Comparative Study of Diabetic Retinopathy Detection Using Machine Learning Methods

Abstract

Untreated diabetic retinopathy, a consequence of poorly managed chronic diabetes, can lead to complete vision loss. Early diagnosis and treatment are crucial to prevent severe complications. Currently, ophthalmologists dedicate significant time to manually diagnose diabetic retinopathy, causing discomfort to patients. Automated technologies offer a promising solution by swiftly identifying diabetic retinopathy and facilitating timely treatment to mitigate further ocular damage. This study proposes leveraging machine learning to extract and classify key features such as exudates, hemorrhages, and microaneurysms using a hybrid classifier combining support vector machines, k-nearest neighbors, and random forests.