Pre-diagnosis of Diabetic Retinopathy using Blob Detection

Diabetic retinopathy is an ailment that harms to the retina because of diabetes mellitus. Starting at now, diagnosing DR is tedious and it is a manual procedure that requires a clinical master to review and break down the fundus pictures. While this strategy is feasible, as the number of individuals with diabetes continues expanding, raises the requirement for programmed DR screening strategies that guarantee serving the mass populace. The proposed method in this paper bring forward an effective feature extraction technique based on blob detection followed by classification of different stages of diabetic retinopathy using machine learning technique. This feature extraction technique could help automatic characterization of retina images for diabetic retinopathy with an accuracy of 83 per cent with the most efficient machine learning classification algorithm, which would help specialists to handily recognize the patient's condition in a progressively precise manner.