Measuring Blood Vessels Pathology

By



ACKNOWLEDGEMENT

I would like to take this chance for thanking my research facilitator, friends & family for support they provided & their belief in me as well as guidance they provided without which I would have never been able to do this research.

DECLARATION

I, (Your name), would like to declare that all contents included in this thesis/dissertation stand for my individual work without any aid, & this thesis/dissertation has not been submitted for any examination at academic as well as professional level previously. It is also representing my very own views & not essentially which are associated with university.

Signature:

Date:

Table of contents

ACKNOWLEDGEMENTii

DECLARATIONiii

Summary1

Chapter 12

Introduction2

1.1Project Overview2

1.2Aim2

1.3Objectives2

1.3.1 Essentials3

1.4Minimum Requirement3

1.5Schedule3

1.5.1Original schedule3

1.5.2Revised schedule3

Chapter 24

Background Reading4

2.1 Definitions4

2.1.1 Brief Definitions of Medical Terms4

2.1.2 Abdominal Aortic Aneurysm4

2.1.3 Thrombus6

2.1.4 Segmentation7

2.2 Accuracy8

2.3 Segmentation methods9

2.4 Manual Segmentation9

2.5 Simple Image Segmentation10

2.5.1 Edge-based Segmentation10

2.5.3 Region-based Segmentation12

2.5.3.1 Region Growing Algorithm12

2.6 Model-Based segmentation13

2.6.1 Deformable models13

2.6.1.1 Active Shape Model (ASM)13

2.8 Useful Programming Languages14

2.8.1 Matlab15

2.8.2 C++15

References16

Summary

The problem regarding the visualization and analysis of internal body tissues have proved to be a big challenge. The challenges regarding these problems in the medical sector of computing, mainly medical images used for visualization and analysis of internal body tissues are addressed in this report. Images presented and discussed in this report are particularly taken for an abnormality (Aneurysm) in the abdominal aorta artery anatomy. Partitioning this artery from other neighboring tissues (such as kidney tissues) is difficult and hard to be accurate. Reasons for this problem being of a major interest will be discussed. As a start, the report declares unfamiliar definitions of medical terms that are used throughout the report. Then, few useful techniques for partitioning (segmentation) are discussed. Some previous segmentation methods from external researches are covered. Finally, Stages for the new developed method and evaluation are described in details.

Chapter 1

Introduction

Project Overview

It is understood that if a normal balloon is blown, you will have a nicely shaped ball. The more air you blow, the bigger the balloon gets. As the balloons become bigger the wall will become thinner. This is an excellent example to understand the structure of the aneurysm. Abdominal Aorta is the largest artery in the body that might contain the largest aneurysm.

Leaving aneurysms without treatment is fatal and should be treated as soon as possible. Abdominal Aorta in the abdomen is not by its own. There are many neighboring organs. Once 'ballooning' arises, it sometimes comes so close to other organs tissues. In this situation, it is hard to decide whether a certain part of the scan belongs to the aneurysm or to other surrounding tissues. This project will introduce and develop a new segmentation method which will reduce this difficulty.

Aim

The aim of this project is to find the thrombus in CT Angiography images for patients with abdominal aortic aneurysms (AAA) by developing a new idea of a segmentation method.

Objectives

Among many segmentation methods, chose the best one that can minimize leaking “segmentation outside the thrombus”.

Raise a new method of segmenting the thrombus using existing ...