Abstract
INTRODUCTION: Maternal mortality remains a critical public health challenge in low- and middle-income countries (LMICs), where over 92% of global maternal deaths occur. Artificial intelligence (AI)-enabled solutions are increasingly recognised for their potential to improve and expand health services delivered to women. Such solutions can accelerate how health systems address gaps in maternal healthcare, including prevention, early detection, intervention and treatment. However, the extent to which AI-enabled solutions have progressed toward real-world application in LMIC healthcare systems remains unclear. This scoping review aims to systematically map the development of AI-enabled solutions for maternal health by applying the Technology Readiness Level (TRL) framework to assess their stage of advancement. It also aims to identify facilitators, barriers and critical research gaps. METHODS AND ANALYSIS: This scoping review will be guided by established methodological frameworks and in line with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews. A comprehensive literature search will be performed across PubMed, EMBASE, IEEE Xplore, CINAHL, CABI and Scopus, as well as grey literature sources. The search will combine controlled vocabulary and keywords related to 'artificial intelligence' and 'maternal health'. Studies reporting any AI-enabled solutions in maternal health, specifically targeting the leading direct causes of maternal mortality (eg, postpartum haemorrhage, hypertensive disorders, sepsis, delivery complications and unsafe abortion), published between 1 January 2015 and 1 August 2025, will be eligible. Two independent reviewers will screen studies, chart relevant data and resolve discrepancies through consensus. Findings will be synthesised using a narrative and tabular approach to map the extent and characteristics of the literature. ETHICS AND DISSEMINATION: Ethical approval is not required as the review involves analysis of publicly available data. The findings will be disseminated through publication in a peer-reviewed journal and presentations at relevant conferences.