Ischemic heart disease detection using support vector Machine and extreme gradient boosting method

Abstract The healthcare practitioners in recent days, there is a challenging task that is predicting and detection of heart disease which is also critical. A timely administration of appropriate treatment is allowed by an accurate and early diagnosis of heart disease and also helps to reduce the mortality. Thus, this paper proposed a swift and exact automatic ischemic heart disease detection using support vector machine (SVM) and extreme gradient boosting (XGD) method with the help of random forest technique for training and testing of dataset. The extraction of T –wave has 164 features and was segmented from averaged MCG recordings. To identify IHD (Ischemic Heart Disease) case, the machine learning classifiers including SVM and XG-Boost are applied with the Z-Alizadeh sani Heart Dataset (HD).Moreover, for parallel selection of features and the cross validation-partitioning were used twice with the optimization of classifier parameters. Then these parameters are trained and tested using a random forest technique. The demonstration of these experiments shown that the accuracy is 93.08% while predicting heart disease detection outcomes among the patients that is provided by the N2Genetic-nuSVM consist of well known Z-Alizadeh Sani dataset.