A Deep Learning Approach to Diabetic Retinopathy Classification

Diabetic retinopathy is a serious ophthalmic disorder, caused by the damaging of retina due to high blood sugar. Depending on the severity of the problem, diabetic retinopathy can be categorized into one of the four classes (i.e. mild, moderate, severe and proliferate). In this work, we have applied Convolutional Neural Network on the retinal fundus images to classify the images into one of the five classes (no diabetic retinopathy, along with the four different classes). Different convolutional neural network architectures were used to correctly classify retinal fundus images into its category. Empirical results indicate that we can classify diabetic retinopathy from the fundus images reliably.