Symptom Analysis of Diabetic Retinopathy by Micro-Aneurysm Detection Using NPRTOOL

Abstract

Diabetic retinopathy is known to be a hazardous eye disease which can cause vision loss or blindness in long run. Recently, this disease has spread dangerously among people who are suffering from diabetes. In this paper, a diabetic retinopathy identification technique has been proposed by utilizing rear end eye image i.e., fundus image obtained from the retina of the patient. This method involves digital image processing on fundus image. Micro-aneurysm, which is considered as first stage of diabetic retinopathy has been diagnosed using MATLAB Neural Network Pattern Recognition Tool (NPRTOOL). The proposed method has been compared to two other existing methods- Naive Bayes method and SVM classifier method. In this research, NPRTOOL tool has been used for calculation and experimental work. Significant improvement has been found in terms of sensitivity and accuracy compared to the aforementioned existing methods.