Diabetic Retinopathy Improved Detection Using Deep Learning

Diabetes is a disease that occurs when the body presents an uncontrolled level of glucose capable damaging retina, leading to permanent damage eyes or vision loss. When diabetes affects eyes, it known as diabetic retinopathy, which became global medical problem among elderly people. The fundus oculi technique involves observing eyeball diagnose check pathology evolution. In this work, we implement convolutional neural network model process image recognize structure and determine presence retinopathy. model’s parameters are optimized using transfer-learning methodology for mapping with corresponding label. training testing performed dataset images severity scale present in labels. separates into five classes, from healthy proliferative retinopathy presence. latter probably blind patient. Our proposal presented accuracy 97.78%, allowing confident prediction images.