Enhanced Myocardial Infarction Prediction Using Machine Learning Algorithms and Gender-Specific Insights

Abstract

A serious medical condition called myocardial infarction (MI), sometimes referred to as a "heart attack," is caused by disruptions in the blood supply to the myocardium. This research examines the efficacy of machine learning (ML) algorithms in forecasting myocardial infarction (MI) using a dataset of 350 records. The study identifies key risk factors for predicting myocardial infarction (MI), such as elevated cholesterol levels, diabetes, advanced age, overall health status, mental well-being, obesity, physical activity, smoking habits, hypertension, and depression. Significantly, gender does not manifest as a predictor of myocardial infarction (MI) when employing various classification methods. The research achieves high accuracy rates of 89.32%, 87.53%, 81.29%, and 76.59% using different machine learning algorithms, including Deep Belief Network (DBN), C4.5, Random Forest (RF), and Bayesian Network (BN), respectively. Algorithm-specific rule sets identify correlations, with the C4.5 algorithm revealing interesting connections between smoking habits and protection against myocardial infarction (MI). Performance metrics like accuracy, precision, sensitivity, and specificity attest to the effectiveness of the proposed technique. The results demonstrate the superior performance of the DBN algorithm, surpassing other algorithms in terms of accuracy (89.32%), precision (84.04%), sensitivity (86.63%), and specificity (82.45%). This paper provides crucial insights into predictive modeling for myocardial infarction (MI), highlighting the importance of various risk factors and advanced machine learning (ML) algorithms. The results offer clinicians and researchers a strong foundation for comprehending and potentially averting myocardial infarction, relying on personalized patient profiles. This paper has the potential to significantly contribute to the field by applying ensemble classifiers and machine learning models to forecast gender-specific myocardial infarctions. As a result, diagnostic precision and patient outcomes could be revolutionized.