Machine Learning Prediction Models for Preeclampsia: Systematic Review and Meta-Analysis
Machine-Learning-Vorhersagemodelle für Präeklampsie: systematische Übersicht und Metaanalyse
Journal of Medical Internet Research · 19.01.2026 · 28 · e78714
DOI: 10.2196/78714 ↗Auf einen Blick
26 Studien mit 31 ML-Modellen wurden ausgewertet. Die gepoolte AUC war hoch, die Heterogenität jedoch extrem; die externe Leistungsfähigkeit war deutlich unsicher.
Fragestellung
Wie gut sagen ML-Modelle Präeklampsie bei Schwangeren voraus?
Methode
Systematische Übersicht und Metaanalyse nach PRISMA; Suche in PubMed, Web of Science, IEEE Xplore und CNKI bis Februar 2025; Biasbewertung mit PROBAST.
Ergebnisse
26 Studien mit 31 ML-Modellen. AUC 0,91 (95 %-KI 0,87-0,92); I2>99 %. 95 %-Prädiktionsintervall der Sensitivität 0,32-0,96. In 6 extern validierten Studien: Sensitivität 0,68; PI 0,25-0,94.
Einschränkungen
Nur 6 Studien mit externer Validierung; extreme Heterogenität; überwiegend interne Leistungsdaten.
Schlussfolgerung der Autor:innen
Die hohe AUC spiegelt laut Autor:innen eher die interne Entwicklungsleistung als eine allgemeine klinische Wirksamkeit in unabhängigen Populationen wider.
Kernaussagen
- AUC 0,91 (95 %-KI 0,87-0,92)
- Extrem hohe Heterogenität (I2>99 %)
- Nur 6 Studien mit externer Validierung
Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.
Kurz gefasst
Metaanalyse in der Zeitschrift Journal of Medical Internet Research, online erschienen am 19.01.2026. Verfasst von Lu Liu und 5 weiteren Autor:innen. Beteiligte Einrichtung: School of Public Health, China Medical University, Shenyang City, Liaoning Province, China. Hauptthemen laut MEDLINE: Machine Learning, Pre-Eclampsia. Die Arbeit wurde bisher 12-mal zitiert (Crossref) und nennt 46 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar.
Abstract (Originalsprache)
Background
Preeclampsia is a severe hypertensive disorder with rising global prevalence. While machine learning (ML) models for predicting preeclampsia are increasingly published, existing evidence shows high heterogeneity, and the distinction between internal performance and external transferability remains unclear.
Objective
This study aims to evaluate the performance of ML models in predicting preeclampsia through a systematic review and meta-analysis, while also exploring their potential clinical application value, in order to specifically enhance the quality of future research and the predictive capability of the models.
Methods
Following PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines and PROSPERO registration, we searched PubMed, Web of Science, IEEE Xplore, and CNKI (China National Knowledge Infrastructure) for studies published through February 2025. We included studies using ML to predict preeclampsia in pregnant women. Bias was assessed using PROBAST (Prediction model Risk of Bias Assessment Tool). We calculated summary estimates using random-effects models and, crucially, computed 95% prediction intervals (PIs) to estimate performance in future clinical settings. Subgroup and meta-regression analyses were conducted to explore heterogeneity.
Results
In total, 26 studies comprising 31 ML models were included. While the pooled area under the receiver operating characteristic curve was high at 0.91 (95% CI 0.87-0.92), extreme heterogeneity was observed (I2>99%). The 95% PI for sensitivity was wide (0.32-0.96), indicating that in some external settings, sensitivity could drop to 32%. Only 6 studies conducted external validation; in these, the pooled sensitivity decreased to 0.68, with a PI of 0.25-0.94.Subgroup analysis suggested that models incorporating laboratory biomarkers and neural networks outperformed others, though CIs overlapped.
Conclusions
Current evidence suggests that a high area under the curve in ML models is more likely to reflect the “performance” of the model on the internal development dataset rather than its universal “effectiveness” and clinical utility in independent, diverse populations. The apparent performance exhibits significant contextual dependence. Future studies should conduct multicenter, prospective external validation and recalibration research to enhance transferability and reliability. Trial Registration PROSPERO CRD420251005830;https://www.crd.york.ac.uk/PROSPERO/view/CRD420251005830
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- School of Public Health, China Medical University, Shenyang City, Liaoning Province, China.
- Department of the Obstetrics and Gynecology, Shengjing Hospital Affiliated to China Medical University, Shenyang City, Liaoning Province China, China.
- School of Engineering, University of Pennsylvania, Philadelphia, PA, United States.
- Department of the Obstetrics and Gynecology, Shengjing Hospital Affiliated to China Medical University, Shenyang City, Liaoning Province, China.
ORCID-Profile: Lu Liu, Qixuan Zhu, Yichi Zong, Xueyuan Chen, Wei Zhang, Jun Wang
Zitieren
APA
Liu, L., Zhu, Q., Zong, Y., Chen, X., Zhang, W., & Wang, J. (2026). Machine Learning Prediction Models for Preeclampsia: Systematic Review and Meta-Analysis. Journal of Medical Internet Research, 28, e78714. https://doi.org/10.2196/78714
Vancouver
Liu L, Zhu Q, Zong Y, Chen X, Zhang W, Wang J. Machine Learning Prediction Models for Preeclampsia: Systematic Review and Meta-Analysis. Journal of Medical Internet Research. 2026;28:e78714. doi:10.2196/78714
BibTeX
@article{Liu2026,
author = {Liu, Lu and Zhu, Qixuan and Zong, Yichi and Chen, Xueyuan and Zhang, Wei and Wang, Jun},
title = {Machine Learning Prediction Models for Preeclampsia: Systematic Review and Meta-Analysis},
journal = {Journal of Medical Internet Research},
year = {2026},
volume = {28},
pages = {e78714},
doi = {10.2196/78714}
}
EUVIMED