The role of artificial intelligence in predicting late-onset intrauterine growth restriction in singleton pregnancies: a systematic review and meta-analysis
- 1Clínica Universidad de Navarra, Pamplona, Spain
- 2Clínica Universidad de Navarra, Madrid, Spain
- 3DatAI, Universidad de Navarra, Pamplona, Spain
Abstract
Purpose Late-onset intrauterine growth restriction (IUGR) is difficult to detect and is associated with adverse perinatal outcomes. This systematic review and meta-analysis evaluated the performance of artificial intelligence (AI) models, including machine learning (ML) and deep learning, for predicting late-onset IUGR in singleton pregnancies. Methods PubMed, Scopus, and Web of Science were searched for studies published during the past decade that used AI to predict IUGR. Nineteen studies including 13,379 pregnancies were analyzed. Sensitivity, specificity, likelihood ratios, and diagnostic odds ratios were pooled using DerSimonian-Laird random-effects models; the area under the summary receiver operating characteristic curve (AUC) was estimated in Stata. Results In the primary analysis (one model per study, k=17), pooled sensitivity was 0.83 (95% confidence interval [CI], 0.74 to 0.89), specificity was 0.89 (95% CI, 0.85 to 0.91), the positive and negative likelihood ratios were 6.6 (95% CI, 5.3 to 8.1) and 0.22 (95% CI, 0.16 to 0.30), and the AUC was 0.93 (95% CI, 0.91 to 0.95). Estimates were unchanged when all model entries were included (k=51). No threshold effect or funnel-plot asymmetry was detected. Subgroup estimates for support vector machines, random forests, and logistic regression were similar but reflected indirect comparisons. Conclusion AI/ML models showed high pooled discriminative accuracy for late-onset IUGR. However, the evidence derives mainly from retrospective, single-center case-control studies without external validation; calibration was reported in six of 19 studies and decision-curve analysis in none. These estimates are an optimistic upper bound, not evidence of superiority over current ultrasound practice.
Related
- Melatonin Supplementation During Middle Gestation Alleviates Intrauterine Growth Restriction in Piglets and Is Associated with Alterations in Placental DNA Methylation
- Predicting follow-up cognition using estimated structural and functional connectivity networks and artificial intelligence in multiple sclerosis
- Predicting Pathological Dental Wear: A Roadmap for Artificial Intelligence in Natural Teeth and Restorations
- Artificial intelligence for predicting chronicity, relapse, and zoonotic risk in brucellosis: toward a one health closed-loop decision-making framework
- Artificial Intelligence in Delirium Care: Predicting, Detecting, and Preventing Delirium in Older Hospitalized Adults.
BibTeX
@article{miranda2026role,
title = {{The role of artificial intelligence in predicting late-onset intrauterine growth restriction in singleton pregnancies: a systematic review and meta-analysis}},
author = {Miriam Turiel Miranda and Alba Etxeandia Atienza and José Nicolás Marín Santos and Lucía Castillo Domínguez and Jesús María Monreal Marquiegui and Laura Emilia Muñoz Saa and Edgar Ricardo Benítez Sastoque and Jesús Fernando López Fidalgo and José Ángel Mínguez Milio and Juan Luis Alcázar Zambrano},
journal = {Ultrasonography},
year = {2026},
month = oct,
doi = {10.14366/usg.25212},
url = {https://doi.org/10.14366/usg.25212},
note = {CC BY-NC},
}
RIS
TY - JOUR
TI - The role of artificial intelligence in predicting late-onset intrauterine growth restriction in singleton pregnancies: a systematic review and meta-analysis
AU - Miriam Turiel Miranda
AU - Alba Etxeandia Atienza
AU - José Nicolás Marín Santos
AU - Lucía Castillo Domínguez
AU - Jesús María Monreal Marquiegui
AU - Laura Emilia Muñoz Saa
AU - Edgar Ricardo Benítez Sastoque
AU - Jesús Fernando López Fidalgo
AU - José Ángel Mínguez Milio
AU - Juan Luis Alcázar Zambrano
PY - 2026
DA - 2026/10/31
JO - Ultrasonography
DO - 10.14366/usg.25212
UR - https://doi.org/10.14366/usg.25212
AB - Purpose Late-onset intrauterine growth restriction (IUGR) is difficult to detect and is associated with adverse perinatal outcomes. This systematic review and meta-analysis evaluated the performance of artificial intelligence (AI) models, including machine learning (ML) and deep learning, for predicting late-onset IUGR in singleton pregnancies. Methods PubMed, Scopus, and Web of Science were searched for studies published during the past decade that used AI to predict IUGR. Nineteen studies including 13,379 pregnancies were analyzed. Sensitivity, specificity, likelihood ratios, and diagnostic odds ratios were pooled using DerSimonian-Laird random-effects models; the area under the summary receiver operating characteristic curve (AUC) was estimated in Stata. Results In the primary analysis (one model per study, k=17), pooled sensitivity was 0.83 (95% confidence interval [CI], 0.74 to 0.89), specificity was 0.89 (95% CI, 0.85 to 0.91), the positive and negative likelihood ratios were 6.6 (95% CI, 5.3 to 8.1) and 0.22 (95% CI, 0.16 to 0.30), and the AUC was 0.93 (95% CI, 0.91 to 0.95). Estimates were unchanged when all model entries were included (k=51). No threshold effect or funnel-plot asymmetry was detected. Subgroup estimates for support vector machines, random forests, and logistic regression were similar but reflected indirect comparisons. Conclusion AI/ML models showed high pooled discriminative accuracy for late-onset IUGR. However, the evidence derives mainly from retrospective, single-center case-control studies without external validation; calibration was reported in six of 19 studies and decision-curve analysis in none. These estimates are an optimistic upper bound, not evidence of superiority over current ultrasound practice.
SN - 2288-5943
SN - 2288-5919
ER -