DIABETES DISEASE PREDICTION USING MACHINE LEARNING ALGORITHMS

Diabetes can lead to a condition known as diabetic retinopathy (DR), which affects the eyes by damaging the blood vessels in the retina’s light-sensitive tissue. When diabetes is poorly managed, DR becomes the leading cause of blindness among working-age individuals. Current methods for detecting diabetic retinopathy rely on manual examination of retinal images by ophthalmologists. The proposed approach aims to automate this detection process. The model was trained on 35,126 retinal scans, accessible publicly on Kaggle through eyePACS, using a GPU. It achieved an accuracy rate of approximately 81%.