Paper Key : IRJ************731
Author: Supriya Raghupati Samant,Vedika Dattatray Gavali,Asawari Bhaskar Kumbhar,Pragati Babaso Gurav,Prof. Sonali Subhash Patil
Date Published: 04 Oct 2024
Abstract
Diabetic retinopathy (DR) is a leading cause of blindness among individuals with diabetes, resulting from damage to the retina's blood vessels. Detection of DR should be done early to prevent severe vision loss. To automate the screening system for detecting DR into five stages, this project is developed using Convolutional Neural Networks (CNNs). The model analyzes retinal images and classifies them into five categories: No DR, Mild, Moderate, Severe, and Proliferative DR. By training the model on a diverse dataset of retinal images, the system can identify early signs of the disease with high accuracy. An accuracy matrix is used to assess the model's effectiveness. The system also includes database to manage patient information, reports, and appointment scheduling, enhancing data accessibility for healthcare providers. This solution offers a cost-effective and scalable approach to screening, enabling timely diagnosis and treatment of DR. Ultimately, this tool has the potential to improve patient care by facilitating early intervention and reducing the incidence of vision loss among diabetic patients.