Machine learning for precision obstetrics: A comprehensive review of risk prediction models for hypertensive disorders and pregnancy-related syndromes
Abstract
Gestational diabetes mellitus (GDM), hypertensive disorders of pregnancy (HDP), preterm birth, and intrauterine growth restriction represent major contributors to maternal and neonatal morbidity worldwide. Traditional screening methods relying on single biomarkers or linear models often demonstrate limited predictive accuracy. This review examines the transformative role of machine learning (ML) in shifting obstetric care toward predictive, preventive, and personalized approaches. Advanced computational models integrating multimodal data sources clinical records, biochemical markers, multi-omics profiles, medical imaging, and lifestyle factors show promising performance. Ensemble methods such as Random Forest and XGBoost, alongside deep learning architectures, frequently outperform conventional logistic regression, achieving AUC values above 0.90 in selected cohorts. Emphasis is placed on preprocessing techniques for class imbalance (e.g., SMOTE), model interpretability via Explainable AI (SHAP), and privacy-preserving strategies like federated learning. While technical and ethical challenges including bias, external validation, and data heterogeneity remain, robust ML frameworks offer substantial potential for early risk stratification and timely intervention in precision obstetrics. Prospective, multicenter validation is essential for clinical translation.
How strong is the evidence?
Evidence Hub reads this as a review, done in computer models (in silico), not in people. Its evidence level is 5 of 5 (Very low · case / narrative / animal / in vitro / in silico). Case reports, narrative reviews, and animal, cell or computer-model studies give the weakest evidence for human health.
Read the paper
Related studies
- Active Learning during Federated Learning for Object Detection
- Fairness-aware federated learning
- Decentralized Federated Learning
- Building a Personally Identifiable Information Recognizer in a Privacy Preserved Manner Using Automated Annotation and Federated Learning
- Preservation of Patient Level Privacy: Federated Classification and Calibration Models
- Efficient explainable federated learning
Go deeper in Evidence Hub
Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.