AI-Driven Approaches for Maternal Health Risk Prediction, Gestational Diabetes Detection, and Postpartum Depression Screening: A Literature Survey
Pregnancy-related complications such as gestational diabetes mellitus (GDM), hypertension, anaemia and postpartum depression (PPD) remain significant contributors to maternal morbidity and mortality, particularly in rural and resource-limited settings. This paper surveys recent literature on the application of Artificial Intelligence (AI), Machine Learning (ML), Explainable AI (XAI) and Natural Language Processing (NLP) to maternal healthcare, covering twenty studies published between 2021 and 2026. The reviewed work is organised into four themes: general maternal-risk classification, GDM prediction, PPD detection, and maternal-health dataset creation. Across these themes, a consistent pattern emerges — most systems solve one problem in isolation, provide little transparency behind their predictions, stop at prediction without offering actionable follow-up, and are validated on data that does not represent rural or low-resource populations. Based on these findings, the paper argues for an integrated, explainable and recommendation-driven maternal-health platform, motivating the design of the proposed system, MaaSuraksha.
Authors
- Suruchi Pandey (ORCID: https://orcid.org/0000-0003-2752-9468)
- Siri P
- Vinutha H K
- Suhani S Rao
Publication Details
- Journal
- Zenodo (CERN European Organization for Nuclear Research)
- Published
- 2026-11-13
- DOI
- https://doi.org/10.5281/zenodo.24184750
- Primary Topic
- Artificial Intelligence in Healthcare
- Type
- article
- Field-Weighted Citation Impact
- 0.00