Detection of Diabetic Retinopathy using Machine Learning Algorithm

Diabetic Retinopathy (DR) is an ocular abnormality caused due to diabetes. In this disease there is a progressive damage to the retina if the high blood glucose levels are not controlled. DR has mainly two stages Non Proliferative Diabetic Retinopathy (NPDR) and Proliferative Diabetic Retinopathy (PDR). In this paper, we propose a robust automated system which detects and classifies the different stages of DR. The optic disc and retinal nerves are being segmented, and using Gray Level Co-occurrence Matrix (GLCM) method the features are being extracted. The classification is done using Fuzzy classifier and Convolutional Neural Network, for detecting different stages of DR. The databases used are STARE, DIARETDB0 and DIARETDB1.