COMBINING MACHINE LEARNING TECHNIQUES WITH STATISTICAL SHAPE MODELS IN MEDICAL IMAGE SEGMENTATION

For more Computer Science projects click here


ABSTRACT


In this thesis, we implemented Point Distribution Model and basic Active Shape Model algorithm and contributed this to the AUST Computer Vision and Machine Learning code library. We applied the Active Shape Model to segmenting lateral ventricles of 2D brain images and used machine learning – specifically K-Nearest Neighbour algorithm- to improve segmentation results. A statistical shape model is created from a training dataset which is used to search for an object of interest in an image. Active shape model has shown over time to be a reliable image segmentation methodology but its segmentation accuracy is hindered especially by poor initialization which can’t be guaranteed to always be perfect. In our methodology, we extract features for each landmark using Haar filters. We train a classifier with these features and use the classifier to classify points around the final points of an Active shape model search. The aim of this approach is to better place points that might have been wrongly placed from the ASM search. We have used the simple, yet effective K-Nearest Neighbour machine learning algorithm, and have demonstrated the ability of this method to improve segmentation accuracy by segmenting lateral ventricles of the brain.


TABLE OF CONTENT

Abstract
Table of Contents
List of Figures

CHAPTER ONE
1.1       Introduction
1.2       Problem Definition
1.3       Outline of Thesis

CHAPTER TWO
2.1       Image Segmentation
2.1.1    Basic Concepts
2.2       First Generation Segmentation Methods
2.2.1    Thresholding
2.2.2    Region Growing
2.2.3    Edge Tracing
2.2.4    Split and Merge
2.3       Second Generation Segmentation Methods
2.3.1    K-means Clustering
2.3.2    Fuzzy C-means Clustering
2.3.3    Active Contour Model (Snakes)
2.4       Third Generation Segmentation Methods
2.4.1    Active Shape Models
2.4.2    Active Appearance Models
2.4.3    Atlas based segmentation

CHAPTER THREE
3.1       Machine Learning
3.1.1    Supervised Learning
3.1.2    Unsupervised Learning
3.1.3    Reinforcement Learning
3.2       Machine Learning Techniques
3.2.1    K-Nearest Neighbour Algorithm
3.2.2    Na├»ve Bayes Classification
3.3.3    Decision Trees
3.3.4    Bagging, Boosting, Random Forest
3.3.5    Clustering
3.3.6    Neural Networks
3.3.7    Support Vector Machines
3.4       Machine Learning in Image Segmentation
3.5       Statistical Shape Models with Machine Learning

CHAPTER FOUR
4.1       Active Shape Models
4.1.1    Landmarks
4.1.2    Aligning the Training Set
4.1.3    Modeling Shape Variation
4.1.4    Using Shape Models in Image Search

CHAPTER FIVE
5.1       ASM with KNN and Haar Features
5.1.1    Haar Feature Extraction
5.2       KNN-ASM
5.3       KNN-ASM Framework
5.4       Results and Discussions

CHAPTER SIX
6.1       Conclusion
6.2       Future Work
REFERENCES


CHAPTER ONE


1.1         INTRODUCTION


In this thesis, our aim is to segment images, specifically, medical images. We aim to implement the popular Active Shape Model algorithm [16] and demonstrate its usefulness in segmenting 2d medical images. Furthermore, we explore improving the results of the Active Shape Model segmentation using machine learning techniques.

Predominantly, the aim of medical image segmentation is to label each pixel in an image to indicate the anatomical structure it belongs to and delineate such structures of interest for the purposes of visualization, diagnosis or medical research. Segmentation is often a crucial first step in patient diagnosis especially when qualitative and quantitative information about appearance, size, or shape of patient anatomy is desired. Results of medical image segmentation are useful for many purposes including image guided surgery, detection of anatomical changes over time, detection of pathological diseases, volumetric measurement, visualization and research. With the increasing importance of the segmentation process in diagnosis, accuracy of the process is important, as this may impact diagnostic accuracy, treatment planning and subsequently treatment.


The segmentation process is unfortunately as difficult as it is important, and the reasons are easy to comprehend. Computers are not half as good as humans when it comes to ill-defined problems such as object recognition, and when these images contain noise, it makes the process even more difficult for a computer. More often than not, images will be noisy, altering the intensity values of some pixels. This could make anatomical structures difficult to separate from their surroundings, and strong edges may not be present around its borders. Sometimes, the intensity level of a single tissue class varies gradually over the image –a phenomenon known as intensity inhomogeneity or non-uniformity – and this doesn‟t make the segmentation task any easier. Other times, an individual pixel may contain mixture of tissue classes such that intensity of a pixel in the image may not be consistent with one class. The gray levels of different tissues, if too close would increase the difficulty of the process. These problems and the variability in the tissue distribution among individuals in the human population means that some degree of uncertainty must be attached to all segmentation results.

Segmentation can be done manually, semi-automatically or can be a fully automated process. Manual segmentation is a time consuming task and with the volume of medical image data needed to be processed, it is highly unlikely to be the ideal method considering the fact that results of such a process would depend on operator variability and thus would be difficult to reproduce. The level of confidence ascribed to manual processes suffers accordingly. Automatic methods overcome these drawbacks and are preferred especially with the computing resources available today and the amount of data needed to be processed. However, accurate automatic segmentation is by no means an easy feat and remains an active area of research in computer vision.....


For more Computer Science projects click here
___________________________________________________________________________
This is a Postgraduate Thesis and the complete research material plus questionnaire and references can be obtained at an affordable price of N3,000 within Nigeria or its equivalent in other currencies.


INSTRUCTION ON HOW TO GET THE COMPLETE PROJECT MATERIAL

Kindly pay/transfer a total sum of N3,000 into any of our Bank Accounts listed below:
·         Diamond Bank Account:
A/C Name:      Haastrup Francis
A/C No.:         0096144450

·         GTBank Account:
A/C Name:      Haastrup Francis
A/C No.:         0029938679

After payment, send your desired Project Topic, Depositor’s Name, and your Active E-Mail Address to which the material would be sent for downloading (you can request for a downloading link if you don’t have an active email address) to +2348074521866 or +2348066484965. You can as well give us a direct phone call if you wish to. Projects materials are sent in Microsoft format to your mail within 30 Minutes once payment is confirmed. 

--------------------------------------------------------
N/B:    By ordering for our material means you have read and accepted our Terms and Conditions


Terms of Use: This is an academic paper. Students should NOT copy our materials word to word, as we DO NOT encourage Plagiarism. Only use as guide in developing your original research work.

Delivery Assurance
We are trustworthy and can never SCAM you. Our success story is based on the love and fear for God plus constant referrals from our clients who have benefited from our site. We deliver project materials to your Email address within 15-30 Minutes depending on how fast your payment is acknowledged by us.

Quality Assurance
All research projects, Research Term Papers and Essays on this site are well researched, supervised and approved by lecturers who are intellectuals in their various fields of study.

No comments:

Post a Comment

Note: Only a member of this blog may post a comment.