Detection of Diabetic Retinopathy Using a Convolutional Neural Network (CNN) Algorithm

Abstract

Diabetic retinopathy is a disease that damages the blood vessels in the retina of the eye. If left untreated, this condition can lead to blindness. This study aims to detect and classify diabetic retinopathy using the Convolutional Neural Network (CNN) algorithm one of the deep learning methods applied in machine learning for image analysis and interpretation. The objective of this research is to enhance the accuracy of predicting and classifying the types of blindness experienced by diabetic patients based on retinal images. The system identifies four retinal conditions: normal retina, glaucoma, cataract, and diseased retina. The CNN model in this study was trained with an input image size of 2464×1632 using 90 training images and 10 testing images, a 3×3 filter, and 800 epochs. The model achieved 90% accuracy in classifying eye images during training and testing. The results demonstrate that CNN is highly effective in detecting diabetic retinopathy and differentiating between various retinal disorders.