A Comparative Study of Bayesian, Tree-Based, and Ensemble Machine Learning Models for Clinical Heart Disease Prediction

Abstract

Cardiovascular disease remains one of the most pressing health challenges globally, contributing significantly to disability and mortality among both men and women. Traditionally, diagnosing heart disease relies heavily on comprehensive clinical assessments, including patient history and physical examination. However, the growing global population has outpaced the availability of cardiologists, especially in regions with the highest demand for medical expertise. This imbalance increases the risk of delayed or missed diagnoses. To address this gap, the use of Artificial Intelligence (AI), particularly Machine Learning (ML), has gained prominence in healthcare. ML-based systems can support or augment traditional diagnostic processes. This study explores the application of three machine learning algorithms, which are naïve Bayes, decision tree, and random forest, for predicting the likelihood of a patient suffering from heart disease. Comparative analysis indicates that the Random Forest algorithm had the best performance in predicting the likelihood of a patient suffering from heart disease, thereby showing the power of an ensemble model and its recommended usage in automated diagnostic systems.