RETINOSCAN: Retinal Image-Based Diabetic Retinopathy Detection

Diabetic Retinopathy (DR) is a leading cause of vision impairment worldwide, necessitating early detection and accurate diagnosis. This study presents a reliable system that leverages advanced image processing and deep learning techniques to identify and classify DR stages. MATLAB is utilized for fundus image preprocessing, integrating a pre-trained Inception-v3 model for feature extraction and stage-wise classification. A web-based graphical user interface enhances clinical applicability. Deployment through MATLAB Compiler and Docker ensures real-world accessibility. The model, trained and validated on a comprehensive retinal image dataset, demonstrates high reliability and accuracy. This system has the potential to improve patient outcomes and establish new standards in automated retinal disease diagnostics.