AI-Assisted Global Regulatory Intelligence for Software as a Medical Device (SaMD): A Traceable, Risk-Based Framework for Multijurisdictional Compliance
Abstract: Background: Software as a Medical Device (SaMD), including software functions that use artificial intelligence (AI) or machine learning (ML), is governed by evolving and partly divergent regulatory requirements. Manufacturers operating across markets must identify applicable requirements, interpret their legal status, map them to quality-system processes, and maintain evidence that requirements remain current. Manual regulatory surveillance is difficult to scale and can lead to missed changes, inconsistent interpretations, and weak traceability. Objective: This paper proposes a human-supervised, AI-assisted global regulatory intelligence (GRI) framework for SaMD. The framework links authoritative-source monitoring, structured requirement extraction, jurisdictional comparison, quality-management-system mapping, risk-based gap assessment, and controlled change implementation. Methods: A conceptual framework was developed through a targeted review of publicly available regulatory and consensus documents from the International Medical Device Regulators Forum (IMDRF), the US Food and Drug Administration (FDA), the World Health Organization (WHO), the Central Drugs Standard Control Organization (CDSCO), and the International Organization for Standardization (ISO). The approach is a narrative, document-based synthesis rather than a systematic review or empirical validation study. Results: The proposed architecture contains eight connected modules: source governance; regulatory change detection; document classification and version control; requirement extraction and semantic normalization; jurisdictional applicability; evidence and QMS mapping; risk prioritization; and human approval with audit-trail retention. A requirement-level data model and a staged validation plan are provided. The framework explicitly separates binding law, regulator guidance, consensus documents, and voluntary standards. Conclusion: AI can improve the speed, consistency, and traceability of SaMD regulatory intelligence, but it cannot replace accountable regulatory interpretation or legal review. A defensible system should use authoritative sources, preserve source-level evidence, disclose uncertainty, prevent unsupported inference, and validate performance against expert-labeled regulatory requirements. The proposed framework is a research and implementation blueprint; its effectiveness remains to be demonstrated empirically.
Keywords: Software as a Medical Device; SaMD; artificial intelligence; regulatory intelligence; regulatory change management; medical-device software
How to Cite?: Dr. Pushpinder Paul, "AI-Assisted Global Regulatory Intelligence for Software as a Medical Device (SaMD): A Traceable, Risk-Based Framework for Multijurisdictional Compliance", Volume 15 Issue 8, August 2026, International Journal of Science and Research (IJSR), Pages: 2204-2209, https://www.ijsr.net/getabstract.php?paperid=SR260821085249, DOI: https://dx.doi.org/10.21275/SR260821085249