Medical image analysis for eye disease

Visual impairment can be caused by illness, deterioration of eyesight or eye related diseases. Eye related diseases such as Diabetic Retinopathy/Maculopathy, Glaucoma, Age-related Macular Degeneration (AMD) are the more common causes of visual impairment in Singapore. AMD is the third leading cause...

Full description

Saved in:
Bibliographic Details
Main Author: Liang, Zi Yang.
Other Authors: Chan Kap Luk
Format: Final Year Project
Language:English
Published: 2010
Subjects:
Online Access:http://hdl.handle.net/10356/40909
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
Description
Summary:Visual impairment can be caused by illness, deterioration of eyesight or eye related diseases. Eye related diseases such as Diabetic Retinopathy/Maculopathy, Glaucoma, Age-related Macular Degeneration (AMD) are the more common causes of visual impairment in Singapore. AMD is the third leading cause of visual impairment in Singapore and around the world. AMD can be detected by the build-up of drusen (yellowish deposits) in the eye fundus image by a certified optician or grader. The aim of this project is to detect the drusen spots from the eye fundus image. An automated detection algorithm will be developed to process the fundus images and detect the region where the drusen are located. In this report, the background and theories behind the process of the automated detection algorithm will be touched on. Various methods will be explored to determine the best method in detecting the drusen spots. For the experiments, a set of images graded by certified graders from the Singapore Eye Research Institute (SERI) will be processed by the algorithm to determine the effectiveness of the algorithm. Those drusen locations that are detected by the algorithm will be recertified by the graders from SERI. Finally, the report will bring up some challenges encountered during the course of this project and will suggest some possibilities for others who are attempting to conduct similar experiments to consider.