A Novel Ensemble Method for Predicting Heart Disease Using Quine McCluskey Binary Classifiers (QMBC)

Abstract

Heart disease is a major public health concern, requiring accurate and interpretable risk assessment models. In order to improve interpretability and prediction performance, we suggest a Novel Ensemble Approach that integrates with the Quine McCluskey Binary Classifier (QMBC). Traditional models often lack transparency, limiting their clinical utility. Our method combines ensemble learning with QMBC’s rule-based decision-making to address this issue. The framework integrates multiple base classifiers trained on different feature subsets from heart disease datasets. A systematic feature selection process identifies key risk factors, ensuring clinical relevance. QMBC then generates interpretable rules, clarifying decision logic. The "johnsmith88 heart disease" dataset is used to validate our methodology, and performance indicators like accuracy, sensitivity, specificity, and AUC-ROC are used. Expert reviews and comparative analyses with existing models assess interpretability and clinical usefulness. This ensures our model's robustness and practical applicability. Further, we evaluate generalizability through cross-validation and external validation across diverse populations and healthcare settings. Scalability is also explored, integrating emerging risk factors and advancements in medical imaging. Our adaptable framework aims to improve heart disease risk assessment, offering a transparent and clinically useful tool for better decision-making in healthcare and public health.