A Comprehensive Analysis on Risk Prediction of Heart Disease using Machine Learning Models

Abstract

Most of the deaths worldwide are caused by heart disease and the disease has become a major cause of morbidity for many people. In order to prevent such deaths, the mortality rate can be greatly reduced through regular monitoring and early detection of heart disease. Heart disease diagnosis has grown to be a challenging task in the field of clinically provided data analysis. Predicting heart disease is a highly demanding and challenging task with pure accuracy, but it is easy to figure out using advanced Machine Learning (ML) techniques. A Machine Learning approach has been shown to predict heart disease in this approach. By doing this, the disease can be predicted early and the mortality rate and severity can be reduced. The application of machine learning techniques is advancing significantly in the medical field. Interpreting these analyzes in this methodology, which has been shown to specifically aim to discover important features of heart disease by providing ML algorithms for predicting heart disease, has resulted in improved predictive accuracy. The model is trained using classification algorithms such as Decision Tree (DT), K-Nearest Neighbors (K-NN), Random Forest (RF), Support Vector Machine (SVM). The performance of these four algorithms is quantified in different aspects such as accuracy, precision, recall and specificity. SVM has been shown to provide the best performance in this approach for different algorithms although the accuracy varies in different cases.