Automated Detection and Classification of Diabetic Retinopathy Using Neural Networks for Ophthalmic Healthcare

Abstract

Automated Detection and Classification of the Diabetic Retinopathy using Neural Networks focuses on identifying diabetic eye disease through intelligent analysis of retinal fundus images. The system aims to reduce manual effort and improve early diagnosis accuracy using deep learning techniques.This project presents a deep learning-based approach that automatically detects and classifies Diabetic Retinopathy into five stages: Healthy, Mild, Moderate, Severe, and Proliferative DR. A Convolutional Neural Network (CNN) is used to extract critical retinal features such as microaneurysms, hemorrhages, and abnormal blood vessel growth. The model is integrated into a Flask-based web application that allows users to upload retinal images and receive instant predictions along with confidence scores and clinical insights. The proposed system provides a fast, costeffective, and reliable screening solution, supporting ophthalmologist’s inn early diagnosis and helping prevent vision loss caused by diabetic retinopathy KEYWORDS Core Medical, Machine Learning & AI, Image Processing, Classification & Evaluation Healthcare & Application.