Classification of diabetic retinopathy using neural networks

Abstract

Classification of the severity of diabetic retinopathy (DR) and quantification of diabetic changes are vital for assessing the therapies and risk factors for this frequent complication of diabetes. A multilayer feedforward network has been developed for the classification of DR. One of its major strengths is that accurate feature extractions and accurate grading of DR lesions are not required. Another strength of this technique is its robustness as the network can also classify DR effectively in noisy environments.