Automated Identification of Diabetic Retinopathy Using Alexnet Architecture

Diabetic Retinopathy (DR) is one of the reasons for causing blindness globally. DR is a retinal disease that is found in diabetic patients. As performing retinal screening examinations is a time taken and unmet need on all diabetic patients, automated medical image analysis helps in identifying the disease severity. In this paper we study about the image classification using deep learning. The purpose of this study was to develop automated diagnostic technology for DR screening. In this paper we used AlexNet design with convolutional neural networks for DR classification. Here our algorithm processes all the fundus images and classified them as affected or not. A total of 800 images were taken from MESSIDOR dataset to train and test a model to separate non-affected from those which are affected. By using this algorithm our model get good results in Accuracy.