Detection of non-proliferative diabetic retinopathy lesions using wavelet and classification using K-means clustering

Abstract

WHO predicts that in year 2012 there are about 347 million people worldwide have diabetes, more than 80% of diabetes deaths occur in different countries. WHO projects that diabetes will be the 7th major cause leading death in 2030. Diabetic Retinopathy caused by leakage of blood or fluid from the retinal blood vessels and it will damage the retina. Non-proliferative diabetic retinopathy (NPDR) is an early stage of diabetic retinopathy and it is categorized into three stages they are mild, moderate and sever NPDR. The characteristic of the Mild; is specified by the presence of minimum microaneurysm, Moderate; specifies the presence of hemorrhages, microaneurysms, and hard exudates where as Severe; determine on the blockage of vessels, depriving several areas of the retina. With their blood supply. These areas of the retina send signals to the body to grow new blood vessels for nourishment. The proposed algorithm tested on online databases like STARE, DRIVE, DiarectDB0, DiarectDB1 and SASWADE (the database collected during the research work). The statistical techniques were applied on NPDR lesion and calculate the mean, variance, standard deviation, & correlation for classification. K-means clustering have been applied on the dataset with extracted features 95% of correct classification have been achieved.