Predicting coronary artery diseases using effective features selected by Harris Hawks optimization algorithm and support vector machine

With 17 million annual deaths, cardiovascular diseases are the leading cause of mortality across the world with coronary artery disease (CAD) as the most prevalent one. CAD is the leading cause of death in industrial countries and at the same time is rapidly spreading in the developing world. Thus, the development and introduction of machine learning methods for the accurate diagnosis of heart diseases, especially CAD, have been an important debate in recent years in order to overcome relevant problems. The aim of this paper was to propose a model for enhancing CAD prediction accuracy. It sought a framework for predicting and diagnosing CAD using the features selection of Harris Hawks Optimization algorithm (HHO) and Support Vector Machine (SVM). The heart disease data set of Cleveland hospital available in the University of California Irvine (UCI) was used as the studied data set. It included 303 cases. Each case had 14 features with the final medical status of cases (CAD or normal case) as one of the features where 165 and 138 cases were diagnosed as CAD and normal, respectively. The results of this study revealed that HHO could enhance CAD diagnosis accuracy.