Machine Learning-Based Prediction of First-Trimester Preeclampsia: A Systematic Literature Review

Abstract

Background: Machine learning (ML) has emerged as a promising approach for the early prediction of preeclampsia, enabling timely intervention during pregnancy. However, evidence regarding the performance and generalizability of ML models remains fragmented.Objective: This systematic literature review aimed to synthesize current evidence on ML algorithms, predictive performance, and factors influencing the prediction of first-trimester preeclampsia.Methods: A systematic literature review was conducted following the PRISMA 2020 guidelines. Articles published between 2016 and 2026 were identified through Scopus, ScienceDirect, and PubMed. Studies investigating ML-based prediction models for first-trimester preeclampsia were included, and data were synthesized narratively.Results: Twenty two studies met the inclusion criteria. Random Forest was the most frequently applied algorithm, although XGBoost, Support Vector Machine, Artificial Neural Network, and ensemble learning methods also demonstrated promising performance. Most models achieved an AUC above 0.85 by integrating maternal characteristics, blood pressure, laboratory findings, biomarkers, and ultrasound parameters. However, most studies relied on internal validation, while external validation remained limited.Conclusion:  Machine learning shows strong potential for first-trimester preeclampsia prediction. Future studies should emphasize multicenter external validation, standardized reporting, and explainable artificial intelligence to improve model generalizability and support clinical implementation.