Enhancing Heart Disease Diagnosis with Hybrid Machine Learning Models: A Case Study

Heart disease remains one of the leading causes of morbidity and mortality worldwide, prompting extensive research into accurate and early detection methods. Recent advancements have highlighted the critical role of Artificial Intelligence (AI) and Machine Learning (ML) in enhancing diagnostic precision. This study explores the effectiveness of hybrid ML models in diagnosing heart disease, specifically focusing on two novel combinations: Random Forest (RF) integrated with Sequential Minimal Optimization (SMO) and J48 decision trees augmented with Logistic Regression (LR). Using a comprehensive heart disease dataset, the models were evaluated based on their classification accuracy, class separability, and risk prediction capability. Both models incorporated advanced preprocessing, cross-validation, and hyperparameter optimization techniques. The RF–SMO model achieved an accuracy of 97% and a Receiver Operating Characteristic (ROC) area of 0.97, while the J48–LR model attained 92% accuracy with a ROC area of 0.96. These findings underscore the potential of hybrid ML approaches to enhance cardiac diagnostics, offering valuable tools for clinical decision support and the advancement of personalized healthcare.