Discovering Optimal Algorithm to Predict Diabetic Retinopathy using Novel Assessment Methods

• Main Authors: Shiva Reddy, Nilambar Sethi, R. Rajender
• Format: Article
• Language: English
• Published: European Alliance for Innovation (EAI) 2021-01-01
• Series: EAI Endorsed Transactions on Scalable Information Systems
• Subjects: diabetic retinopathy; random forest; decision tree; adaptive boosting; bagging; support vector machine (svm) using gaussian kernel (gk); accuracy; youden’s j index; concordance; somers’ d statistic and balanced accuracy
• Online Access: https://eudl.eu/pdf/10.4108/eai.13-7-2018.165505

Diabetic retinopathy is a diabetes complication that effects eyes. It disrupts the vasculature of the sensitive tissue present atthe back of the eye. If this complication is untreated it may lead to blindness. The aim of this work is to train a model thatefficiently predicts diabetic retinopathy. Machine learning techniques like Decision tree, Random forest, Adaptiveboosting and Bagging are used as primary algorithms to train predictive models. An algorithm namely ‘Support VectorMachine using Gaussian kernel for retinopathy prediction’ is proposed in this work. The proposed algorithm is comparedwith the primary algorithms based on five evaluation metrics namely accuracy, Youden’s J index, concordance, Somers’ Dstatistic and balanced accuracy. From the results obtained the proposed algorithm obtained better values for all consideredevaluation metrics. Thus the use of SVM with Gaussian kernel is proposed to be used for prediction of diabeticretinopathy.