AI Model for Diabetic Retinopathy and Hypertensive Retinopathy Classification on Fundus Images

Early detection of DR and HR, two significant causes of vision loss, can prevent more serious issues down the road. The study introduces an AI system that uses images of the back of the eye to identify and evaluate both DR and HR. A dataset comprising seven significant categories—Mild DR, Moderate DR, Moderate HR, No DR/HR, Proliferate DR, Proliferate HR, and Severe HR/DR—was used by the scientists to train their algorithm. After investigating several deep learning models for retinal features, they selected one that was able to identify the majority of the crucial information. Images of the eyes are uploaded, the disease is automatically predicted, the severity is explained, and a personal risk score is provided. Other components of the system include a module that helps patients comprehend their symptoms, makes therapy recommendations, and gives physicians a report that can be downloaded. The findings demonstrate that the approach is very precise and effective for various disease stages, making it beneficial for early screening and remote eye care. This work provides a useful, user-friendly AI tool that provides crucial information for evaluating DR and HR in a clinical environment.