K7. Retinal Blood Vessel Image Segmentation using Rotating Filtration to Help in Early Diagnosis and Management Diabetic Retinopathy

A new technique has been proposed for early detection of diabetic retinopathy. Retinal blood vessel appearance in retina is good clinical sign of diabetes. This paper-by use of digital image processing segmentation with rotating filtration; minute and big retinal blood vessels are isolated and visualized. Four different edge detection filters are used individually. Each filter was applied on the retinal image. This filtration process been repeated with rotating filter pattern each time. Literature survey for the work done in this trend is mentioned. The algorithm and previous work performance was evaluated numerically and the achieved ROC curve demonstrates the high accuracy of using Laplacian of Gaussian filter. And it is expected to be welcomed by ophthalmologists to be used in diagnosing early stages of Diabetic Retinopathy.