Diabetic retinopathy grading system based on transfer learning

Much effort is being made by the researchers in order to detect and diagnose diabetic retinopathy (DR) accurately automatically. The disease very dangerous as it can cause blindness suddenly if not continuously screened. Therefore, many computers aided diagnosis (CAD) systems have been developed various DR grades. Recently, CAD based on deep learning (DL) methods adopted get merits diagnosing pathological abnormalities of disease. In this paper, we present a full based-DL system depending multi-label classification. proposed DL system, customized efficientNet model early advanced grades Learning transfer useful training small datasets. We utilized IDRiD dataset. It experiments manifest that robust, reliable, deigns promising results detecting grading DR. achieved accuracy (ACC) equals 86%, Dice similarity coefficient (DSC) 78.45.