Automated detection of retinal disorder using hybrid computational techniques

Abstract

Bio-Medical-Healthcare is a sector that is of high priority where the newlinepeople expect a very high level of care and service. The most essential sense newlinethat keeps us in touch with the environment is eyesight. The retina is the most newlinecrucial part of the human eye and one of its major problems is known as newlineDiabetic Retinopathy (DR) affects the vision of humans. The main stages of newlineDR are Non-Proliferative Diabetic Retinopathy (NPDR) and Proliferative newlineDiabetic Retinopathy (PDR) that depend upon the presence of clinical newlinefeatures such as Micro aneurysms (MAs), Hemorrhages (HAs) and Exudates. newlineDiabetic Retinopathy (DR) eye disease is considered a long-standing diabetes newlineproblem. As the primary signs of Diabetic Retinopathy are Exudates, early newlinedetection of DR is required to restrict the progress of disease. The detection newlinemethod called Ophthalmoscopy can be used to detect the disease at an early newlinestage which requires skilled professionals and more time to analyze the results newlineof the retinal disease. An accurate and economic detection method can help newlinethe ophthalmologists in the diagnosis of the disease with less time and low newlinecost. newlineThe main focus of this research is to identify the retinal problems newlineand perform disease classification using automated computational techniques newlinewith hybrid algorithms. These computational techniques can assist physicians newlineto examine their patients with advanced diagnostic tools and evaluate their newlineprogress more efficiently and providing successful treatment in preventing newlinevision loss. newline