Convolutional Neural Networks in Detecting Diabetic Retinopathy

Abstract

Diabetic Retinopathy (DR) is an eye disease, which is caused because of Diabetes and it leads to cause Blindness. Early identification and timely treatment are essential in order to prevent Vision loss. But in countries like India Africa where there are very a smaller number of doctors available when compared with the patients regular testing of eyes and thorough consultation is very difficult. But to overcome the problem of DR a continuous monitoring is necessary, so for this Computer based diagnosis is considered as effective method. In this Computer based Diagnosis many different approaches are there, but here we are considering Deep Learning approach for detecting of DR. In Deep learning again many approaches are there, but we are considering Convolution Neural Networks (CNN) method in early detection of DR. A CNN based approach identifies Exudates, Micro Aneurysms and Hemorrhages in retina image which are essential in identifying Diabetic retinopathy.