Ch01.

qxd 7/5/04 10:06 AM Page 3

Chapter 1
Introduction to
CT physics
Image generation

What is computed tomography (CT)?

Since the first CT scanner was developed in 1972 by Sir Godfrey Hounsfield,
the modality has become established as an essential radiological technique
applicable in a wide range of clinical situations.
CT uses X-rays to generate cross-sectional, two-dimensional images of the
body. Images are acquired by rapid rotation of the X-ray tube 360° around the
patient. The transmitted radiation is then measured by a ring of sensitive radiation
detectors located on the gantry around the patient (Fig. 1.1). The final image is
generated from these measurements utilizing the basic principle that the internal
structure of the body can be reconstructed from multiple X-ray projections.
Early CT scanners acquired images a single slice at a time (sequential
scanning). However, during the 1980s significant advancements in technology
heralded the development of slip ring technology, which enabled the X-ray tube
to rotate continuously in one direction around the patient. This has contributed
to the development of helical or spiral CT.
In spiral CT the X-ray tube rotates continuously in one direction whilst the
table on which the patient is lying is mechanically moved through the X-ray
beam. The transmitted radiation thus takes on the form of a helix or spiral. Instead
of acquiring data one slice at a time, information can be acquired as a continuous
volume of contiguous slices (Fig. 1.2a, b). This allows larger anatomical regions
of the body to be imaged during a single breath hold, thereby reducing the
possibility of artefacts caused by patient movement. Faster scanning also increases
patient throughput and increases the probability of a diagnostically useful scan
in patients who are unable to fully cooperate with the investigation.
3
Ch01.qxd 7/5/04 10:06 AM Page 4

Image generation continued

Ring of fixed detectors Rotating X-ray tube and

a fan beam of X-rays
Fan angle

Patient

Fig. 1.1 Ring of detectors (‘fourth generation’). Patient in cross-section.

The next generation of CT scanners is now commercially available. These

multislice or multidetector machines utilize the principles of the helical scanner
but incorporate multiple rows of detector rings. They can therefore acquire
multiple slices per tube rotation, thereby increasing the area of the patient that
can be covered in a given time by the X-ray beam (Fig. 1.3a, b).

4
Ch01.qxd 7/5/04 10:06 AM Page 5

Image generation continued

Patient/table
movement

Fig. 1.2 (A) Single-slice system (one ring). (B) Single-slice helical CT. The
X-ray tube rotates continuously and the patient moves through the X-ray beam
at a constant rate.

5
Ch01.qxd 7/5/04 10:06 AM Page 6

Image generation continued

Patient/table
movement

Fig. 1.3 (A) Multidetector system (four rings shown here). (B) Multislice
helical CT.

6
Ch01.qxd 7/5/04 10:06 AM Page 7

Image generation continued

How is a CT image produced?

Every acquired CT slice is subdivided into a matrix of up to 1024 × 1024 volume

elements (voxels). Each voxel has been traversed during the scan by numerous
X-ray photons and the intensity of the transmitted radiation measured by
detectors. From these intensity readings, the density or attenuation value of the
tissue at each point in the slice can be calculated. Specific attenuation values are
assigned to each individual voxel. The viewed image is then reconstructed as a
corresponding matrix of picture elements (pixels).

What is a Hounsfield unit or CT number?

Each pixel is assigned a numerical value (CT number), which is the average of all
the attenuation values contained within the corresponding voxel. This number is
compared to the attenuation value of water and displayed on a scale of arbitrary
units named Hounsfield units (HU) after Sir Godfrey Hounsfield.
This scale assigns water as an attenuation value (HU) of zero. The range of CT
numbers is 2000 HU wide although some modern scanners have a greater range
of HU up to 4000. Each number represents a shade of grey with +1000 (white)
and –1000 (black) at either end of the spectrum (Fig. 1.4).

Air Water

–1000 –500 0 +500 +1000

Lung Fat Soft tissue Bone

Bone +400 +1000

Soft tissue +40 +80
Water 0
Fat –60 –100
Lung –400 –600
Air –1000

Fig. 1.4 The Hounsfield scale of CT numbers.

7
Ch01.qxd 7/5/04 10:06 AM Page 8

Image generation continued

Window level (WL) and window width (WW)

Whilst the range of CT numbers recognized by the computer is 2000, the human
eye cannot accurately distinguish between 2000 different shades of grey.
Therefore to allow the observer to interpret the image, only a limited number of
HU are displayed. A clinically useful grey scale is achieved by setting the WL and
WW on the computer console to a suitable range of Hounsfield units, depending
on the tissue being studied.
The term ‘window level’ represents the central Hounsfield unit of all the
numbers within the window width.
The window width covers the HU of all the tissues of interest and these are
displayed as various shades of grey. Tissues with CT numbers outside this range
are displayed as either black or white. Both the WL and WW can be set
independently on the computer console and their respective settings affect the
final displayed image.
For example, when performing a CT examination of the chest, a WW of 350
and WL of +40 are chosen to image the mediastinum (soft tissue) (Fig. 1.5a),
whilst an optimal WW of 1500 and WL of –600 are used to assess the lung fields
(mostly air) (Fig. 1.5b).

What is pitch?

Pitch is the distance in millimetres that the table moves during one complete
rotation of the X-ray tube, divided by the slice thickness (millimetres). Increasing
the pitch by increasing the table speed reduces dose and scanning time, but at the
cost of decreased image resolution (Fig. 1.6a, b).

Image reconstruction

The acquisition of volumetric data using spiral CT means that the images can be
postprocessed in ways appropriate to the clinical situation.
● Multiplanar reformatting (MPR) – by taking a section through the three-
dimensional array of CT numbers acquired with a series of contiguous slices,
sagittal, coronal and oblique planes can be viewed along with the standard
transaxial plane (Fig. 1.7).

8
Ch01.qxd 7/5/04 10:06 AM Page 9

Fig. 1.5 These two images are of the same section, viewed at different
window settings. (A) A window level of +40 with a window width of 350
reveals structures within the mediastinum but no lung parenchyma can be seen.
(B) The window level is –600 with a window width of 1500 Hounsfield units.
This enables details of the lung parenchyma to be seen, at the expense of the
mediastinum.

9
Ch01.qxd 7/5/04 10:06 AM Page 10

Image generation continued

Fig. 1.6 (A) Pitch is low. The table moves less for each tube revolution. The
image is sharper. (B) Pitch is high. The table moves further for each revolution
so the resulting image is more blurred. The helix is stretched.

10
Ch01.qxd 7/5/04 10:06 AM Page 11

Image generation continued

B C

Fig. 1.7 The three images demonstrate a haemoperitoneum, shattered

right kidney and a lacerated spleen in axial (A), sagittal (B) and coronal
(C) planes.

11
Ch01.qxd 7/5/04 10:07 AM Page 12

Image generation continued

● Three-dimensional imaging – using reconstructed computer data enables the

external and internal structure of organs to be viewed. The data can be
projected as a three-dimensional model to display spatial information or
surface characteristics (volume and surface rendering) (see Fig. 2.10). This is
becoming increasingly useful for patients unable to have invasive endoscopy.
● CT angiography (CTA) – following intravenous contrast enhancement,
images are acquired in the arterial phase and then reconstructed and displayed
in either a 2D or 3D format. This technique is commonly used for imaging the
aorta, renal and cerebral arteries. In addition there is increasing interest in the
use of CTA to image the coronary and peripheral vessels (Fig. 1.8).

A B

Fig 1.8 The image on the right is a two-dimensional ‘angiogram’ derived

from a three-dimensional reconstruction, showing a small left anterior descending
artery with a severe proximal stenosis. The image on the left is a comparative
image from the invasive traditional coronary angiogram of the same patient,
confirming the lesion.

Contrast media

Contrast between the tissues of the body can be improved by the use of various
contrast media. These mostly contain substances with a high molecular weight
and thus increase the attenuation value of the organ they opacify.

12
Ch01.qxd 7/5/04 10:07 AM Page 13

Contrast media

● Oral contrast – the bowel is routinely opacified in almost all CT examinations

of the abdomen and pelvis because the attenuation value of the bowel is
similar to other surrounding structures and therefore pathological processes
may be obscured (see Fig. 2.19 – contrast in stomach in upper image). Substances
used include dilute barium or iodine-based preparations, which are normally
given to the patient to drink 24 hours and 1 hour prior to examination to
opacify the distal and proximal gastrointestinal tract respectively.
● Intravenous contrast – these are usually iodine-based media. They can be
injected to opacify the vascular tree in different phases, depending on the rate
and volume of contrast injection and the timing of image acquisition. Arterial
opacification is maximal at approximately 20 seconds with venous enhancement
rising to a peak after approximately 70 seconds. The level of intravascular
enhancement then declines as the contrast equilibrates with the tissues of the
body before being excreted by the kidneys into the ureters and bladder.
These different phases of enhancement are used to image various organs
depending on the indication for the CT examination. Spiral scanners, because
of their speed, are able to acquire images during each phase of enhancement,
thus increasing the information obtained from the study. For example, when
imaging the pancreas for a suspected tumour, thin sections are initially
acquired during the arterial phase to best demonstrate the tumour and also to
diagnose any arterial involvement of the tumour. A second scan during the
portal venous enhancement phase is then performed to optimally demonstrate
any liver metastases and tumour invasion into venous structures (see Fig. 2.19).
The high density of contrast can be utilized to display only the blood vessel
lumen when reconstructing axial data with a maximum intensity projection
(MIP). This is used to look for vascular disease.
● Air – can also be used as a negative contrast agent, for example in imaging of
the large bowel in CT colonography (see Fig. 2.21).

Dual-modality imaging

Positron emission tomography (PET)/CT

PET allows the detection of glucose uptake in tissues. The radiopharmaceutical
utilized is 18-FDG (18-fluorodeoxy-D-glucose). The technique has particular use

13
Ch01.qxd 7/5/04 10:07 AM Page 14

Dual-modality imaging continued

in the detection of malignant and inflammatory lesions where glucose uptake is

increased.
PET and CT imaging can be performed in the same machine, thus providing
superimposed CT and PET images of pathology in a single examination. This
results in a complementary, overlapping display of anatomical and functional/
metabolic information, which can be used in the accurate and efficient diagnosis
and follow-up of cancer patients (Fig. 1.9).

Fig. 1.9 (A) CT of thorax showing soft tissue masses (lymphadenopathy –

white arrowheads) adjacent to the pulmonary vessels. (B) PET scan at the same
axial level showing increased metabolic activity (white arrowheads). Note
high metabolic activity within the left ventricle (black arrowhead).
Continued

14
Ch01.qxd 7/5/04 10:07 AM Page 15

Dual-modality imaging continued

Fig. 1.9 cont’d. (C) PET/CT fused image correlating increased metabolic
activity within lymph node masses implying pathological enlargement.
Courtesy: Nuclear Medicine, University Hospital, Zurich, Switzerland.

Advantages and disadvantages of CT

Advantages
● CT is readily available in most hospitals.
● It is an increasingly rapid imaging modality with excellent image resolution,
enabling faster and more accurate diagnostic evaluation of patients over a
wide spectrum of clinical indications.
● The data acquired in one scan can subsequently be manipulated to provide
multiplanar and 3D reconstructions.

Disadvantages
● Radiation – although CT scans account for only 4% of X-ray examinations,
they contribute to more than 20% of the radiation dose to the population by
‘medical X-rays’. For typical doses of common radiological examinations, see
Table 1.1.

15
Ch01.qxd 7/5/04 10:07 AM Page 16

Advantages and disadvantages of CT

Table 1.1 Radiation doses (mSv – millisieverts) (from Royal

College of Radiologists 2003 Making the best use of a
department of clinical radiology. Guidelines for doctors,
5th edn. Royal College of Radiologists, London)

Approx. equiv.
Diagnostic Typical effective Equiv. no. period of
procedure dose (mSv) of CXR background radiation

CXR 0.02 1 3 days

CT head 2.0 100 10 months
CT chest 8 400 3.6 years
CT abdomen/pelvis 10 500 4.5 years

UK average background radiation = 2.2 mSv per year; regional averages range from
1.5 to 7.5 mSv per year.

● Artefacts – an artefact is a feature or appearance that is seen on an image,

which does not actually exist. They occur in all imaging modalities and are
often unavoidable. Recognizing the presence of artefacts is important in order
to avoid confusion with pathology. However, with the increasing speed of
image acquisition in a single breath hold by the most modern scanners, many
artefacts are being minimized or eliminated. Types of artefact include:
1. motion – from patient movement during a scan, commonly due to breathing
2. streak (beam hardening) – dark ‘streaks’ behind high-density objects, e.g.
dental amalgam and metallic joint replacements
3. partial voluming – different tissue densities within a single voxel lead to
‘averaging’ of data. For example, a small black object within a larger white
space would look like a shade of grey.
● Relatively poor tissue contrast when compared to MRI can be a problem,
despite the use of oral and IV contrast. This may occur in thin adults and
children due to the lack of intra-abdominal fat separating the various tissue
planes.
● CT has a relatively high cost and limited portability.
● There can be contrast media-related complications, including allergic reactions
and renal toxicity.

16