Cardiovascular disease prognosis using effective classification and feature selection technique

Abstract

Cardiovascular disease is a worldwide health problem and according to American Heart Association (AHA), it also causes an approximate death of 17.3 million each year. Therefore early detection and treatment of asymptomatic cardiovascular disease which can significantly reduce the chances of death. An important fact regarding such life-threatening disease prognosis is to identify the patient's physical state (healthy or sick) based on the analysis of health checkup data. This paper aims at optimized cardiovascular disease prognosis using different data mining techniques. We also provide a technique to improve the accuracy of proposed classifier models using feature selection technique. Patient's data were collected from Department of Computing of Goldsmiths University of London. This dataset contains total 14 attributes in which we applied SMO (SVM - Support Vector Machine), C4.5 (J48 - Decision Tree) and Naïve Bayes classification algorithms and calculated their prediction accuracy. An efficient feature selection algorithm helped us to improve the accuracy of each model by reducing some lower ranked attributes. Which helped us to gain an accuracy of 87.8%, 86.80% & 79.9% in case of SMO, Naïve Bayes and C4.5 Decision Tree algorithms respectively.