: sdc.
• Keys on the workstation keyboard appear within angle brackets: <Enter>.
• A key combination such as <Ctrl-T> indicates holding down the <Ctrl> key,
pressing the <T> key, then releasing the <Ctrl> key.
• Mouse buttons are underlined: left
Whenever this manual refers to “clicking”, “selecting”, “pressing a button”, etc. with
the mouse, this always means using the left mouse button, unless the middle or
right mouse button is specifically mentioned.
Safety notice legends:

THIS INDICATES A POTENTIALLY HAZARDOUS SITU-

ATION WHICH, IF NOT AVOIDED, COULD RESULT IN
WARNING! DEATH OR SERIOUS INJURY.

This indicates a potentially hazardous situation which, if not

CAUTION avoided, may result in minor or moderate injury.

This indicates a non-hazardous situation which, if not

NOTICE avoided, could result in equipment damage, lost time, or
reduced image quality.

1-3
Advanced Vessel Analysis

Blank page.

1-4
CHAPTER 2 Advanced Vessel Analysis

CHAPTER 2 - INTRODUCTION
Advanced Vessel Analysis is an optional software extension of the Volume
Analysis application for CT systems.
It is used for quantitative analysis, diagnosis and therapy planning of vascular
diseases using 3D angiography data.
You define the vessel to be analyzed by marking successive points in the 3D vol-
ume. Advanced Vessel Analysis then automatically computes the centerline of
the vessel, and the “Lumen” view shows an “unfolded” view of the vessel along
the centerline. The view can be rotated interactively around this centerline.
You can now use the Lumen view together with the other view types available in
Volume Analysis to identify and examine pathologies such as a stenosis or
aneurysm. In particular, the associated 3D view is “clipped” automatically so as
to show only the vessel, and two additional oblique view modes allow you to dis-
play either the cross section of the vessel at right angles to the centerline at the
location of the 3D cursor, or the corresponding longitudinal section. You can
also use the Navigator and Volume Rendering.
You can perform cross section area and diameter measurements by placing
measurement points along the centerline. Measurements can be shown both as
absolute values, and as percentages relative to a reference location.
An automatic report generation feature allows you to rapidly record the results of
your analysis.

2-1
Advanced Vessel Analysis

1 OVERVIEW

Advanced Vessel Analysis has been developed as a tool for rapid, consistent and
repeatable quantitative analysis of 3D angio data (CT Angio).

• The user identifies the vessel to be analyzed by marking points inside the vessel
(start and end of the section, and one or more intermediate points as required).
• The software then automatically detects the vessel centerline and computes
cross section area and mean and minimum diameter at each point.
• Using the “Lumen”, 3D and oblique views (see below) the user can now interac-
tively explore the anatomical context, and if necessary edit the results from the
automatic centerline and cross section computation. Vessel dimensions (diame-
ter, cross section area) are displayed as a curve in the Lumen view, and the cor-
responding numerical values can be read at any given point by moving a cursor
line.
• To produce a report, the user marks significant points on the views, such as start
and end of an aneurysm or stenosis, position of supplied/supplying vessels,
bifurcations, etc. The report will contain both the measurement values concern-
ing these points and a full set of correlated images (including 3D and oblique
views). Once the user has marked the measurement points, the report is gener-
ated automatically by the software. Reports can be filmed and/or saved on the
workstation.
By performing the “mechanical” tasks of vessel analysis (centerline detection,
quantification, report generation) automatically, the Advanced Vessel Analysis soft-
ware allows you to:
− Analyze 3D angio data rapidly and consistently,
− Improve the repeatability of the analysis, in particular between pre and post-sur-
gical exams, but also between different users,
− Produce reports faster, and in a standardized format and style that will help the
referring physician to interpret and use the measurements and images in a con-
sistent manner.

2-2
 Advanced Vessel Analysis

2 PROTOCOLS
The Vessel Analysis protocols consist of comprehensive instructions, combined
with the tools needed to perform the analysis. They are contained in protocol pan-
els, in the same manner as other protocols in Volume Analysis .
This minimizes the need to memorize the procedures, or to have to refer continu-
ously to the user documentation.
Advanced Vessel Analysis is supplied with a set of pre-defined protocols.
The user can use these protocols directly, or modify and edit them to adapt them to
his own specific requirements, then save such customized protocols under a new
name.

3 VIEWS
Advanced Vessel Analysis introduces three new view types, in addition to those
already available with Volume Analysis:
− The “Lumen” view shows the complete length of the selected vessel, “unfolded”
along the centerline,
− The oblique cross section view (X-section) shows a true cross section of the ves-
sel (perpendicular to the centerline) at any given point,
− The oblique longitudinal section view (L-section) shows a longitudinal section of
the vessel (parallel to the centerline) at any given point,
Advanced Vessel Analysis generates a special 3D view showing only the selected
vessel. Curved reformatted views that use the centerline as the reference trace
can also be generated.
The oblique X-section and L-section views can be positioned interactively at any
point along the centerline using the 3D cursor. The Lumen, oblique L-section and
curved views can all be rotated interactively. This allows you to rapidly and com-
pletely explore the anatomical context (lesions, calcifications, branches, nearby
anatomy such as ostiums).

The geometrical transformations necessary to “unfold” the

CAUTION vessel, or, in other words, to transform the 3-dimensional
centerline into a straight line, result in a degree of geometri-
cal distortion of the anatomical features in the Lumen view.
The 3D view is “clipped” to clearly show the vessel to be
analyzed among the surrounding tissue, and may not
include all parts of the vascular anatomy.
For these reasons the Lumen and 3D views should be used
for orientation purposes only, NOT for diagnostics.

2-3
Advanced Vessel Analysis

Blank page.

2-4
CHAPTER 3 Advanced Vessel Analysis

CHAPTER 3 - SUMMARY OF OPERATIONS

This chapter summarizes the sequence of operations you will use with
Advanced Vessel Analysis:
− Selecting and loading the exam,
− Selecting a vessel for analysis,
− Editing the result of the automatic vessel identification and quantification, if
necessary,
− Exploring the anatomical features and possible pathologies in the selected
vessel,
− Marking the locations where measurements are required in the final report,
− Filming or saving the report.

The procedures are described in full in Chapter 4 “Procedures”, the different

views and controls in Chapter 5 “Views and Controls”.
Chapter 6 “Reports” lists the measurement and images that will be included in
the reports, and discusses measurement accuracy.
The customizing of an existing protocol and saving it as a new protocol are
described in Chapter 7 “Protocols”.

3-1
Advanced Vessel Analysis

1 OVERVIEW
Advanced Vessel Analysis is based on the use of protocols.
Each protocol consists of a number of steps. The panels associated with each step
contain both the instructions on how to perform each step and the necessary con-
trols (buttons, menus, etc.).
This chapter summarizes the operations you will use with Advanced Vessel Analy-
sis.

2 SUMMARY OF OPERATIONS
2-1 Selecting and Loading the Exam,
To start using Advanced Vessel Analysis:
• Select the exam, and if necessary the series within the exam, in the Image Work
Browser.
Advanced Vessel Analysis accepts CT modality exams.
• Click "VA" button then Start Volume Analysis .
• Select the protocol category, then the appropriate Vessel Analysis protocol. This
can be one of the pre-defined protocols supplied with the application, or a cus-
tom protocol defined by the user.
See Chapter 4 “Procedures” for details.

2-2 Vessel Selection

At this stage Step 1 of the protocol is displayed together with three baseline views
(axial, sagittal, coronal) in the case of a CT exam.
You are now prompted by the protocol to place the points that are necessary to
define the section of the vessel to be analyzed.
• To move through the 3D volume you use the same controls as with Volume Anal-
ysis: 3D cursor, active annotations and keyboard arrow keys.
• To place each point, you click with the left mouse button.
• The protocol will prompt you for each point in succession, but when necessary
you can return to a point, clear it and place it again.
When all points are placed to your satisfaction, you can switch to the next step in
the protocol.
The protocol prompts you to mark those points that are conventionally associated
with a specific type of analysis (stenosis, aneurysm). You can, however, configure
a protocol by adding, modifying or removing intermediate points and branch points
as required for a particular type of pathology. Such a re-configured protocol can

3-2
 Advanced Vessel Analysis

either be applied to the current session only, or saved and re-used as a new custom
protocol.

2-3 Editing
• In Step 2 of the protocol, select (Accept).
This starts the automatic process of computing the vessel centerline (identification)
and vessel cross section at each point along the centerline (quantification).
When the process is complete, a 3D view, the “Lumen” view and an oblique cross
section (X-section) view are displayed. The “Lumen” view shows the “unfolded”
vessel, and the oblique X-section view shows the vessel cross section perpendicu-
lar to the centerline at the position of the cursor.
• At this stage, you should verify, that the software has:
− Identified the vessel correctly.
− Quantified the vessel cross section at each location correctly.
The limited resolution of the data set, and features such as closely adjacent ves-
sels, may occasionally lead to the software identifying the wrong vessel. Use the
available views to check that the displayed centerline does correspond to the
desired vessel.
The limited resolution may also lead to cross section quantification errors. These
can be identified on the oblique X-section view as irregularities in the contours that
do not match the vessel outline. You can check the successive cross sections by
moving the 3D cursor along the centerline.
• Both the centerline and the cross section contours can be edited at this stage if
necessary (see Chapter 4 “Procedures” for details).
Note that editing the centerline can be effective for minor errors. If a more signifi-
cant error occurs, e.g., selection of a wrong branch, it is usually easier to repeat
Step 1 (paragraph 2-2) and add intermediate points if required.
• If you have edited the centerline or the cross section contours, repeat the compu-
tation by again selecting (Accept) before going to the next step.

3-3
Advanced Vessel Analysis

2-4 Exploring the Exam

Once you have verified the correct identification of the vessel and the cross section
quantification, you dispose of a range of tools and controls to explore the anatomi-
cal features and possible pathologies in the selected vessel:
− You can move to any location in and around the vessel using the 3D cursor,
− The 3D view is “clipped” so as to clearly show the selected vessel among the
surrounding tissue. The centerline and cross section at the current cursor posi-
tion are also shown.
As in Volume Analysis, you can rotate the displayed 3D volume in all directions,
change the rendering mode, use cut planes, etc.
− The Lumen view shows the “unfolded” vessel, and a graph alongside the view
can show diameter or cross section area as required. You can rotate the Lumen
view around the centerline, and adjust the width of the view to display more or
less of the adjacent anatomy.
− By default, the oblique view shows the vessel cross section perpendicular to the
centerline at the position of the cursor (X-section mode). It can also show a lon-
gitudinal cross section (L-section mode) tangent to the centerline at the position
of the cursor. The L-section view can be rotated around the centerline.
− While using Advanced Vessel Analysis, you can switch any of the views to one of
the other types available in Volume Analysis, such as baseline and oblique refor-
matted views, curved reformatted views, Volume Rendering and Navigator
views, etc.
See Chapter 5 “Views and Controls”.

2-5 Measurements
• In Step 3 you mark the locations where measurements are required in the final
report.
The protocol prompts you to place measurement points. Measurements linked to
these points such as diameters and cross section area, and measurements
between these points such as distance, angle and volume, will be included in the
report.
See Chapter 4 “Procedures”.
• The protocol prompts you for those measurement points and measurements that
are conventionally associated with a specific type of analysis (stenosis, aneu-
rysm). You can, however, configure a protocol by adding, modifying or removing
measurement points and measurements as required for a particular type of
pathology. Such a re-configured protocol can either be applied to the current ses-
sion only, or saved and re-used as a new custom protocol.
See Chapter 7 “Protocols”.

3-4
 Advanced Vessel Analysis

• To quickly add one or more measurements to be used only in the current ses-
sion, you can use Step 4 of the protocol. In this step you can define measure-
ment points and measurements (diameter, area, distance, angle volume) that will
be included in the current report, but cannot be saved in a custom protocol.
See Chapter 4 “Procedures” paragraph 4-5 “Adding Quick Measurements”.

2-6 Reports
• In Step 5 of the protocol you can examine in tabular form all measurement infor-
mation that will be included in the report. At this stage you can still return to the
earlier steps in the protocol and make changes, add measurements, etc. as
required.
• Once you are satisfied with the report contents, you can either film (print) the
report, or save it as a new series on the image disk of the workstation.
The filmed or saved report consists of two main parts: the tables of measurement
results and a set of images (screen captures) that is linked to the measurements.
The number and type of images in the report may differ depending on the protocol,
but generally includes 3D, Lumen, oblique and curved reformatted images.
• During the analysis you can perform additional screen captures and “queue” the
resulting images. These will then be added automatically at the end of the filmed
or saved report.
See Chapter 4 “Procedures” and Chapter 6 “Reports”.

2-7 Customizing Protocols

Advanced Vessel Analysis is supplied as a set of pre-defined protocols.
You can use these protocols as they are, or use them as a starting point to config-
ure your own custom protocols, to account for individual differences in patient
pathology.
See Chapter 7 “Protocols”.

3-5
Advanced Vessel Analysis

Blank page.

3-6
CHAPTER 4 Advanced Vessel Analysis

CHAPTER 4 - PROCEDURES
Advanced Vessel Analysis accepts CT exams. These have to meet certain
requirements.

Advanced Vessel Analysis consists of of a set of pre-defined protocols, each for

a particular type of vessel analysis. Each protocol contains the successive steps
required to perform an analysis: vessel selection, vessel identification and quan-
tification, selection of sections of interest (measurement points), adding quick
measurements, report review, and saving or filming of the report.

The user can configure a protocol, then save it as a new custom protocol.

This chapter sets out the image requirements and describes the use of the pro-
tocols in general and the procedures in each protocol step in particular.

To define custom protocols, refer to Chapter 7 “Protocols”.

4-1
Advanced Vessel Analysis

1 OVERVIEW
The CT exams used with Advanced Vessel Analysis have to meet certain image
requirements. These are detailed in paragraph 2 “Image Requirements”.
The Advanced Vessel Analysis protocols have been developed for the analysis of
blood vessel pathologies. Each protocol provides you with the necessary tools and
instructions for a specific analysis such as a stenosis or aneurysm. See paragraph
3 “Using Protocols”.
The procedures you will use to start Advanced Vessel Analysis and perform the
analysis through the successive steps in the selected Vessel Analysis protocol are
described in paragraph 4 “Procedures”.

2 IMAGE REQUIREMENTS
Advanced Vessel Analysis accepts CT image sets. These must meet the same
image requirements as those for the basic Volume Analysis application.
A CT image set to be used with Advanced Vessel Analysis should meet at least the
following requirements:
• Field of view, matrix size and display center must be the same for all images in
the set,
• Orientation and gantry tilt should be the same for all images in the set,
• Tilted acquisitions are not supported for right/left decubitus patient orientations,
• There MUST NOT be two images corresponding to the same location in the set,
• The set should include only AXIAL, SAGITTAL, CORONAL or OBLIQUE images.
Other types such as screen saves, etc. are not supported,
• Inter-slice distance MUST be less than 10 mm.
When loading a CT image set, the Advanced Vessel Analysis software uses the
FIRST selected image in the Browser as a basis for using/discarding the other
images selected for building a 3D model. For example, any images having a matrix
size or gantry tilt different from that of the first selected image in the Browser are
discarded.
Although Advanced Vessel Analysis will accept and load CT exams with inter-slice
distances up to 10 mm, in practice exams with such large inter-slice distances are
not suitable for vessel analysis, because many vessels that are essential for the
analysis have diameters that are significantly less than 10 mm. Inter-slice dis-
tances in the order of 1 to 2mm are generally acceptable.

4-2
 Advanced Vessel Analysis

It remains the responsibility of the physician to determine

CAUTION whether the inter-slice distance of a given exam is accept-
able for use with Advanced Vessel Analysis.
Always bear in mind that, within an exam, details with
dimensions in the order of or less than the inter-slice dis-
tance cannot be identified with an acceptable degree of reli-
ability.

It is recommended NOT to use Advanced Vessel Analysis in

CAUTION cases where an opacified catheter is present in the vessel
segment to be analyzed (see Chapter 6 “Reports”, paragraph
2-3).

4-3
Advanced Vessel Analysis

3 USING PROTOCOLS
You use the Advanced Vessel Analysis protocols for rapid and consistent analysis
of blood vessel pathologies such as aneurysms and stenoses, and for the planning
of interventions such as the placement of a stent, or angioplasty.
A protocol provides you with the necessary tools and instructions for a specific
analysis such as a stenosis or aneurysm, either in a single vessel or in a vessel
with bifurcations.
All protocols follow the same overall procedure:
• Definition of the vessel(s) to be analyzed,
• Automatic computation of the vessel centerline and cross section,
• Validation of the results, and editing of the centerline and cross section if
required,
• Placing measurement points,
• Reviewing the measurements, and saving and/or filming the results.
The Advanced Vessel Analysis package is provided with a set of pre-defined proto-
cols. You can use these as supplied, or use them as “templates” to configure your
own custom protocols.
Because each protocol is defined for a particular task, the protocols differ individu-
ally, in particular in the instructions on how and where to place the points required
to define the vessel(s) and the measurement points. These differences are clearly
set out and documented in the protocols themselves. Therefore, this User Guide
describes the common features of the Advanced Vessel Analysis protocols such as
the overall procedures, available controls, views and techniques, but it does not
describe each separate protocol in full detail.
In most cases you will be using the views and tools available in the Advanced Ves-
sel Analysis protocols. However, all the other standard and optional tools, view
types, etc. that are part of the Volume Analysis application remain fully available.

4-4
 Advanced Vessel Analysis

4 PROCEDURES
4-1 Select and Start
You use the Advanced Vessel Analysis protocols in the same manner as the other
protocols available in Volume Analysis:
• Select the exam to be analyzed in the Image Work Browser,click the button of
"VA" and start Volume Analysis,
• Select the protocol category (anatomical region),
• Select the appropriate Vessel Analysis protocol,
• Follow the instructions in the protocol panels. The Vessel Analysis protocols all
consist of five steps, described below.
Note: When the Advanced Vessel Analysis option is installed, you can also define
a 3D trace within Volume Analysis, then display a “Lumen” type view based
on this trace without starting a Vessel Analysis protocol, either from the (Dis-
play Tools) > [View Type] menu, or by selecting “Lumen” in the view type
active annotation on-view menu (see the Volume Analysis user documenta-
tion for details).
You can manipulate (resize, rotate) such a Lumen view in the same manner as in a
Vessel Analysis protocol, but you cannot perform any measurements on such a
view, so it is of limited usefulness.

4-5
Advanced Vessel Analysis

4-2 Step 1 - Define Section to Analyze

Current point label and abbreviation

(tick mark indicates point has been
placed)
Instructions

Current point (to make corrections, Clear point

use arrow buttons to select the point)
Configure (customize) protocol

Close protocol Go to next step in protocol

4-2-1 Overview
At startup, for a CT exam the three baseline views (axial, sagittal and coronal) are
displayed. You are prompted by the protocol to place a number of points to define
the section of the vessel to be analyzed (pencil-shaped mouse cursor).
You can also configure the points contained in the protocol to define and save a
new custom protocol.

4-2-2 Marking Points

You move the 3D cursor as usual by holding down the <Shift> key and moving the
mouse on the views. To move the plane of a view “back and forth”, you can also
use the <left> and <right> arrow keys on the keyboard or the active annotations.
To mark each point, click with the left mouse button. Note that the point is marked
at the current location of the mouse cursor, not of the 3D cursor. You do not need
to mark the points exactly in the center in the vessel, but they should clearly desig-
nate the vessel(s) to be analyzed. After placing each point, the protocol automati-
cally prompts you for the next point.

4-6
 Advanced Vessel Analysis

You can adjust window W/L on the views as needed: the window W/L settings do
not affect the subsequent automatic vessel analysis computation.

• In each case you will have to define the start and

Point 1/3 the end of the section. Depending on the protocol,
you will be prompted for other points, such as bifur-
cations (branches) and optional intermediate
points. You can skip one or more of these other
Clear Point points if you do not require them for the current
analysis. Click on the right arrow next to the Point
... legend to skip a point.
Back Next
• To correct a point, select it with the arrow buttons
at either side of the Point ... legend, then select
(Clear Point) and place the point again.
• After placing all requested points, select (Next).

Note: When placing the points on the views, they are labeled automatically. If any
of the labels overlap or hide features on the views, you can click and drag on
them to move them to a different position.

4-2-3 Adding and Modifying Points

A protocol prompts you to mark the points that are associated with a specific type of
analysis. However, for a particular analysis you may at times want to add, modify
or remove one or more points (intermediate or branch points). To do so:

Configure
• Select (Configure Protocol).
Protocol This function allows you to configure the protocol to your require-
ments, by adding, modifying and removing points used to define
the vessel and also measurement points (see paragraph 4-4-4).
The resulting protocol can then either be applied to the current
session only, or saved and re-used as a new custom protocol.
See Chapter 7 “Protocols” for full details.

4-7
Advanced Vessel Analysis

4-3 Step 2 - Identify and Edit

4-3-1 Identify

Accept vessel selection and

start computation

Close protocol Go to previous or next step in protocol

• To start the vessel analysis computation, select (Accept).

Accept
The other buttons on the panel are inactive at this stage.
Advanced Vessel Analysis now first computes the centerline of the
vessel (identification), then the cross section at each point along
the centerline (quantification).

After vessel identification the displayed views switch from the initial baseline or 3D
views to a 3D view, a “Lumen” view, and an oblique view.
− In the 3D view, the 3D volume is clipped, so as to display only a “tube” along the
centerline that contains the vessel (volume of interest).
− In the Lumen view, the full length of the vessel section is displayed “unfolded”
along the centerline, together with a graph that can show either the section area
or the mean, minimum or maximum diameter at each point on the centerline. If
you are analyzing a vessel with bifurcations, an active annotation on the view
allows you to select which branch is displayed.
− By default, the oblique view is automatically oriented perpendicular to the vessel
centerline at the current cursor position (X-section mode). Alternatively, you can
orient the oblique plane parallel to the centerline (L-section mode).
The projections onto the views of the centerline, and of the cross section contour of
the vessel at the current cursor position, are displayed on all views except the
Lumen view.
Because of the way the Lumen view is generated, by “unfolding” the vessel as a
projection at each point of the vessel centerline, the Lumen view by itself is unsuit-
able for diagnostics. This also applies to the 3D view which is generated by “clip-
ping” the 3D volume so as to show the vessel clearly and therefore may not show
all of the vascular anatomy.

4-8
 Advanced Vessel Analysis

4-3-2 Verify

3D view Lumen view

Protocol panel Oblique cross section view

After vessel analysis computation is complete, you must verify the results.

In difficult situations, such as a vessel with a complex trajec-

CAUTION tory, or vessels touching each other in the images, the auto-
matic centerline algorithm may not always follow the correct
trajectory. The user should always examine the computed
trajectory before moving on to the next step in the protocol,
and if necessary correct the result either by modifying the
centerline trace or by adding intermediate points and repeat-
ing the centerline detection process.

Correct vessel quantification is critically dependent on such

CAUTION factors as acquisition image quality and voxel size (image
resolution and inter-slice distance), and anomalies may
occur because of limitations in the available data. It is the
responsibility of the user to verify the result of the vessel
identification and quantification process before using the
data for analysis.

Verify that the vessel centerline does correspond to the ves-

Edit Center sel(s) to be analyzed. In the case of complex vascular struc-
tures, it is possible for the automatic centerline computation
to follow an incorrect trajectory or to identify a wrong branch
in the structure.
In the case of minor corrections, you can use the (Edit Cen-
ter) function to locally edit the centerline as required. For
major corrections it is usually advisable to return to Step 1
and define the vessel again, adding intermediate points as
required.

4-9
Advanced Vessel Analysis

Verify the vessel quantification, as shown by the cross sec-

Edit Section tion contours on the oblique view in X-section mode and by
the graph in the Lumen view.
Anomalies may occur in the vessel quantification because of
limitations in the available data. You can remove such
anomalies by using the (Edit Section) function to locally edit
the cross section contour as required.
When analyzing a vessel with bifurcations, make sure to ver-
ify each branch. You can switch between branches by using
the active annotation on the Lumen view (see Chapter 5).

4-3-3 Edit

After editing, accept edit and

repeat computation

Edit centerline Edit cross section

Undo edit
Close protocol Go to previous or next step in protocol

When you edit the vessel centerline with (Edit Center) on

Edit Center the 3D view, the corrections are made in the plane perpen-
dicular to the current viewing direction. Rotate the 3D view to
make corrections in all three dimensions.

• To edit the vessel centerline, select (Edit Center), to edit

Edit Section a cross section contour, select (Edit Section). You can
now edit the centerline or contour by pointing the mouse
cursor onto the section of the trace you wish to modify
and holding down the left mouse button, then dragging
the trace to the desired shape.

• Select (Undo Edit) to revert to the previous trace if the

Undo Edit result of the edit is not satisfactory.

4-10
 Advanced Vessel Analysis

• After editing either the centerline or the cross section con-

Accept tours, select (Accept) again to update the vessel analysis
computation.

• Once you are fully satisfied with the results, select (Next)
Back Next
to go to Step 3.

4-4 Step 3 - Select Section of Interest

Current point label and abbreviation

(tick mark indicates point has been
placed) Instructions

Current point (to make corrections,

use arrow buttons to select the point) Clear current point

Select view type / view layout Configure (customize) protocol

Close protocol Go to previous or next step in protocol

4-4-1 Overview
Once vessel identification and quantification (Step 2) are satisfactory you can
explore and analyze the exam using the tools available in Volume Analysis.
To prepare the exam report you place a number of measurement points as defined
in the current protocol.
You can also configure the measurements in the protocol to define and save a new
custom protocol.

4-11
Advanced Vessel Analysis

4-4-2 Explore and Analyze

You explore and analyze the exam using the same range of tools as available in
Volume Analysis.

You will mostly use the 3D, Lumen and oblique views (X-
Advanced Views section and L-section views), but all the other Volume Anal-
ysis view types are also available. The (Advanced Views)
button in the Step 3 protocol panel allows you to rapidly
access the most frequently used view types:
• [VR]: switch the 3D view to Volume Rendering mode,
• [VOI]: return the 3D view to the original volume of inter-
est (“tube” containing the vessel),
• [X-Section]: switch the oblique view to X-section
(oblique cross section) mode,
• [Navigator]: display a Navigator view (upper left view),
• [MIP]: switch the 3D view to MIP rendering mode,
• [L-Section]: switch the oblique view to L-section
(oblique longitudinal section) mode.

You can move through the exam using the 3D cursor and the Lumen view cursor
line, and rotate the 3D, Lumen and oblique L-section views as required.
For a full description of the specific features and controls available with Advanced
Vessel Analysis (in particular on the Lumen and oblique views) refer to Chapter 5
“Views and Controls”.

4-12
 Advanced Vessel Analysis

4-4-3 Measurements
Overview
You use Step 3 to mark measurement points along the vessel centerline that define
the measurements to be included in the exam report. Each protocol will prompt you
to place the specific points that are relevant to the type of analysis being performed.
The points are annotated automatically by the software.
For each measurement point, the corresponding measurements (diameters, cross
section area) are added to the report. Depending on the protocol, measurements
between the points (distance, angle and/or volume of the vessel section between
the points) are also added.
In general, you will be prompted to mark the start and end points of the pathology to
be analyzed (aneurysm, stenosis), and its maximum or minimum cross section (this
last point may be computed automatically by the software, but you can define it
manually if required).
You may be prompted to place other points. Not all of these may be relevant to the
particular analysis being performed. You are not obliged to mark all points: if you
do not mark a particular point, the corresponding measurements will not appear in
the report.
As an example, the Stenosis Analysis protocol proposes a reference point, used to
compute relative measurements for diameters and area expressed in % of the ref-
erence (in addition to the absolute measurements in mm and mm2). If you do not
mark the reference point, these relative measurements will not appear in the report.
Similarly, depending on patient pathology, you may not always require the full set of
measurement points available in the Aorta Analysis protocol. Again, by not mark-
ing the points you do not need, you exclude the corresponding measurements from
the report.

Procedure
You mark the points in the same manner as when placing points to define the ves-
sel (Step 1). Mostly, you will be using the Lumen view, but you can use other views
as required.
You move the 3D cursor as before by holding down the <Shift> key and moving the
mouse on the views. To move the plane of a view “back and forth” on baseline and

4-13
Advanced Vessel Analysis

oblique views, you can also use the <left> and <right> arrow keys on the keyboard
or the active annotations.

• To mark each point, click with the left mouse button.

Point 1/4 Note that the point is marked at the current location of
the mouse cursor, not of the 3D cursor.
• You can skip one or more points if you do not require
Clear Point
them for the current analysis. Click on the right arrow
next to the Point ... legend to skip a point.
• To make corrections, use the arrow buttons at either
side of the Point ... legend to select the point, remove
it by selecting (Clear Point) and mark it again by click-
ing with the left mouse button. If required you can cor-
rect points placed automatically by the software (e.g.,
the minimum cross section of a stenosis) in the same
manner.

Note: When placing the points on the views, they are labeled automatically. If any
of the labels overlap or hide features on the views, you can click and drag on
them to move them to a different position.

4-14
 Advanced Vessel Analysis

4-4-4 Adding and Modifying Measurements

After placing the pre-defined measurement points, at times you may want to add
one or more measurements, or modify an existing one.

• To quickly add one or more basic measurements (such as

Back Next
diameter, area, distance, angle or volume) select (Next)
to go to Step 4 in the protocol. See paragraph 4-5 “Add-
ing Quick Measurements” below.

Configure
For more elaborate function, such as adding, modifying,
Protocol renaming and deleting measurements, select (Configure
Protocol).
This function allows you to configure the protocol to your
requirements, by adding, modifying and removing measure-
ment points and related measurements, and also points
used to define the vessel (see paragraph 4-2-3). The result-
ing protocol can then either be applied to the current session
only, or saved and re-used as a new custom protocol.
See Chapter 7 “Protocols” for full details.

4-15
Advanced Vessel Analysis

4-5 Step 4 - Adding Quick Measurements

Select measurement type,

then follow instructions

Modify previous measurements

Close protocol Go to previous or next step in protocol

4-5-1 Overview
In Step 3 of the protocol you were prompted to place a number of pre-defined mea-
surement points.
Step 4 of the protocol allows you to quickly add one or more measurements differ-
ent from those defined in the protocol. These measurements will be included in the
report.

If you do not require any such additional measurements, you

Back Next
can skip this step (click on (Next) to go to Step 5).

4-5-2 Measurements
You can add the following types of measurements:
• (Auto Diameter): mean, minimum and maximum diameter and cross section
area for a point along the vessel centerline,
• (Manual Diameter): distance between two points. This will usually be a diame-
ter (e.g., the actual vessel diameter in the presence of a thrombus where the
automatic diameter measurement is that of the lumen) but you can also use this
function to measure e.g., the distance between the vessel wall and nearby bone,
• (Length): distance between two points located on the vessel centerline and
measured along the centerline,
• (Volume): volume of the section of the vessel between two points,
• (2 Point Angle): angle between a section of the vessel, defined by two points,
and the vessel centerline,
• (3 Point Angle): angle between two sections of the vessel, defined by three
points. The measurement is the angle between the segment defined by the first
and second point and the segment defined by the second and third point.

4-16
 Advanced Vessel Analysis

4-5-3 Procedure
To add a measurement, select the type of measurement you require. You will be
prompted to place the necessary point(s). A default name (label) is automatically
added in the list in the protocol panel and a default abbreviation on the Lumen view.
To modify or remove a quick measurement:

Modify Previous Select (Modify Previous Measurements).

Measurement
A list of the existing quick measurements is displayed.

To modify a quick measurement:

• Select it in the list by clicking on the label.

Modify
• Select (Modify) to change the name or the abbreviation.
Film • Switch (Film) on or off to determine whether the images
associated with the measurement should be included in
the report.

To remove a quick measurement:

• Select it in the list by clicking on the label.

Remove
• Select (Remove) to delete the measurement from the list.

Note: To define additional measurements that are to be included in a custom pro-

tocol, you should use (Configure Protocol) in Step 3 (see Chapter 7 “Proto-
cols”). The quick measurements added during Step 4 are used only during
the current session, and will not appear in a custom protocol.

4-17
Advanced Vessel Analysis

4-6 Step 5 - Reports

Report contents:
exam data (illustrated) and
measurements (following pages)

Move through report pages Save report

Film report
Close protocol Go to previous step in protocol

Step 5 of the protocol allows you to review the exam and

Page 1/ 5 measurement information that will be included in the
report. Use the arrow buttons at either side of the Page
... legend to move through the pages.
Refer to Chapter 6 “Reports” for a description of the
report contents and a discussion of the accuracy of the
measurements included in the report.

If during the review you notice any measurements you may

Back Next
want to correct, you can return to Step 3 or Step 4 using the
(Back) button and make the necessary corrections (e.g.,
move a measurement point, or change an annotation).

You can save or film the report as soon as you are satisfied
Save with the measurement information that will be included in the
report.

• Select (Save) or (Film) in the protocol panel as required.

Film

If you are filming the report, you can select the film format
Film Format (layout of the images on the film in columns and rows).
• Open the (Film Format) menu and select [2x2], [2x3],
[2x4], [4x4] or [4x5] as required.

4-18
 Advanced Vessel Analysis

The software automatically composes and formats the report.

Close
Once the report is composed, you can select (Close) in the proto-
col panel to end the analysis, and select the next exam in the Study
List.

4-19
Advanced Vessel Analysis

Blank page.

4-20
CHAPTER 5 Advanced Vessel Analysis

CHAPTER 5 -VIEWS AND CONTROLS

Advanced Vessel Analysis introduces three new view types:
− Lumen view, showing the vessel “unfolded” along the centerline.
− Oblique cross section (X-section) view, showing the vessel cross section at
the position of the cursor.
− Oblique longitudinal section (L-section) view, tangent to the vessel centerline
at the position of the cursor.
The full range of views already available in Volume Analysis (such as 3D, base-
line, oblique and curved reformatted views, Navigator and Volume Rendering)
can also be used with Advanced Vessel Analysis.
This chapter describes the specific Advanced Vessel Analysis views and the
controls available on these views, and briefly summarizes the other views.

5-1
Advanced Vessel Analysis

1 OVERVIEW
1-1 General
During the initial step of the Vessel Analysis protocols (vessel selection), for a CT
exam the three baseline views (axial, sagittal and coronal) are displayed.
After vessel identification and quantification the view layout changes to a 3D, a
Lumen and an oblique view. The latter can be switched between vessel cross sec-
tion (X-section) and longitudinal section (L-section) mode.
You will mostly be using these view types to explore the exam and to place mea-
surement points. However, at any time during the analysis, you can use any of the
other view types available in Volume Analysis.

1-2 Controls
The controls on the views used with Advanced Vessel Analysis are mostly identical
to those that you will be familiar with from Volume Analysis. Refer to the Volume
Analysis user documentation. Controls that are specific to Advanced Vessel Analy-
sis are described in this chapter.

The window width and level (W/L) settings determine how

CAUTION clearly pathologies and other anatomical structures
present in the current 3D model can be discerned on the
views. Incorrect W/L settings may result in pathologies
and other essential anatomical structures not being dis-
played correctly, or even not being displayed at all.
A single W/L setting cannot always clearly display all fea-
tures present in an exam. Where necessary, use several
different settings to explore the exam data.
Also note that thresholding (see the Volume Analysis user
documentation) removes all voxels with values outside
the selected threshold range from the 3D volume.
Any pathologies or other anatomical features removed in
this manner can no longer be displayed, irrespective of
the W/L settings.

A 3D view is a two-dimensional projection on the screen

CAUTION of the 3D volume. There is no indication on a 3D view of
how “deep” inside the 3D volume the 3D cursor is located.
To prevent misinterpretation of the cursor location on the
axis perpendicular to the viewing position, the user
should always verify the cursor position by correlation
with the baseline and reformatted views.

5-2
 Advanced Vessel Analysis

To prevent misinterpretation when rotating oblique or 3D

CAUTION views, always use the orientation annotations and the ref-
erence image (orientation indicator) to keep track of the
orientation of the slice plane or view direction with respect
to the patient’s body (RAS coordinate system).

The user should be aware of the limited spatial and den-

CAUTION sity resolution of the clinical images processed with the
Advanced Vessel Analysis protocol. When manipulating
the data with different display modes, make sure that
pathologies and other essential anatomical structures
remain present.

2 LUMEN VIEW
2-1 Overview
The left side of the Lumen view shows the vessel “unfolded” along the centerline,
with the centerline straightened out. You can display either a reformatted “thin”
strip (section), or the “tube” around the centerline in MIP mode. If you have defined
a vessel with bifurcations, you can switch the display between the bifurcations.
The graph on the right side of the Lumen view can show the section area or the
mean, minimum or maximum diameter.

View type Patient name

Branch
Angle of rotation
around centerline
Linear position of cursor
from start of centerline Graph
Field of view
Rendering mode
(Rfmt or MIP)
Width of Lumen "stripe" Curve type
(section area, mean,
min or max diameter)

Lumen Cursor line and

measurement

Window Width/Level
Centerline (not displayed on view)
Lumen view

5-3
Advanced Vessel Analysis

A cursor line allows you to correlate the Lumen view with the other views.
The Lumen view can be rotated around the centerline, and you can change its
width if required.

2-2 Views

3D, Lumen and curved reformatted view

5-4
 Advanced Vessel Analysis

2-2-1 Geometry
At first sight, a Lumen view appears similar to a curved reformatted view, so that it
is important to be aware of the differences.
A Lumen view is constructed by transforming the three-dimensional centerline into
a straight line and then displaying for each point of the centerline the intersection
with the oblique plane perpendicular to the centerline (see illus.).
This transformation results in geometrical distortion of the displayed anatomical
features because the successive intersections are not parallel.
This is the reason why the Lumen view should NOT be used for diagnostic pur-
poses by itself, but always in combination with other views.
A curved reformatted view is generated by constructing a curved plane that inter-
sects the vessel centerline (or more generally a 3D trace), then “flattening out” this
plane for display on the screen. In this case the geometrical relations in the plane
are maintained (see illus.). Of course, due care should still be taken when inter-
preting a curved view, because it displays the intersection of the anatomical fea-
tures with a curved cut plane.

2-2-2 Measurements
You can display measurements of section area, mean, minimum and maximum
diameter at the cursor position on the Lumen view.
You can also use the tools available in Volume Analysis to perform measurements
on the Lumen view. However, because of the distorted geometry of the Lumen
view, only two types of measurements are possible:
− Distance measurements along a horizontal line (e.g., vessel diameter, or dis-
tance between vessel and nearby feature),
− Distance measurements vertically along the centerline.
Any other measurements are meaningless.
For other measurements, you use the Lumen view to ‘navigate’ along the vessel,
and place the measurement points on the cross section oblique view and/or base-
line views where their position in all three dimensions can be determined unambig-
uously.

5-5
Advanced Vessel Analysis

2-3 Graph
The graph on the right side of the Lumen view can show the section area or the
mean, minimum or maximum diameter.
Note: The section area at a given point is the true cross section of the lumen at
that point as computed by the software and as indicated by the green outline
on the cross section oblique view. The mean diameter at a given point is
defined as the diameter of a circle with the same area as the section area at
that point. The minimum and maximum diameter at a given point are
defined as the smallest and largest value, respectively, obtained when rotat-
ing the diameter measurement around the centerline.
You use the drop-down menu linked to the curve type active annotation on the
right of the view to select which value is displayed. From this menu you can also
rescale the curve to display a range of values around that corresponding to the cur-
rent cursor position, or reset the scale to display the full range.

2-4 Cursor Line

The horizontal graduated cursor line on the Lumen view indicates the intersection
with the plane containing the 3D cursor and displayed in the oblique X-Section
view.
To position the 3D cursor using the Lumen view, the view must be selected. You
can then move the 3D cursor as usual by pointing with the mouse while holding the
<Shift> key.
You can also move along the centerline in steps by means of the left and right arrow
buttons on the keyboard (the mouse should be positioned on the Lumen view or on
an oblique view).
On the Lumen view, the 3D cursor is shown as a red tick mark on the cursor line.
When you position the 3D cursor using one of the other views, the red tick mark on
the cursor line is not displayed.

5-6
 Advanced Vessel Analysis

2-5 Active Annotations

You use the active annotations of the Lumen view in the same manner as else-
where in Volume Analysis (see the Volume Analysis user documentation).
The use of the View Type, DFOV, W/L and Patient Name active annotations is iden-
tical.
If you are analyzing a vessel with bifurcations (e.g., right and left iliac arteries in the
aorta), you use the branch annotation to select which branch of the bifurcation is
displayed in the Lumen view. If you are analyzing a single vessel, this annotation
reads “End of Section”.
If you are also displaying a curved view that uses the vessel centerline as the refer-
ence trace (see paragraph 4-1), the curved view will switch automatically to display
the same branch as the Lumen view.
The Angle annotation allows you to rotate the Lumen view around the centerline of
the vessel.
The rendering mode annotation allows you to select either Reformat (Rfmt) or MIP
as the rendering mode for the Lumen view.
In Reformat mode the Lumen view shows the “strip” or “ribbon” along the centerline
that corresponds to the current angle setting, in MIP mode the view shows the
“tube” around the centerline.
The LP (linear position) annotation allows you to move the 3D cursor along the cen-
terline (the value shown is the 3D distance between the cursor line and the starting
point, measured along the centerline).
Note: You can also move along the centerline in steps by means of the left and
right arrow buttons on the keyboard (the mouse should be positioned on the
Lumen view or on an oblique view).
The Width annotation allows you to vary the width of the “strip” along the centerline
(or the diameter of the “tube” in MIP rendering mode), in order to observe features
further away from the centerline.
For the curve type annotation, see paragraph 2-3 above.

5-7
Advanced Vessel Analysis

2-6 On-View Menu

You use the on-view menu by clicking and holding right anywhere on the Lumen
view (except for the red annotations or the 3D cursor mark) and moving down to the
required menu item.
[Save Image]: saves the full contents of the view (Lumen view, graph, annotations)
as a screen save.
[Queue Report Image]: places the same information in a “queue” of screen saves,
that will be added at the end of a report, when it is saved or filmed.
[Clip to Width]: clips the associated 3D view to a “tube” around the vessel center-
line with the same diameter as the current width of the Lumen display (also see
paragraph 3-4).
[Enlarge]: enlarges the view, so that it takes up the entire viewing area. The menu
item changes to [Reset Size] to return to the original size (one quarter of the view-
ing area).
[Reset Pointer]: returns the 3D cursor to the center of the 3D volume.

3 ACCESSORY VIEWS
3-1 Overview
For vessel analysis, you initially use the three baseline views or set of 3D views
(depending on the exam type), then a 3D, Lumen and oblique view. The Lumen
view is described above (paragraph 2). The features of the other views that are
specific to their use within Advanced Vessel Analysis are described in this para-
graph.

3-2 Baseline Views

To select and mark a vessel for analysis, you initially use the three baseline views
(axial sagittal, coronal) with a CT exam, or a set of 3D views for a 3D XR exam.
After vessel identification and quantification the view layout changes to a 3D,
Lumen and oblique view. You can display one or more of the baseline views again
at any time if required. Once the vessel is defined, a projection of the current vessel
cross section (at the position defined by the cursor line on the Lumen view) is
shown also on the baseline views (green contour).

5-8
 Advanced Vessel Analysis

3-3 Oblique View

3-3-1 Overview
The oblique reformatted views that you use with Advanced Vessel Analysis are the
same as in Volume Analysis, with the addition of a number of automatic modes to
define the slice orientation:
− An X-Section view shows a reformatted slice that is perpendicular to the vessel
centerline and contains the 3D cursor.
− A basic L-Section view shows a reformatted slice tangent to the vessel center-
line at the location of the cursor line on the Lumen view, and parallel to the
Lumen view at that point. When the Lumen view is rotated around the centerline,
the L-section view rotates with it.
− The Best L-Section view is oriented so as to show as much of the vessel as
possible for a given cursor position (plane of maximum curvature). The Dmin
and Dmax L-Section views are oriented parallel to the plane containing the min-
imum and maximum vessel diameter, respectively, for a given cursor position.
The vessel cross section contour, and the minimum and maximum diameter are
shown in green on the X-section view.
Otherwise, the views are the same as in Volume Analysis, i.e. they are plane refor-
matted slices, either “thin” (one voxel thick) or “thick” (using MPVR, see Volume
Analysis user documentation).

3-3-2 Controls
You control the orientation mode by means of the section active annotation to the
right of the slice thickness/rendering mode at lower left of the view. Click on the
annotation and select the mode from the drop-down menu:
• [No Lock]: unlocks the orientation of the oblique view from that of the Lumen
view. You can now use the on-view cube or cross reference lines to change ori-
entation as with any conventional oblique reformatted view.
• [XSection]: perpendicular to vessel centerline
• [LSection]: tangent to vessel centerline
• [Best LSection]: tangent to vessel centerline in plane of maximum curvature of
the vessel
• [Dmin LSection]: tangent to vessel centerline in plane of minimum diameter
• [Dmax LSection]: tangent to vessel centerline in plane of maximum diameter
The [Align 3D Views] menu item in this menu allows you to align the 3D view(s)
with the current orientation of the oblique view.

5-9
Advanced Vessel Analysis

The LP (linear position) active annotation allows you to move the 3D cursor along
the centerline (the value shown is the 3D distance between the cross-section plane
containing the 3D cursor and the starting point, measured along the centerline).
Note: You can also move along the centerline in steps by means of the left and
right arrow buttons on the keyboard (the mouse should be positioned on the
Lumen view or on an oblique view).
On the X-section view, annotations showing the value of the minimum and maxi-
mum diameter are added on the view. If such an annotation masks essential fea-
tures of the image:
• Click and drag on the annotation to move it to a different position.
The on-view menu contains an additional item:
[Queue Report Image]: places the contents of the view in a “queue” of screen
saves, that will be added at the end of a report, when it is saved or filmed.
The other controls on the oblique views are the same as in Volume Analysis. This
applies both to the other active annotations (view type, image location, DFOV, slice
thickness, rendering mode for “thick” slices, W/L, patient name and orientation
annotations) and to the remainder of the on-view menu.

3-4 3D View
Initially the 3D view shows only the volume of interest (VOI), i.e., essentially a tube
around the centerline containing the vessel. If you have identified a vessel with
bifurcations, all branches are displayed (contrary to the Lumen view which can only
display one branch at a time).
The vessel centerline is shown on the view. A green contour indicates the position
of the cross section currently displayed in the oblique X-section view.
The on-view menu contains an additional item:
[Queue Report Image]: places the contents of the view in a “queue” of screen
saves, that will be added at the end of a report, when it is saved or filmed.
The other controls on the 3D view (active annotations, remainder of the on-view
menu, rotation) are the same as in Volume Analysis.
Additionally:
• To change the diameter of the volume displayed in the 3D view to the same value
as the width of the Lumen view, select [Clip to Width] in the Lumen on-view
menu,
• If you want to reset the 3D view to display the original volume of interest (VOI),
move to Step 3 of the protocol, select (Advanced Views), then select the [VOI]
view mode,

5-10
 Advanced Vessel Analysis

• To align the view direction in the 3D view with that of the current oblique view,
click on the section annotation on the oblique view and select [Align 3D Views]
in the drop-down menu (see paragraph 3-3-2).

The 3D view is intended for orientation purposes only, NOT

CAUTION for diagnostic use. In some cases it may not include all
parts of the vascular anatomy. In the case of CT images, it is
specifically NOT intended to display the tissue around the
enhanced lumen.

4 OTHER VIEWS
While using a Vessel Analysis protocol, you can at all times use one of the other
view types available in Volume Analysis, and in particular curved reformatted views,
3D views with volume rendering, and Navigator views.
For instructions on how to use these view types, refer to the Volume Analysis user
documentation.

4-1 Curved Views

Once you have identified a vessel centerline, you can display a curved reformatted
view that uses the centerline as the reference trace. In the case of a vessel with
bifurcations, the curved view will show the same branch as the Lumen view.
The controls on the curved view are the same as in Volume Analysis.

When filming or saving the result of a curved reformatting

CAUTION operation, always include the view on which you have
defined the trace in the record. Without this information it is
impossible to interpret a curved reformatted view correctly.

Note: When using Advanced Vessel Analysis this applies in particular to filming or
saving curved reformatted views separately, in addition to those included in
the report.
In the automatically generated report the curved views are always accompanied by
the corresponding 3D views showing the centerline trace.

5-11
Advanced Vessel Analysis

4-2 Volume Rendering

You can use the Volume Rendering mode on the 3D view(s) during vessel analysis.
The Volume Rendering controls are the same as in Volume Analysis.

When using Volume Rendering, you should be fully familiar

CAUTION with its functioning and the relevant safety information as
set out in the Volume Analysis user documentation. In par-
ticular:
An incorrect threshold setting, incorrect curve type (thresh-
old mode), or incorrect opacity setting, can result in essen-
tial anatomy, or pathologies, not being visible on the views
displayed in volume rendering mode.

4-3 Navigator
You can use the Vessel Analysis views (Lumen, oblique views) in correlation with
the Navigator view to perform a “virtual endoscopy”.
The Navigator controls are the same as in Volume Analysis .

When using the Navigator software, you should be fully

CAUTION familiar with its functioning and the relevant safety informa-
tion as set out in the Volume Analysis user documentation.
In particular:
Using an incorrect threshold setting or incorrect threshold
mode can result in pathologies or other essential anatomy
not being visible on the navigator views.
A navigator view is a two-dimensional projection on the
screen of a 3D volume. On such a view there is no indica-
tion of how “deep” inside the 3D volume the 3D cursor is
located. To prevent misinterpretation of the cursor location
on the axis perpendicular to the viewing position, the user
should always verify the cursor position by correlation with
the other views.
The Navigator software uses a conical projection (perspec-
tive). For this reason, the assessment of dimensions and
distances on navigator views tends to be subjective. The
user should always correlate apparent dimensions and dis-
tances on the navigator view with the other views.

5-12
CHAPTER 6 Advanced Vessel Analysis

CHAPTER 6 - REPORTS
When you have completed a vessel analysis, you can film the resulting report
and/or save it on the image disk.
The report contains the tables of measurements that you have defined in the
protocol, and a set of images associated with the measurements. It is com-
posed and formatted fully automatically, without requiring any user intervention
other than selecting “film” or “save”.
The accuracy of the measurements performed with Advanced Vessel Analysis is
dependent on a number of factors. Refer to paragraph 2 “Measurement Accu-
racy” in this chapter.

1 REPORTS
1-1 Overview

Report contents:
exam data (illustrated) and
measurements (following pages)

Move through report pages Save report

Film report
Close protocol Go to previous step in protocol

In Step 5 of the protocol you can examine in tabular form all measurement infor-
mation that will be included in the report. At this stage you can still return to the
earlier steps in the protocol and make changes, add measurements, etc. as
required.
Once you are satisfied with the report contents, you can film the report and/or
save it as a new series on the image disk of the workstation.
The process of composing and formatting the report is fully automatic.

1-2 Report Contents

The filmed or saved report consists of:
− The tables of measurement results.
− A set of images (screen captures) that is linked to the measurements.
− Any screen saves that have been “queued” during the analysis.

6-1
Advanced Vessel Analysis

1-2-1 Measurements
The first part of the report consists of a text page with the exam data (exam name,
patient name, hospital name, acquisition parameters) and text pages with tables of
measurement results.
The contents and format of these tables depend on the protocol, and also on the
exam type.
Note: If a reference point is defined in the protocol, measurements are also listed
expressed in % relative to the reference section, in addition to the absolute
values.
For both CT exams, the measurements (diameters, area) are computed from the
vessel quantification data (cross section contour) at the measurement point.

1-2-2 Images
The second part of the report consists of a set of images related to the measure-
ments.
The number and type of images in the report depend on the protocol and on the
number and type of measurements that have been added to the analysis.
They may include 3D, Lumen and curved reformatted images at several angles,
and oblique X-Section and L-Section images.
Extra annotations are added automatically on all images that allow you to deter-
mine unequivocally the location and orientation of each image, and to which mea-
surement the image refers.

1-2-3 Screen Saves

During the analysis you can “queue” additional screen captures by means of the
[Queue Report Image] menu item in the on-view menus (also see Chapter 5
“Views and Controls”). These will be added automatically at the end of a report,
when it is saved or filmed.

6-2
 Advanced Vessel Analysis

1-2-4 Annotations
The annotations (such as system annotations, text annotations, etc.) that are dis-
played on the screen will also appear on the filmed or saved images. Extra annota-
tions that identify the images are added automatically by the software (see above).
The saved images are of type SCPT (secondary captures). This means that all
annotations become part of the saved image and cannot be modified by another
viewing application at a later stage.
If the annotations on the screen have been turned off (see the Volume Analysis
user documentation) they will not be present on the filmed or saved images either.

Although images without annotation may be suitable

CAUTION for teaching purposes, diagnosis should not be per-
formed with such images.

While working on an exam, you can hide the patient name on the views for
increased confidentiality. If you have done so, make sure to show the patient name
again on the views BEFORE filming or saving images for diagnostic purposes.

When filming or saving images for diagnostic pur-

CAUTION poses, always make sure the patient name is displayed
on all views.

1-3 Procedure
You can film or save the report as soon as you are satisfied with the measurement
information that will be included in the report, as displayed in Step 5 of the protocol.
• Select (Film) or (Save) in the protocol panel as required.
If you are filming the report, you can select the film format (layout of the images on
the film in columns and rows).
• Open the (Film Format) menu and select [2x2], [2x3], [2x4], [4x4] or [4x5] as
required.
The software automatically composes and formats the report. The display will
cycle automatically through the successive images that will be included in the
report. Once the report is composed, you can select (Close) in he protocol panel to
end the analysis, and select the next exam in the Study List.
If you choose to film the report, it is sent to the same laser camera linked to your
workstation that you use to film images from Volume Analysis or other applications
Refer to the Image Works user documentation for details. Filming is performed as

6-3
Advanced Vessel Analysis

a background task: the workstation is available for the next exam as soon as the
data have been “queued” for transmission to the laser camera.
If you choose to save the report, it is saved on the image disk as a new series,
attached to the current exam. It consists of a set of secondary captures (type
SCPT) that can be reviewed using the Image Work Browser and filmed as required
at a later time.

2 MEASUREMENT ACCURACY

The software calculates and displays measurements

CAUTION with a resolution of one decimal place (such as 0.1
mm, 0.1 degree, etc.).
You should be aware that the real measurement accu-
racy is generally considerably less for a number of dif-
ferent reasons.
To assess the accuracy of the measurements per-
formed with Advanced Vessel Analysis, you should be
fully familiar with the section “Measurement Accuracy”
in the chapter “Measurements” of the Volume Analysis
User Guide.

This section summarizes the aspects concerning measurement accuracy that are
applicable in particular to the measurements performed with Advanced Vessel
Analysis.

6-4
 Advanced Vessel Analysis

2-1 Voxel Dimensions

The image set resolution, i.e., the dimensions of the voxels (volume elements) that
constitute the image set, is determined by the size of the field-of-view, the matrix
size and the inter-slice distance.
In a typical CT image set to be used for vessel analysis, the voxel cross section in
the acquisition plane will be in the order of 0.3x0.3 to 0.5x0.5mm for a 512x512
acquisition matrix and a field of view in the order of 15 to 25cm.
Ideally, voxels should be isotropic (with the same dimensions along all three axes),
i.e., the inter-slice distance should be the same as the voxel dimension in the acqui-
sition plane. In practice, however, considerations such as patient irradiation dose
levels will usually lead to the choice of a larger inter-slice distance.
To reliably identify and measure small diameter vessels, that are significant for the
analysis, an inter-slice distance in the order of 1 to 2mm is generally acceptable.
You should be aware that details with dimensions in the order of or less than the
inter-slice distance cannot be identified or measured with any degree of reliability.
Note: Volume Analysis accepts inter-slice distances up to 10mm, but such large
inter-slice distances are of little if any use for vessel analysis.

2-2 Geometrical accuracy

For CT image sets, the largest dimension of a voxel (normally the inter-slice dis-
tance) determines the geometrical accuracy:
− For a distance measurement, the geometrical accuracy of the displayed length is
equal to +/- largest voxel dimension,
− For an angle measurement, the geometrical accuracy depends on the length of
the segments, and improves as the length of the segments increases. As an
example, for an angle measured between segments which are five times larger
than the largest voxel dimension, the geometrical accuracy of the displayed
angle value is equal to +/- 10 degrees.
− For an area measurement, the geometrical accuracy of the displayed area value
is equal to +/- the circumference of the region of interest multiplied by (largest
voxel dimension)2 / 2.
The geometrical accuracy defines a lower bound on the overall accuracy that can
be obtained. Further limiting factors are the vessel analysis quantification algrithm,

6-5
Advanced Vessel Analysis

acquisition accuracy, partial volume effects, display settings and display resolution.

Distance, angle and area measurements are valid only if

CAUTION all segments defining the measurement are longer than
the inter-slice distance.

The geometrical accuracies defined above, when related

CAUTION to vessel measurements, are valid only if the vessel diam-
eter is larger than 2mm.

2-3 Quantification Algorithm

The vessel quantification algorithm computes the vessel cross section contours at
each point along the vessel centerline, on which the diameter, cross section area
and volume measurements are based.
This algorithm provides a best fit to the data available in the exam. However, the
user should be aware that the limited spatial and density resolution of the clinical
images processed with Advanced Vessel Analysis also imposes limitations on the
attainable measurement accuracy.
Correct vessel quantification is critically dependent on such factors as acquisition
image quality and voxel size (image resolution and inter-slice distance), and anom-
alies may occur because of limitations in the available data. It is the responsibility
of the user to verify the result of the vessel identification and quantification process
before using the data for analysis.

It is recommended NOT to use Advanced Vessel Analysis

CAUTION in cases where an opacified catheter is present in the ves-
sel segment to be analyzed.

2-4 Acquisition Accuracy

Any errors resulting from the acquisition process that are present in the original
image set (calibration and slice interpolation errors, motion artifacts) will be added
to the same extent to the measurement error.

6-6
 Advanced Vessel Analysis

2-5 Partial Volume Effects

In CT exams, the value of a voxel is the weighted average for all materials in the
voxel. Because of the high attenuation coefficient of calcium, even a small amount
of calcium in a voxel will weigh its value towards that of calcifications or bone, so
that the entire voxel appears to be calcifications or bone.
In vessel analysis, this partial volume effect will tend to make calcifications in the
vessels appear larger than they are in reality, and hence affect the measurement
accuracy.

2-6 Display Settings and Display Resolution

Since anatomical features are rarely of a uniform density, the apparent dimension
of an anatomical feature can change when you change the display settings (win-
dow width and level).
The measurements computed by the Advanced Vessel Analysis software are not
affected by this, because the software does not rely on the display settings for ves-
sel identification and quantification.
However, when you place the measurement points yourself, e.g., for an additional
diameter measurement, the apparent diameter can differ by one or more voxels
depending on the W/L settings, thereby adding another factor of uncertainty.

When four views are displayed on the workstation screen, each view has a display
resolution of 512x512 pixels, and you obviously cannot place a manual measure-
ment point with a precision better than a single pixel. Since most exams used for
vessel analysis are based on a 512x512 acquisition matrix, display resolution nor-
mally does not impose a further limitation on accuracy, the more so because in
most cases the display field of view (DFOV) for an analysis is smaller than the
acquisition field of view.

6-7
Advanced Vessel Analysis

Blank page.

6-8
CHAPTER 7 Adcanced Vessel Analysis

CHAPTER 7 - PROTOCOLS

Advanced Vessel Analysis is supplied with a set of pre-defined protocols, des-

tined for the general analysis of a stenosis, the specific analysis of a stenosis in
the carotids, and the analysis of an aneurysm in the aorta.
You can use these protocols as they are, or use them as a starting point to con-
figure your own custom protocols, to account for individual differences in patient
pathology.
This chapter briefly lists the pre-defined Vessel Analysis protocols, and
describes how to configure your own custom protocols as required, by adding
vessel definition points and adding, modifying or removing measurements to be
included in the saved or filmed report.

7-1
Advanced Vessel Analysis

1 PRE-DEFINED PROTOCOLS
1-1 Stenosis Analysis
The Stenosis Analysis protocol has been defined for the general analysis of a
stenosis.
The vessel definition points in this protocol comprise the start and end of the sec-
tion to be analyzed, and optional intermediate points.
The measurement points comprise a reference point, the start and end of the
stenosis, and the smallest section of the stenosis.
The resulting measurements in the report include the mean, minimum and maxi-
mum diameters and the cross section area at each of these measurement points
(expressed both in mm / mm2 and in % relative to the reference point) and the
length of the stenosis.

1-2 Carotid Analysis

The Carotid Analysis protocol (in the Neck protocol category) is similar to the gen-
eral Stenosis Analysis protocol, but has been configured specifically for the analy-
sis of a stenosis in the carotids.
The measurement points and measurements in the report are the same as in the
Stenosis Analysis protocol.

1-3 Aorta Analysis (Abdomen)

The Aorta Analysis protocol (in the Abdomen protocol category) has been defined
specifically for the analysis and therapy planning of an aneurysm in the aorta.
The vessel definition points in this protocol comprise the start of the section to be
analyzed, and branches such as the celiac trunk, superior mesenteric artery, right
and left renal and right and left external iliac (the latter point acts as the end of the
section).
The measurement points correspond to those conventionally used for analysis of
an aneurysm in the aorta, such as position of the renal arteries, start, end and larg-
est enhanced section of the aneurysm, locations of the aorta bifurcation and the left
and right iliac bifurcations.
The resulting measurements in the report include the mean, minimum and maxi-
mum diameter and the cross section area at each of these measurement points,
and basic distance measurements.

7-2
 Advanced Vessel Analysis

2 CUSTOM PROTOCOLS
2-1 Overview
When a patient’s pathology requires it, you can re-configure the pre-defined proto-
cols (see above) to meet your own particular requirements.
For incidental changes, you simply re-configure the protocol before or during the
current analysis session. To set up your own custom protocols, you configure an
existing protocol in the same manner (using a representative exam) then save the
protocol under a new name.
To add, modify or remove points used to define the vessel to be analyzed (“track-
ing” points), you can re-configure the protocol in Step 1 “Define Section to Analyze”.
See paragraph 2-2.
To add, modify or remove measurement points or related measurements between
points, you can re-configure the protocol in Step 3 “Select Section of Interest”. See
paragraph 2-3.
If your changes are only to be used for the current exam, you can now use the
modified protocol to perform the analysis, then save or film the report as required.
If you also intend to re-use the modified protocol for future sessions, you can save
the protocol under a new name before closing it (see paragraph 2-4).
Note: To quickly add one or more basic measurements (such as diameter, area,
distance, angle or volume), to be included in the report only for the current
session, you do not need to re-configure the protocol: you can use Step 4
“Add Quick Measurements” of the protocol. See Chapter 4 paragraph 4-5.

2-2 Configuring Vessel Definition (“Tracking”) Points

To configure the vessel definition points:
• Display Step 1 “Define Section to Analyze” (if necessary use the (Back) button).
• Select (Configure Protocol) to open the Configure Tracking Points panel.

List of tracking points

Click on label to select (when modifying Point type
or removing a point)

Scroll (if list is larger than window) Select (Add) , (Modify) or (Remove)
Save protocol
Close protocol Return to previous panel

The list of existing vessel definition points (“tracking” points) is displayed.

7-3
Advanced Vessel Analysis

You can distinguish four types of points:

Root
− Root: this is the start of the section to be analyzed. This
Step 1 point cannot be modified or removed.
− Step: this type designates optional regularly spaced points
Step2
along the vessel.
− Intermediate: you use this type to better define the vessel,
Intermediate and in particular to remove ambiguities where the vessel
identification algorithm might select a wrong branch.
Branch − Branch: this type allows you to define branches. The ves-
Branch (end sel identification algorithm will track the vessel up to the
of section) defined branch point, then return to the “main” vessel.
You will note that the end of the section is also of type
“Branch”.
Note: When you add or modify points you only have the
options “Intermediate” and “Branch”.

To add a new point:

• Select the point in the list under which you want to add a point, by clicking on its
label.
• Select (Add) to open the Add a Tracking Point panel.

Enter name for point Enter abbreviation for point

Select tracking type from menu

Enter prompt text for point

Cancel and return Accept entries and

to previous panel return to previous panel

• Select the Tracking type: [Intermediate] or [Branch].

• Enter the necessary information in the three text fields. Click inside each field to
select it, then enter the text from the keyboard. For corrections, you can move
the text cursor with the left and right arrow keys or by clicking with the mouse.
You can insert characters from the keyboard, or delete the character before the
text cursor with the <BackSpace> or <Del> key.
• Point name: enter the full descriptive name of the point. Try to avoid names
longer than the width of the text entry field (40 characters max.).

7-4
 Advanced Vessel Analysis

• Abbreviation name enter an abbreviation for the name (or the full name if it is
short enough). The abbreviation will be displayed on the views. Limit the length
to about ten to twelve characters at most.
• Prompt text for point: enter the instructions (“prompt”) that will be shown in the
“Define Section to Analyze” panel, indicating how and where to place the point.
• This text does not need to fit inside the text field: when you reach the right edge it
will scroll automatically. You can type the text as one long sentence: when dis-
played afterwards it will wrap round automatically. Do not use the <Enter> key.
Note: If you are adding a new point for the current session only, the prompt text
can be a brief reminder to yourself, such as “Above bifurc.”. However, if you
intend to save the protocol for future use, it is advisable to write out the
prompt text fully, such as “Click with the left mouse button to select a point
just above the bifurcation”. This avoids any ambiguity during later use of the
protocol, either by yourself or by another user.
• Select (Accept) to validate the new point and return to the Configure Tracking
Points panel, or (Cancel) to annul the operation.
Note: If you entered a name or abbreviation that already exists, a message alerts
you to this and the panel remains displayed, so that you can make the cor-
rection.
To modify a point:
• Select the point in the list by clicking on its label.
• Select (Modify) to open the Modify a Tracking Point panel.
• You can now modify the Tracking type and the information in the three text
fields (point name, abbreviation and prompt text) in the same manner as
described above for a new point.
• Select (Accept) to validate the modification and return to the Configure Track-
ing Points panel, or (Cancel) to annul the operation.

To remove a point:
• Select the point in the list by clicking on its label.
• Select (Remove).The point is deleted from the list.
Note: Take care when removing points. Once deleted they cannot be restored.
You will note that you cannot remove the starting point (type “Root”).
At this stage you can save the protocol (see paragraph 2-4) or return to the Step 1
“Define Section to Analyze” panel by clicking on the (Back) button.

7-5
Advanced Vessel Analysis

2-3 Configuring Measurement Points and Related Measurements

2-3-1 Overview
When configuring the protocol, a distinction is made between the measurement
points and related measurements.
The measurement points are located on the centerline of the vessel and associated
with measurements such as diameters and cross section area at that point.
The related measurements refer to measurements between points, such as angles,
or the length or volume of a section between two points.
You add, remove or modify measurement points and related measurements sepa-
rately (see paragraphs 2-3-2 and 2-3-3, respectively).

2-3-2 Configuring Measurement Points

To configure the measurement points:
• Display Step 3 “Select Section of Interest” (if necessary use the (Back|Next) but-
tons).
• Select (Configure Protocol) to open the Configure Protocol panel.

Configure points

Save protocol
Close protocol Return to previous panel

• Select (Configure Points) to open the Configure Points panel.

List of measurement points

Click on label to select (when modifying Measurement point type
or removing a measurement point)

Scroll (if list larger than window) Select (Add) , (Modify) or (Remove)

Close protocol Return to previous panel

7-6
 Advanced Vessel Analysis

The list of existing measurement points is displayed. You can distinguish three
types of points:
− No Diameter: no diameter or cross section area measurements will be associ-
ated with the point. You may want to define such a point for a related measure-
ment (such as an angle, length or volume, see further), where the diameter and
area measurement at the point itself are not required in the report.
− Auto Diameter: all diameter and area measurements for the point will be com-
puted automatically by the software and included in the report.
− Manual Diameter: only a manually defined diameter (distance) measurement
will be associated with the point. To perform such a measurement during analy-
sis, you move the 3D cursor to the appropriate cross section, then place two
points on the oblique X-section view. You may want to include these instructions
in the “prompt” text (see below). An example would be the presence of a throm-
bus or calcification, where an automatic measurement will show the diameter of
the lumen, and the user may want to add a measurement of the actual vessel
diameter.

To add a new point:

• Select (Add) to open the Add a New Point panel.

Enter name for point Enter abbreviation for point

Select measurement type
("diameter") from menu

Enter prompt text for point

Cancel and return Accept entries and

to previous panel return to previous panel

• Select the Diameter Type: [No Diameter], [Auto Diameter] or [Manual Diame-
ter].
• Enter the necessary information in the three text fields. Click inside each field to
select it, then enter the text from the keyboard. For corrections, you can move
the text cursor with the left and right arrow keys or by clicking with the mouse.
You can insert characters from the keyboard, or delete the character before the
text cursor with the <BackSpace> or <Del> key.
• Point name: enter the full descriptive name of the point. Try to avoid names
longer than the width of the text entry field (40 characters max.).
• Abbreviation name: enter an abbreviation for the name (or the full name if it is
short enough). The abbreviation will be used in the report lists, and displayed on
the views. Limit the length to about ten to twelve characters at most.

7-7
Advanced Vessel Analysis

• Prompt text for point: enter the instructions (“prompt”) that will be shown in the
“Select Section of Interest” panel, indicating how and where to place the point.
This text does not need to fit inside the text field: when you reach the right edge it
will scroll automatically. You can type the text as one long sentence: when dis-
played afterwards it will wrap round automatically. Do not use the <Enter> key.
Note: If you are adding a new point for the current session only, the prompt text
can be a brief reminder to yourself, such as “Mark bifurc.”. However, if you
intend to save the protocol for future use, it is advisable to write out the
prompt text fully, such as “Click with the left mouse button to select a point
just above the bifurcation”. This avoids any ambiguity during later use of the
protocol, either by yourself or by another user.
• Select (Accept) to validate the new point and return to the Configure Points
panel, or (Cancel) to annul the operation.
Note: If you entered a name or abbreviation that already exists, a message alerts
you to this and the panel remains displayed, so that you can make the cor-
rection.

To modify a point:
• Select the point in the list by clicking on its label.
• Select (Modify) to open the Modify a Point panel.
• You can now modify the Diameter type and the information in the three text
fields (point name, abbreviation and prompt text) in the same manner as
described above for a new point.
• Select (Accept) to validate the modification and return to the Configure Points
panel, or (Cancel) to annul the operation.

To remove a point:
• Select the point in the list by clicking on its label.
• Select (Remove). The point is deleted from the list.
Note: Take care when removing points. Once deleted they cannot be restored.

After adding, modifying or removing points in the Configure Points panel, click on
(Back) to return to the Configure Protocol panel.
At this stage you can configure the related measurements (see paragraph 2-3-3),
save the protocol (see paragraph 2-4) or return to the Step 3 “Select Section of
Interest” panel by clicking again on the (Back) button.

7-8
 Advanced Vessel Analysis

2-3-3 Configuring Related Measurements

To configure the related measurements:
• Display Step 3 “Select Section of Interest” (if necessary use the (Back|Next) but-
tons).
• Select (Configure Protocol) to open the Configure Protocol panel.

Configure related measurements

(between points)
Save protocol
Close protocol Return to previous panel

• Select (Configure Measurements) to open the Configure Related Measure-

ments panel.

List of measurements
Click on label to select (when modifying Measurement type
or removing a measurement)

Scroll (if list is larger than window) Select (Add) , (Modify) or (Remove)

Close protocol Return to previous panel

The list of existing measurements is displayed. You can distinguish three types of
measurements:
− Length: distance measured along the vessel centerline between the two points.
− Volume: volume of the section of the vessel between the two points.
− Angle: angle between a straight line connecting the two points and the vertical
(RAS coordinate system).

7-9
Advanced Vessel Analysis

To add a new measurement:

• Select (Add) to open the Add a New Measurement panel.

Enter name for measurement Enter abbreviation for measurement

Select measurement type from menu Images in report on/of f

Select start point in "From point" list Select end point in "Topoint" list

Scroll list if necessary Scroll list if necessary

Cancel and return Accept entries and
to previous panel return to previous panel

• Enter the necessary information in the two text fields. Click inside each field to
select it, then enter the text from the keyboard. For corrections, you can move
the text cursor with the left and right arrow keys or by clicking with the mouse.
You can insert characters from the keyboard, or delete the character before the
text cursor with the <BackSpace> or <Del> key.
• Measurement name: enter a descriptive name for the measurement. Try to
avoid names longer than the width of the text entry field (40 characters max.).
• Abbreviation name: enter an abbreviation for the name (or the full name if it is
short enough). The abbreviation will be used in the report lists, and displayed on
the views. Limit the length to about ten to twelve characters at most.
• Select the measurement type: [Length], [Volume] or [Angle].
• To include only the measurement values in the report: set (Film) to off. To include
both the measurement values and the corresponding images in the report: set
(Film) to on.
• In the side-by-side lists, select (highlight) the start point in the From Point list,
and the end point in the To Point list. If necessary you can use the up/down but-
tons under each list to scroll through the list. You can only set up measurements
between existing points. To set up measurements between points not yet in the
list, first return to the Configure Points panel and create the new points as
described in paragraph 2-3-2.
• Select (Accept) to validate the new measurement and return to the Configure
Related Measurements panel, or (Cancel) to annul the operation.

7-10
 Advanced Vessel Analysis

To modify a measurement:
• Select the measurement in the list by clicking on its label.
• Select (Modify) to open the Modify a Measurement panel.
• You can now modify the information in the two text fields (measurement name
and abbreviation) and the measurement type, and select whether to include the
images in the report ( (Film) button), in the same manner as described above for
a new measurement.
• Select (Accept) to validate the modification and return to the Configure Related
Measurements panel, or (Cancel) to annul the operation.

To remove a measurement:
• Select the measurement in the list by clicking on its label.
• Select (Remove). The measurement is deleted from the list.
Note: Take care when removing measurements. Once deleted they cannot be
restored.
After adding, modifying or removing points in the Configure Points panel, click on
(Back) to return to the Configure Protocol panel.
At this stage you can configure the measurement points (see paragraph 2-3-2),
save the protocol (see paragraph 2-4) or return to the Step 3 “Select Section of
Interest” panel by clicking again on the (Back) button.

7-11
Advanced Vessel Analysis

2-4 Saving the Protocol

If you intend to use the modifications of the protocol only for the current analysis
session, you can return to Step 1 or Step 3 after you have made the modifications,
perform the analysis and save or film the report, then close the protocol. The mod-
ifications of the protocol will not be saved.
If you want to save the modified protocol for future use, you can do this at any stage
before you actually close the protocol. To do so:
• Select (Save Protocol) in the Configure Tracking Points or Configure Proto-
col panel. If necessary, first move to either Step 1 or Step 3 and select (Config-
ure Protocol).
• Click inside the Name: text entry field, and enter the new name from the key-
board.
• Select (Save) to save the protocol, or (Cancel) to annul the operation. If the
name already exists, a message will alert you to this. In the message window,
you can select (Cancel) to repeat the operation with a different name, or (Over-
write) to replace the existing protocol with the modified one.
The protocol will be saved in the same Volume Analysis protocol category as the
original protocol.
After saving the protocol, you can return to the Step 1 or Step 3 panel using the
(Back) buttons.

7-12
 Advanced Vessel Analysis

2-5 Deleting a Protocol

If you regularly create customized protocols, you may at some point want to delete
one or more protocols, in particular if you have created successive versions of the
same protocol and you want to delete the earlier versions.
To delete a protocol:
• From the Browser: first start Volume Analysis. If Volume Analysis is already run-
ning, close the current protocol and select (New Protocol).
• In the Select a protocol category panel, select the required category.
• In the Select a protocol panel, click on (Delete Protocol) under the list. The
panel changes to Select a protocol to be removed.
• Click on the icon of the protocol you want to delete. A confirmation message “Do
you really want to delete this protocol?” is displayed. Click on (Yes) to confirm,
or on (No) to cancel the operation. If required you can remove more than one
protocol by selecting them in succession.
• Click on (Cancel) to close the panel and return to the Select a protocol cate-
gory panel.

7-13
 Advanced Vessel Analysis

GLOSSARY
Note: Words in italics refer to terms defined elsewhere in the Glossary.

3D Model - The representation of the 3D (three-dimensional) image data in the

workstation computer memory. 3D models can be saved, archived and networked
(3D OBJ format).
Usually, a 3D model is created by the software at startup from a CT or MR
image set. However, saved 3D models, or 3D models generated by Advan-
tage 3DXR can also be loaded and processed by the Advanced Vessel
Analysis application.
3D OBJ - Computer file format (“Type” in Browser series list), used to store a 3D
Model on the workstation. Also used by Advantage 3DXR to store 3D data directly.
3DXR (3D X-ray) - Process of deriving anatomical information by computer synthe-
sis of X-ray data acquired at multiple angles by means of a conventional X-ray sys-
tem.
Active Annotation - An system annotation on a view that can be modified by the
user to control certain viewing parameters (e.g., window width and level). Active
annotation are displayed in red.
Advantage 3DXR - Software application used to derive a 3D data set from X-ray
data acquired at multiple angles by means of a conventional X-ray system. A data
set generated by Advantage 3DXR uses the 3D OBJ format (3D object).
Algorithm - A step-by-step process used to solve a problem. In Advanced Vessel
Analysis this refers in particular to the process used to identify (track) and quantify
the vessel section to be analyzed.
Annotation - Generally, workstation-supplied text which accompanies an image
when it is displayed on-screen, describing when and how that image was acquired,
with what parameters. Also, text and measurement information added on a view by
the user.
Artifact - Feature in an image resulting either from the initial data acquisition or
subsequent computer processing that does not correspond to a real feature in the
original anatomical structure. Also see Partial Volume Effect.
Baseline view - A basic axial, coronal or sagittal view, aligned parallel to the main
axes of the RAS coordinate system.
Browser - The window used to select available images for display and manipula-
tion.
CT (Computed Tomography) - Process of deriving anatomical information by
computer synthesis of X-ray data, acquired by means of a CT scanner in the form
of parallel “slices”.

1
Advanced Vessel Analysis

CTA (CT Angiography) - The use of CT techniques optimized for angiography.

DFOV (Display Field Of View) - The real dimensions of a view (width and height)
with reference to the RAS coordinates.
DICOM - Abbreviation for Digital Imaging and Communications in Medicine. Stan-
dard for the formatting and exchange of medical images and associated informa-
tion.
Display matrix - The number of pixels in a displayed image, expressed in terms of
number per axis - e.g., 512 x 512.
Exam - In MR, a single study, including all its component series and scans. In CT,
all images made from data taken of a patient after entering a particular scan cycle.
In 3DXR, a 3D model generated by Advantage 3DXR.
Field of View (Acquisition FOV) - The area of the anatomy being imaged, usually
expressed in centimeters. FOV image size is a function of the acquisition matrix
times the pixel size.
HU (Hounsfield Unit) - Scale unit denoting voxel density in a CT data set.
Image - In this document the term “image” is used to designate the part of the exam
data being processed and displayed on the workstation screen. Depending on the
display settings, a view (q.v.) can display an entire image, or part of it (zoom).
Image Display Area - During use of a viewing application, the portion of the screen
where images are displayed.
Measurement Annotation - A user annotation on a view that shows the result of a
measurement.
MPVR (Multi-Projection Volume Reconstruction) - A technique that allows you
to define and display a “thick” slice that encompasses a feature of interest, instead
of using baseline and oblique views that represent slices that are only one voxel
thick.
MR, MRI (Magnetic Resonance Imaging) - Creation of images using the magnetic
resonance phenomenon. Most of the current applications involve imaging the distri-
bution of hydrogen nuclei (protons) in the body.
MRA (MR Angiography) - The use of MR techniques optimized for angiography.
On-view menu - Menu displayed either on a view or on a particular feature of a
view such as a trace, a reference image, etc. by pressing the right mouse button.
Partial Volume Effect - The appearance of voxels with intermediate values at the
interface (separating surface) between two tissue types with clearly distinct and dif-
ferent densities, where these voxels do not correspond to a real feature in the origi-
nal anatomical structure.
Pixel - Abbreviation for “picture element,” the smallest unit a computer screen can
display.

2
 Advanced Vessel Analysis

RAS - Abbreviation for Right/Anterior/Superior. Designation for the patient-linked

coordinate system used in CT and MR data sets.
Reference Image - Small image (normally displayed in the bottom right corner of a
view) that graphically indicates the orientation of the image displayed in the view
with respect to a baseline image (axial, coronal or sagittal).
Rendering - Techniques used to represent a three-dimensional object on a two-
dimensional surface.
System Annotation - An annotation on a view added by the system software, con-
taining data concerning the displayed image. Certain system annotations can be
active, i.e., they can be modified by the user.
Text Annotation - A user annotation on a view containing text. Text annotations
can be used to add comments, or to add a legend to an anatomical feature on a
view.
Title Bar - A bar at the top of a window that provides information about that window,
and also allows you to move the window to a different position on the screen by
clicking and dragging on it with the mouse.
Toggle - The act of switching a function from on to off, or off to on, with a single
mouse click of the pointer on the function’s button. Toggle buttons usually appear
within windows and other monitor screen areas, but some keys on the keyboard
may also provide toggle functions.
User Annotation - An annotation on a view added by the user, containing either
text or the result of a measurement.
View - Part of the workstation screen, used to display image data. The view area of
the Advanced Vessel Analysis screen normally contains four views. A view can
display an entire image, or part of it (zoom).
View Area - During use of a viewing application, the portion of the screen(s) where
images are displayed. The view area normally contains four views, but a single
view can be enlarged so as to take up the entire view area.
VOI (Volume Of Interest) - In Advanced Vessel Analysis, the volume displayed in
the 3D view that contains the selected section of the vessel to be analyzed.
Voxel - Abbreviation for “volume element,” the basic element in a CT or MR data
set.
Window (1) - Describes the range of pixel values that are assigned a shade of
grey. Narrow windows offer greater resolution and contrast of anatomy having simi-
lar densities. It also helps you find the values for anatomy in which you are inter-
ested.
Window (2) - ”Window” is also the term used for an on-screen graphical tool used
to display information.

3
Advanced Vessel Analysis

Window Width and Level (W/L) - In this context, “window” refers to the range of
pixel values within the image data, that is assigned a shade of grey for display.
“Level” refers to the center value, “width” refers to the range of pixel values dis-
played around this central level.
The adjustment is marginally similar to adjusting brightness and contrast
controls: a narrow window (low width) translates to a high contrast of the dis-
play, and similarly a low level translates to a high value of brightness.