CLASSIFICATION OF EYE DISEASES WITH MOBILENETV3 AND EFFICIENTNETB0 MODELS

Abstract

Diabetes increases the risk of eye complications such as diabetic retinopathy, cataracts, and glaucoma. It's the leading cause of blindness in working-age adults. Regular eye examinations and proper care can prevent most cases of blindness. Cataracts and glaucoma are also more common in people with diabetes. Glaucoma can result from increased pressure inside the eye and has two main types: open-angle and angle-closure. This thesis proposed the use of MobileNetV3 and EfficientNetB3 models for classifying eye diseases. The MobileNetV3 model achieved an accuracy of 73%, while the EfficientNetB3 model achieved an accuracy of 94%. The results demonstrate the potential of these models for accurately identifying eye diseases, which can have significant implications for early detection and treatment.