Feature Selection for Disease Monitoring by Means of Support Vector Machine

Abstract

Eye disease is a major cause of preventableblindness in the world. It is a complication of diabetes whichcan also affect various parts of the body. When the smallblood vessels have a high level of glucose in the retina, thevision will be blurred and can cause blindness eventually.This is known as diabetic retinopathy. Regular screening isessential in order detect the early stages of diabeticretinopathy for timely treatment to prevent or delay further deterioration. This project describe a method for automatically detecting new vessels on the optic disc using retinalphotography. Abnormal vessels are segmented using watershed segmentation. Features are extracted from the image. The main aim of this project is to develop a computer-based approach to detect the different diabetic retinopathy stages using fundus images.