Heart Disease Prediction Using Machine Learning Techniques: A Quantitative Review

Abstract

Heart diseases or the cardiovascular diseases are the main reasons for a large number of deaths in the world today. Heart disease affects the functioning of blood vessels and can cause coronary artery infections that in turn lead to weakening of patient’s body. Therefore, there is a need for reliable, accurate, and feasible system that can diagnose the heart disease on time so that the cardiac patient is given an efficient treatment before it leads to a severe complication, finally resulting in a heart attack. During the last many years, machine learning (ML) algorithms and techniques have been applied to various available heart disease datasets for automatic prediction, diagnosis, and treatment of the heart disease. This paper presents a thorough survey of various machine learning techniques and analyzes their performances which are used for efficient prediction, diagnosis, and treatment of various heart diseases. Some machine learning techniques used for the prediction of the occurrence of heart diseases that are surveyed in the proposed paper are support vector machine (SVM), decision tree (DT), Naïve Bayes (NB), K-nearest neighbor (KNN), artificial neural network (ANN), etc. Then, the average prediction accuracy was calculated for each technique to find out overall best and worst performing technique. According to the results, the highest average prediction accuracy was achieved by ANN (86.91%), whereas C4.5 decision tree technique came up with the lowest average prediction accuracy of 74.0%.