diabetic retinopathy detection and classification based on deep learning: a review

Diabetic retinopathy can be defined as an eye disease that occurs specifically in diabetic patients and results in damaging the small blood vessels of the retina because of high and low blood sugar. delayed detection and treatment often lead to blindness, so one of the most significant issues is early detection of this disease, which is necessary for successful treatment. many deep learning methods have been suggested for diabetic retinopathy detection and classification. manual inspection of the fundus images to check diabetic retinopathy is highly tedious and time-consuming work. thus, an automatic method for early diabetic retinopathy diagnosis utilizing the fundus images is very useful tool that helps experts. in this review paper, several deep learning models that are widely used in literature investigated diabetic retinopathy classification will be presented and discussed. in addition, a comparative analysis for classification performance accuracy of diabetic retinopathy using these deep learning models will be reviewed comprehensively. thus, an automatic approach for early diabetic retinopathy diagnosis utilizing the fundus images is a very useful tool that helps the experts.