Real-World Evidence and Adaptive Governance Paradigms for AI-Enabled Medical Devices: A Framework for Post-Market Safety Surveillance and Precision Regulatory Science

ABSTRACT Background and Objective: AI medical devices are applied for diagnosis, patient monitoring, risk stratification, and therapy selection. These devices can be updated after their release through software changes and algorithm improvement. Thus, their initial clinical evaluation cannot entirely guarantee post-market safety and efficacy. This study aims to assess the role of real-world data (RWD), real-world evidence (RWE), and artificial intelligence in the risk stratification and post-market surveillance of AI-based medical devices. Critical evidence gaps and regulatory needs were identified. Methods: A narrative review informed by a literature search was undertaken in PubMed, Scopus, Web of Science and Google Scholar. In parallel, regulatory guidelines issued by the U.S. Food and Drug Administration (FDA), the European Medicines Agency (EMA), the Medical Device Regulation (MDR) and the International Medical Device Regulators Forum (IMDRF) were retrieved and scrutinized. After screening, the evidence was thematically synthesized in terms of validation methods, surveillance targets, regulatory activities and risk levels. Results: AI-driven analytical methods can contribute in clinical classification of healthcare data and provide relevant real-world evidence for post-market surveillance, indication expansion, safety-signal detection, risk prediction and regulatory decision-making. The main challenges include limited prospective and external validations, population heterogeneity, algorithmic bias, interoperability, cybersecurity issues and inconsistent post-market surveillance practices between jurisdictions. Conclusion: The convergence of artificial intelligence (AI), real-world data (RWD), and real-world evidence (RWE) can enable life cycle–oriented, adaptive regulation of medical devices. This Review proposes an Evidence Maturity Matrix and a Regulatory Readiness Framework to guide evidence generation and regulatory preparedness across the product life cycle. Keywords: Artificial intelligence; Real-world data; Real-world evidence; AI-enabled medical devices; post-market surveillance; Risk stratification; Precision regulatory science; Regulatory readiness.