BLOOD VESSEL MOLECULAR SEGMENTION AND ANALYSIS IN DIABETIC RETINAL IMAGES IN IMAGE PROCESSING USING MATLAB

Abstract

Diabetes retinopathy is a long-term condition that damages the retina and other portions of the diabetic patient's body, including the eyes. The people with Diabetic retinopathy who have been suffering for a long time would go blind even if it reached its extreme. Diabetic retinopathy's devastating effects can be minimised if individuals are diagnosed as soon as possible. Diabetic retinopathy must be detected early if a person is to have a chance of regaining their vision and receiving proper treatment. Diabetic retinopathy can be detected from retinal fundus images using image processing and deep learning, as shown in this work. An extraction phase and a classification phase are both included in the proposed study. By segmenting blood vessels and recognising micro aneurysms in digital fundus images, we were able to extract the most relevant information. After a Convolution Neural Network was used to analyse the images, the classification was carried out. Diabetic retinopathy can be diagnosed from retinal fundus pictures using the proposed method, according to the results.