Artificial Intelligence in Drug-Coated Cardiovascular Devices: A Narrative Review

Drug-coated cardiovascular devices (DCCDs), including drug-eluting stents (DESs) and drug-coated balloons (DCBs), have significantly advanced interventional cardiology by reducing restenosis and improving long-term outcomes. However, their effectiveness is limited by challenges such as patient-device mismatch, variability in drug delivery kinetics, and dependence on operator experience. Traditional strategies for device selection and performance evaluation are often inadequate to address patient-specific complexities. This narrative review aims to explore how artificial intelligence (AI) can improve the design, deployment, and monitoring of DCCDs, focusing on personalized treatment strategies, regulatory implications, and future innovations in interventional cardiology. A targeted literature search was conducted in PubMed, Scopus, and Web of Science between 2020 and 2025 using keywords such as "artificial intelligence", "drug-eluting stents", "cardiovascular devices", "machine learning", and "intravascular imaging". Studies were included based on their relevance to AI applications in DCCD design, procedural support, or post-procedural monitoring. AI has demonstrated significant potential throughout the DCCD lifecycle. In design, machine learning models enable optimization of drug release kinetics and device geometry. During procedures, AI improves real-time intravascular imaging interpretation and provides guidance for precise device placement. Post-intervention, predictive analyses using patient data can aid in the early detection of complications such as in-stent restenosis. Furthermore, technical, regulatory, and ethical challenges remain, including model validation, data bias, and the need for transparency in decision-making algorithms. AI-driven approaches offer a promising paradigm for advancing cardiovascular device technology toward more adaptable, personalized, and efficient care. Integrating explainable, clinically validated AI systems with DCCDs can improve outcomes, reduce procedural variability, and support value-based care. Future research should prioritize real-time intraoperative feedback systems, adaptive AI models based on longitudinal patient data, and regulatory compliance and fairness strategies.