Multiple eye disease detection using deep learning

Abstract

Human eyes are vulnerable to several abnormalities because of trauma, aging and disease like diabetes. The main factors of blindness around the world are glaucoma, cataract, macular degeneration and diabetic retinopathy etc. These eye diseases need to be detected and diagnosed timely with appropriate treatment for the solution of this problem. Multiple eye disease detection by analyzing various medical images can provide a timely diagnosis of eye diseases. The steps that are involved in multiple eye disease detection using deep learning are the acquisition of images, region of interest extraction, extraction of features and classification or detection of a particular disease. In this paper, diseases like uveitis, glaucoma, crossed eyes, bulging eyes and cataracts have been detected using deep learning models like Resnet and vgg16 model. We have obtained 92% accuracy using Resnet50 and 79% accuracy using the vgg16 model.