From Prediction to Prevention: The Role of AI in Transforming Coronary Artery Disease Risk Assessment

Background: Artificial intelligence (AI) is reshaping the landscape of coronary artery disease (CAD) prevention through its ability to enhance risk prediction, early detection, and individualized interventions. Objective: This narrative review examines the current role of AI-based models in CAD prevention, evaluating their predictive accuracy, clinical applications, and implementation challenges. Methods: We synthesized evidence from recent systematic reviews, meta-analyses, and original studies on machine learning (ML) and deep learning (DL) techniques using multimodal data such as electronic health records (EHR), electrocardiograms (ECG), and imaging. Key Findings: AI models consistently outperform traditional risk scores like Framingham and ASCVD in predictive performance, especially when multimodal data integration is applied. These models show particular promise in high-risk and complex populations. Additionally, AI tools contribute to clinical decision-making, including revascularization planning and precision phenotyping. However, critical limitations remain—most notably limited external validation, opacity in model explainability, and bias stemming from non-representative datasets. Conclusions: While AI offers transformative potential in CAD prevention, responsible deployment requires addressing ethical, technical, and systemic challenges. Key strategies include improving model transparency, ensuring fairness across populations, and embedding AI tools seamlessly into clinical workflows. The success of future systems will depend on explainability, human-AI collaboration, and meaningful stakeholder engagement.