Identification of Proliferative Diabetic Retinopathy Using Texture Segmentation

Abstract

In this study, it is proposed fast texture segmentation of detection of proliferative diabetic retinopathy. proliferative diabetic retinopathy is characterized by the revascularization on the optic disc or on the retinal area or both, as a result of diabetes mellitus. It may leads to the blindness if it is not detected and treated in proper time. This study proposes a novel method to detect the vessel growth on the optic disc using texture segmentation. The vessels that grow on the optic disc as a result of proliferative diabetic retinopathy will be more convoluted than the normal vessel. The segmentation of the optic disc is made using the texture descriptors. Next, the segmentation of blood vessel is made by Ridge strength and Watershed Transformation. Fifteen features which represent the shape, gradient of the vessel are calculated. The calculated features are separated using a hyperplane to classify it as normal or abnormal by support vector machine classifier. We have observed a sensitivity and specificity of 88.89 and 91.3%.