文章基本信息

标题：Detection of Diabetic Retinopathy using Image Processing and Machine Learning
本地全文：下载
作者：Salman Sayed ; Dr. Vandana Inamdar ; Sangram Kapre 等
期刊名称：International Journal of Innovative Research in Science, Engineering and Technology
印刷版ISSN：2347-6710
电子版ISSN：2319-8753
出版年度：2017
卷号：6
期号：1
页码：99
DOI：10.15680/IJIRSET.2017.0601013
出版社：S&S Publications
摘要：Diabetes is a chronic disease caused by either failure of insulin production in body or inability to resistinsulin in the body. Complications of Diabetes leads to heart disorders, vascular disease and stroke, Kidney disease,Neuropathy and Diabetic eye disease known as “diabetic retinopathy”. Diabetes for a prolonged time damages theblood vessels of retina and thereby affecting seeing ability of a person and leading to diabetic retinopathy. Diabeticretinopathy is classified into two categories, non-proliferative diabetic retinopathy (NPDR) and proliferative diabeticretinopathy (PDR). In this paper, detection of diabetic retinopathy in fundus image is done by image processing andmachine learning techniques. Probabilistic Neural Network (PNN) and Support vector machines (SVM) are the twomodels adopted for detection of diabetic retinopathy in fundus image and their results analyzed and compared.Accuracy of detection in SVM is 90% and that of PNN is 80%. Thereby SVM model is outperforming the PNN model.
关键词：Diabetic Retinopathy; Probabilistic Neural network; Support vector machine; Sensitivity; Specificity.