Deep Neural Network-Based Early Detection of Diabetic Retinopathy from Retinal Fundus Images

Diabetic retinopathy is a leading cause of preventable vision loss worldwide, and early detection is critical for reducing the risk of blindness. However, manual screening is time-consuming, expensive, and often unavailable in low-resource settings. This paper proposes a deep neural network-based framework for the automated detection of diabetic retinopathy from retinal fundus images. The method leverages convolutional neural networks to learn discriminative visual features from images and classify disease severity levels. By combining image preprocessing, feature extraction, and deep classification, the proposed approach aims to improve diagnostic accuracy and support large-scale screening programs. Experimental evaluation on publicly available retinal image datasets demonstrates that deep learning can achieve strong performance in terms of accuracy, sensitivity, and specificity, making it a promising tool for clinical decision support. The results suggest that neural network-based systems can assist ophthalmologists in identifying diabetic retinopathy at an early stage, thereby improving patient outcomes through timely intervention.

کلیدواژه ها:

Artificial Intelligence ، Deep Learning ، Neural Networks ، Diabetic Retinopathy ، Retinal Fundus Images ، Medical Image Analysis ، Convolutional Neural Networks ، Early Diagnosis