A Systematic Review of Machine Learning Approaches for Cardiovascular Disease Diagnosis

Abstract

Diseases (CVD) are the world's biggest health issue and the leading cause of mortality, after cancer and diabetes. Early detection and prediction of CVDs are crucial for addressing the issue because they can drastically lower rates of morbidity and mortality. Physicians can diagnose a variety of cardiac conditions, including heart failure and valve dysfunction, with the use of computer-aided procedures. We live in the "information age," when millions of bytes of data are created daily. By employing the data mining technique, we can transform this data into knowledge for clinical research. Based on various risk factors, machine learning algorithms have demonstrated encouraging outcomes in the prediction of heart disease. Our goal in this study is to evaluate and analyze the results produced by machine learning methods, such as support vector machines, artificial neural networks, logistic regression, random forests, and decision trees, in order to predict CVDs. The accuracy of several machine learning algorithms in predicting cardiac issues is highlighted in this literature review, which can also serve as a foundation for developing a clinical decision-making tool to identify and stop heart illness early.