DOI: 10.12688/f1000research.193096.1 ISSN: 2046-1402

The Role of Artificial Intelligence in Reducing Maternal Mortality: A Systematic Review

Elizabet C. Jusuf, Iin Fadhilah Utami Tammasse, Johnsen Mailoa
Show PDF Cite
Abstract* Introduction Maternal mortality remains a critical global health issue, particularly in low- and middle-income countries (LMICs), where the majority of deaths are preventable. Despite extensive efforts, progress in reducing maternal mortality has been slow and uneven. Recent advances in artificial intelligence (AI) and machine learning (ML) offer promising new strategies for addressing persistent gaps in maternal healthcare, particularly through early risk identification and optimized clinical decision-making. Materials and methods A comprehensive literature search was conducted across PubMed, Scopus, ScienceDirect, ProQuest, and Medline May 2020 to June 2025, following PRISMA guidelines. Registered in PROSPERO with CRD420251051696. Eligible studies included randomized controlled trials, cohort, or case-control studies investigating AI-based approaches to reduce maternal mortality among pregnant women. Data extracted included study characteristics, AI model details, and clinical outcomes. Study quality was assessed using the ROBINS-I tools. Results Eleven studies met the inclusion criteria, encompassing diverse settings and populations. AI and ML models, such as random forest, CatBoost, Naïve Bayes, and xgboost, were used to predict outcomes including obstetric complications, gestational diabetes, sepsis, hypertensive disorders, and postpartum hemorrhage. Many models demonstrated high predictive accuracy (AUC > 0.80), enabling early identification of high-risk pregnancies and supporting timely interventions. Integrating clinical, laboratory, and social determinants of health data improved model performance. However, challenges remain regarding data quality, infrastructure, ethical considerations, and model interpretability. Conclusion AI have significant potential to reduce maternal mortality by facilitating early risk stratification and personalized care. To realize this potential, future research should focus on prospective studies, integration of multimodal data, and development of explainable, equitable AI models. Collaborative efforts across disciplines and stakeholders are essential to address implementation barriers and achieve global maternal health targets.

More from our Archive