Recent evidence in the literature suggests that Artificial intelligence (AI) is rapidly becoming more clinically relevant with expanding applications across cardiovascular medicine and cardiothoracic surgery. Advances in computational power and the widespread digitization of clinical data have enabled AI models to identify complex, nonlinear patterns across multimodal datasets, positioning them as powerful tools for diagnosis, risk stratification, and procedural decision support. This review examines the current and emerging landscape of AI in cardiac care, with a particular focus on valvular heart disease. We synthesize evidence spanning diagnostic applications such as electrocardiographic and echocardiographic interpretation, preoperative planning, and risk prediction for surgical and transcatheter interventions, and real-time intraoperative decision support. Across these domains, AI systems frequently demonstrate performance comparable to or exceeding conventional approaches, particularly in automating standardized tasks and enabling personalized risk assessment. However, most evidence to date derives from retrospective studies, and challenges related to generalizability hold significant barriers to widespread adoption. We further discuss ethical considerations necessary for safe and equitable implementation. Overall, AI shows substantial promise to augment cardiovascular care across the continuum of practice, but its successful translation into routine clinical use will require rigorous prospective validation, transparent model development and interpretability, and carefully designed integration into existing clinical workflows.
Anand et al. (Mon,) studied this question.