A Novel Hybrid Model for Predictive Analysis of Myocardial Infarction using Advanced Machine Learning Techniques

Cardiovascular disease (CVD) remains a leading cause of mortality, posing challenges for early detection and prediction. The recent survey of interest among researchers focuses on advanced machine-learning (AML) models due to their impressive precision, accurate classification, and predictive capabilities. This area has particularly significant implications within medical cardiology, as it aims to promptly identify CVD. This study introduces an effective and precise system for detecting Myocardial Infarction (MI). The system leverages three distinct feature selection approaches-filter methods, wrapper methods, and embedded methods—in conjunction with eight ML algorithms: logistic regression (LR), k-nearest neighbors(KNN) classifier, support vector classifier (SVC), decision tree (DT), random forest (RF), gradient boosting (GB), ada boost (AB) classifier, and xgb classifier based on the performance compare all. Incorporating these methods enhances the classification model's performance while reducing computational complexity. The proposed model is evaluated using a standardized dataset, demonstrating superior predictive capabilities in terms of accuracy, sensitivity, precision, f1-score, auc, and specificity. This novel approach outperforms existing research in this domain, further underscoring its potential in advancing MI prediction and diagnosis for elder and newborn babies.