Detection of Diabetic Retinopathy using Optimized Back-Propagation Neural Network (Op-BPN) Algorithm

Abstract

Among many people with diabetics, Diabetic Retinopathy is a human eye disorder that causes damage to the retina of the eye. If not treated in a timely manner, the patient can ultimately reach a condition called total blindness. Although it needs early detection and constant monitoring of diabetic patients, successful therapies for DR are accessible. For the diagnosis of diabetic retinopathy, certain physical exams, such as visual acuity checks, pupil dilation, and optical coherence tomography may also be used but they consume more time. The main aim of this study is to improve the diagnostic accuracy of diabetes retinopathy disease and reduce the time taken to detect using Optimized Back-propagation Neural Network (Op-BPN) algorithm based on the features extracted from different retinal image output. The outcome of the proposed classification model (Op-BPN) is measured to build its dominance in terms of accuracy, error, precision and recall. Simulation result has been proved that, the proposed algorithm performs efficiently when compared with the existing algorithm.