Theoretical Perspectives on Machine Learning in the Diagnosis of Coronary Heart Disease: A Comparative Framework for Clinical Decision Support

Abstract

Coronary Heart Disease (CHD) remains one of the most prevalent causes of global mortality, posing a continuous challenge to modern healthcare systems. Traditional diagnostic methods, while effective, are often limited by subjectivity, time constraints, and the increasing complexity of patient data. Machine learning (ML) offers a theoretical framework that can revolutionize CHD diagnosis by enabling automated, data-driven decision-making. This paper explores the theoretical basis of using ML algorithms—specifically Logistic Regression, Random Forest, and Support Vector Machine (SVM)—for CHD diagnosis. We present a conceptual comparison of these models in terms of learning mechanisms, data interpretability, clinical applicability, and model complexity. Rather than focusing on numerical results, this paper provides a high-level analysis of how these models theoretically contribute to improving diagnostic accuracy, patient stratification, and clinical workflow efficiency.