Detection of diabetic retinopathy disease by classifying fundus images using deep learning

Diabetic retinopathy, which is due to the side effects of diabetes, one of today's common diseases, is one of the most serious diseases in the world. Diabetic retinopathy (DR) is diabetes-related damage to the blood vessels in the retina of the eye. It develops due to diabetes. Diabetic retinopathy diagnosis is usually based on expert examination of fundus images. The fact that the examination process is only expert-based makes disease diagnosis difficult. Only experts managing this process can prolong the diagnosis and treatment process of the disease. Diagnosing and treating diabetic retinopathy in the early stages is vital to preserve the vision of diabetic patients and slow the progression of retinopathy. The aim of this study is to investigate the classification and detection of diabetic retinopathy, which doctors diagnose by examining fundus images, using a deep learning model. The research aimed to increase the performance of the model by using techniques such as image enhancement and data augmentation.