AI-Enabled Clinical Decision-Support Interventions in Maternal Health: A Scoping Review of Reviews on Implementation Readiness, Ethical and Governance Considerations, and Equity Implications across Low- and Middle-Income Countries

Background: Globally, approximately 267,000 women die each year from preventable maternal causes, with more than 94% of these deaths occurring in low- and middle-income countries (LMICs). Clinical decision support systems (CDSS) have been developed to improve adherence to evidence-based guidelines; however, their effectiveness is constrained by incomplete data, lack of contextualization, and usability challenges. Emerging applications of artificial intelligence (AI), including machine learning, natural language processing, and large language models, offer opportunities to strengthen CDSS through risk prediction, context-aware recommendations, and real-time support for health providers. However, integrating AI into maternal health CDSS also raises important questions about data quality, transparency, equity, and governance. This scoping review maps and critically synthesizes the global evidence on AI-enabled clinical decision-support interventions in maternal health, with particular attention to implementation readiness, ethical and governance considerations, and equity implications.Methods: The review follows the six-stage methodological framework of Arksey and O’Malley and the JBI 2020 methodology for scoping reviews and will be reported according to the PRISMA extension for Scoping Reviews (PRISMA-ScR). A librarian-conducted search will cover PubMed, Embase, Cochrane Library, Scopus, CINAHL, Web of Science, Global Health (CABI), and PsycINFO from inception to June 2026, including preprint databases (arXiv, medRxiv) and the WHO Global Index Medicus. Published reviews (systematic reviews, scoping reviews, narrative reviews, rapid reviews, and meta-analyses) published in English or French will be included. Two independent reviewers will screen, critically appraise, and extract data. Conflicts will be resolved by a third reviewer. Two independent reviewers will screen, appraise, and extract data. Conflicts will be resolved by a third reviewer. Data will be extracted using a standardized, pre-piloted extraction form; an AI lifecycle framework and clinical maturity scale will assess implementation readiness. A narrative synthesis supported by descriptive statistics and visualizations, where applicable, will be produced.Anticipated results: This review is expected to provide a comprehensive map of AI applications in maternal health CDSS across the care continuum, including antenatal, intrapartum, and postnatal care, and HIV-related prevention of mother-to-child transmission (PMTCT) services. Anticipated outputs include: (1) a typology of AI approaches used; (2) the clinical and health system functions they support; and (3) the outcomes most frequently assessed, including implementation readiness, ethical and governance considerations, and equity implications.