Scribd
Upload
89 views

MiniProject PPT

PPT on project -Pneumonia Detector .Helps in early detection of pneumonia by uploading the patients Chest Xray.

Uploaded by

AISHWARYA HOYSAL
Copyright
© © All Rights Reserved
89 views

MiniProject PPT

PPT on project -Pneumonia Detector .Helps in early detection of pneumonia by uploading the patients Chest Xray.

Uploaded by

AISHWARYA HOYSAL
Copyright
© © All Rights Reserved
You are on page 1/ 15

KARAVALI INSTITUTE OF TECHNOLOGY

NEERUMARGA,MANGALORE
Department of Computer Science and Engineering
A
Presentation
on

PNEUMONIA DETECTOR

Under the Guidan Submitted by


ce of
AISHWARYA HOYSAL (4KM21CS001),
ASHOK SURESH (4KM21CS002),
Ms. ASWANI M BHAVYA KESHAV (4KM21CS003),
Assistant Professor, BHOOMIKA H R (4KM21CS004)
Dept of CSE, KIT,
Mangalore
V I S V E S V A R AYA T E C H N O L O G I C A L U N I V E R S I T Y
JNANA SANGAMA BELAGAVI – 590018
PNEUMONIA
DETECTOR
Early identification of lung inflammation via
chest X-rays
CONTENTS

⚬Abstract
⚬Introduction
⚬Objectives
⚬Problem statement
⚬Requirement
⚬Methodology
⚬Application
⚬Conclusion
ABSTRACT
⚬This project works on Convolutional Neural Networks (CNNs), a type of deep learning model
particularly well-suited for image analysis tasks.

⚬CNNs are designed to automatically and adaptively learn spatial hierarchies of features, making them
effective in identifying patterns in medical images.

⚬In this project, the CNN model is trained on a large dataset of chest X-ray images, labeled as either
NORMAL or indicative of PNEUMONIA.

⚬The trained model can then analyze new X-ray images, detecting subtle patterns and
anomalies that might indicate the presence of pneumonia, thus providing a probability score of the
disease.

⚬ This automated approach aims to assist medical professionals in making quicker and more accurate diagnoses.

0
1
INTRODUCTION
Convolutional Neural Networks (CNNs) :
⚬A deep learning model designed to process and analyze visual data.
⚬Used for feature extraction and classification in image-based tasks.

How It Works:
⚬The CNN processes input X-ray images through multiple layers.
⚬Convolutional layers extract important features (e.g., edges, shapes).
⚬Pooling layers reduce the size of feature maps, highlighting essential information.
⚬Fully connected layers interpret the extracted features to predict the probability of pneumonia.
⚬The final output indicates the likelihood of pneumonia, assisting in diagnosis.

0
2
OBJECTIVES
Automate Pneumonia Detection: A system that can analyze chest X-ray
images to detect signs of pneumonia.

Enhance Diagnostic Accuracy: Use of deep learning techniques to improve


the accuracy of pneumonia diagnosis compared to traditional methods.

Provide Quick Results: Fast and efficient way to assist healthcare professionals
in diagnosing pneumonia.

User-Friendly Interface: An accessible platform for users to upload images and


receive diagnostic results.

0
3
PROBLEM STATEMENTS

Early and
accurate
Shortage of diagnosis of
medical pneumonia is
experts,leading to crucial but
delayed diagnosis challenging, often
and treatment. requiring expert
radiologists to
interpret chest X-
0
ray images. 4
REQUIREMENTS
HARDWARE REQUIREMENTS SOFTWARE REQUIREMENTS PRE REQUISITE KNOWLEDGE

• Processor: Multi-core CPU


COMPUTING RESOURCES
• HPC/Cloud: Powerful CPUs (Intel
• Memory (RAM): At least 8 GB, • HTML 5
i7/i9, AMD Ryzen 7/9) and GPUs ideally 16 GB or more for faster • CSS (cascading style sheets)
(NVIDIA RTX 2080, A100). model training and inference • JavaScript
• Storage: SSDs, minimum 1TB.
• Flask
• Graphics Card (GPU): NVIDIA
IMAGING HARDWARE GPU with CUDA support for • Python
• Access to high-quality chest X-ray accelerated TensorFlow operations • Machine Learning Algorithms
machines (optional but recommended) – CNN
USER END EQUIPMENTS • Storage: SSD with sufficient
• Computers: Quad-core CPU space for datasets, software, and the
(Intel i5/i7), 8GB RAM, 500GB following packages: Flask,
storage. TensorFlow, Pillow, Requests,
• OS: Windows 10/11, macOS, or NumPy and Wheel
recent Linux.
• Display: Minimum 1920x1080 • Operating System: Compatible
with major OS like Windows,
resolution.
macOS, or Linux 0
METHODOLOGY
1 2 3

Data Collection Data Preprocessing Data Visualisation


Gather data from various Perform data cleaning, handling Visualize the data to
sources, images gain
including medical data in different format/Rotated insights.
images etc

4 5

Model Development Model Evaluatiom


evaluate their
Train ML models using performance using
the suitable evaluation metrics
preprocesse data while
testing the model
0
6
Input -> Conv + ReLu -> Max Pooling -> Conv + ReLu -> Max
Pooling -> ... -> Flatten -> Fully Connected + Softmax -> Output
APPLICATIONS
• Early Diagnosis: Enables prompt detection of pneumonia from chest X-rays, facilitating early intervention and
treatment.

• Remote Healthcae: Assists healthcare professionals in remote or underserved areas where access to specialists
may be limited.

• Cost-Effective Screening: Provides an affordable solution for mass screening in high-risk populations or
during outbreaks.

• Healthcare Efficiency: Enhances the efficiency of medical workflows by reducing the time needed for manual
X-ray analysis and prioritizing cases.

• Integration with Telemedicine: Can be integrated into telemedicine platforms to support remote
consultations and diagnostic services.

• Data Collection and Research: Facilitates the collection of large datasets for research on pneumonia and
related respiratory conditions, contributing to advancements in medical knowledge.
0
7
FUTURE SCOPE
• Integration with Electronic Health Records (EHR): Allow automatic uploading of results to
patient health records for better tracking and management.

• Mobile Application Development: Create a mobile app version for on-the-go access and
convenience for healthcare providers and patients.

• Support for Additional Diseases: Extend the system to detect other chest-related conditions such as
tuberculosis or lung cancer.

• User Interface and Experience Enhancements: Improve the user interface to make it more
intuitive and accessible for users with varying levels of tech proficiency.

• Multilingual Support: Add support for multiple languages to make the tool accessible to a global
audience.

0
8
THANK YOU

You might also like