Analysis of Retinal Detachment Severity using RDidean Evaluative System and Classifier Assessment Implementation Models

Abstract

Vision loss can be a permanent disability of a human, that could be attributed to due to Retinal Detachment (RD). This is perilous disorder that could be caused due to dealignment of the layers in a retina. The choroid supplies oxygen and nutrients to the outer segments of the photoreceptors. The retina's photoreceptors will stop working if the choroid separates from the retina. The degree of dependence on the choroid is high, due to its supply of oxygen to the fovea that increases the breathability of the retinal blood vessels. This tapering of the oxygen levels can paramount to macula detachment that can be an irreversible damage to the cones and rods at the posterior pole, thereby leading to blindness. If the retina is quickly reattached, good vision can be preserved even if the macula is not removed. Retinal detachment hitherto has been studied through conceptual data processing and image processing techniques. However, this paper analyzes the retinal detachment for an individual through the various features, and the same is encompassed in an equational form to be termed as the “RDidean” evaluative system. The evaluative model thus explicitly categorizes the retinal database into normal and abnormal images based on the value obtained from the system. The performance of this system is then effectuated through diverse tree classifier models and the deep learning AlexNet classifier in MATLAB to comprehend the precision of classification. While effectively entailing another pool of algorithmic models like the SVM variations and the Naïve Bayesian methods to cognize the accuracy of retinal detachment severity that the evaluative system rendered. This indagation analyzes focusses to establish a corroborative and impeccable prediction system for the classification of normal and abnormal eye through color fundus images. Thereby aiding to improve the ergonomic environment of clinicians to improve the treatment plan, along with delivering complementary clinical decisions, and in institutionalizing affordability for patients through optimal cost for agnizing retinal detachments.