Diabetic retinopathy is a leading cause of blindness in people with diabetes. Proliferative diabetic retinopathy is characterized by neovascularization of the retina as a result of a severe vascular problem. The automatic detection of such new vessels would be helpful in assessing the severity of diabetic retinopathy, and it is an important element of the screening procedure to identify those who may have the disease. Their diabetic retinopathy necessitates rapid care. The early and precise identification of proliferative diabetic retinopathy is critical for the patient's eyesight protection. Automated techniques for detecting proliferative diabetic retinopathy in digital retinal images should be able to distinguish between normal and pathological vessels. Using a multivariate m-Mediods-based classifier, statistical texture analysis (STA), high order spectrum analysis (HOS), and fractal analysis (FA), we suggested a new method for detecting aberrant blood vessels and evaluating proliferative diabetic retinopathy in this paper. The system extracts the vascular pattern and optic disc, using a multilayered thresholding technique and the Hough transform