Automatic Detection and Analysing Severity of Diabetic Retinopathy Disease using Retinal Images

Abstract

Diabetic Retinopathy is an eye threatening disease which not treated properly may lead to loss of vision. This may occur to patients who are suffering from diabetes for a long duration. The less controlled of blood sugar level may eventually lead to Diabetic Retinopathy. Extraction of blood vessels, hard exudates and microaneurysms are the three ways of identifying this disease. The area of these extracted features is calculated for examining whether the person is affected by the disease. Also the severity of the disease can also be found by the number of white pixels appearing in the extracted image wherein the area of the hard exudates and microaneurysms exceeds a certain threshold, provides a corresponding treatment for the disease. Fundus camera captures the retinal images. With the help of acquired retinal images automated screening of these images can be accomplished. Also it is possible to rectify the associated complications in the retina of the eye. The advantage is that the detection and analysing of this disease can prevent the permanent visual loss and provide a proper treatment to the patients who are suffering from this disease. Keywords:diabetes, blood vessels, hard exudates, microaneurysms, fundus camera