Detection of Diabetic Retinopathy via Image Processing Using Deep Neural Networks

Abstract

Diabetic Retinopathy (DR) is a major cause of vision impairment worldwide. Early detection and classification of DR can
significantly reduce the risk of vision loss. This paper presents an implementation of a deep learning-based system for detecting DR using convolutional neural networks (CNNs). The proposed method utilizes retinal fundus images for automated classification of different DR stages. The system incorporates preprocessing techniques, data augmentation, and transfer learning with a pre-trained CNN model to enhance accuracy. Experimental results demonstrate the model’s effectiveness in identifying diabetic retinopathy with high sensitivity and specificity.