Automated Diabetic Retinopathy Detection Based on Convolutional Neural Network

Abstract

Diabetic Retinopathy is a serious eye disease that originates from diabetes mellitus and it is the most common cause of blindness in the developed countries. Exudates, the most prevalent clinical signs of diabetic retinopathy and the detection of exudates in the diagnosis of diabetic retinopathy has vital clinical significance in monitoring the progress of the desease. The convolutional neural network is adopted to diagnose the diabetic retinopathy from the fundus images. Convolutional Neural Network architecture is developed with data augmentation to identify the pixel-wise exudate identification for a better result and accuracy.