A Novel approach for the detection of diabetic retinopathy disease

Diabetic retinopathy is the most common cause of blindness of the eye depend on diabetes. In this work, a novel approach is presented for the detection of diabetic retinopathy diseases from the retina images. For this purpose, firstly regions which are probably diseased are found and features are extracted from these regions by applying Discrete Wavelet Transform. Afterwards the number of found features is reduced by Principal Component Analysis and Naïve Bayes is used for the classification of them. This approach differs from the similar works by the way Region of Interest is found and the automatic selection of features instead of using hand-picked ones. It has been shown that the proposed system achieves an accuracy rate up to the 95% in the detection of the diseased retinas.