Research Appraisalother

Artificial Intelligence in Obstetrics: Current Applications, Opportunities, and Clinical Implementation Challenges

Clin Obstet GynecolDeuster, Eileen, Khalil, Asma1 Mar 2026DOI

Clinical Snapshot

25CEBM

Evidence: Weakother

PICO Framework

P — Population	Pregnant women and obstetric patients across various clinical scenarios
I — Intervention	Artificial intelligence applications (deep learning algorithms, automated diagnostic tools)
C — Comparator	Standard obstetric care and experienced clinician assessment
O — Outcomes	Diagnostic accuracy, clinical implementation feasibility, pregnancy complication prediction, fetal surveillance outcomes

Bottom Line

This narrative review examines AI applications in obstetrics but lacks the systematic methodology required for evidence-based clinical recommendations. While the authors report that deep learning algorithms achieve diagnostic accuracy comparable to experienced clinicians, the absence of quantitative synthesis, quality assessment, and systematic search strategy significantly limits the reliability of these conclusions. The review appropriately identifies critical implementation barriers including limited external validation and algorithmic bias. For Australian obstetricians, any AI tools would require rigorous validation studies, TGA approval, and careful consideration of equity issues before clinical adoption. The current evidence base appears insufficient to support routine clinical implementation, and clinicians should await more robust systematic reviews and randomised controlled trials before considering practice changes.

Evidence: Weak

Key Findings

P Value: Not reported
Effect Size: Comparable to experienced clinicians (specific measures not provided)
Primary Outcome: Diagnostic accuracy of AI applications in obstetrics
Nnt Or Sensitivity: Diagnostic accuracy metrics not specified in abstract
Confidence Interval: Not reported

Clinical Application

Significant implementation barriers identified including technical, ethical, and regulatory challenges Would require TGA approval for medical devices, integration with existing ultrasound systems, and validation in Australian obstetric populations before clinical adoption Limited applicability due to validation gaps and algorithmic bias concerns

Abstract

Artificial intelligence is transforming obstetric practice through applications in diagnostic imaging, risk prediction, and clinical decision-making. Deep learning algorithms have achieved diagnostic accuracy comparable to that of experienced clinicians. However, gaps persist between algorithmic capability and clinical implementation. Critical challenges include limited external validation and algorithmic bias. This review examines current AI applications in obstetrics across multiple clinical domains: automated fetal biometry, structural anomaly detection, prediction of pregnancy complications, and intrapartum fetal surveillance. It highlights persistent technical, ethical, and implementation barriers. Key recommendations include multicenter validation across diverse populations, explainable AI approaches, and creating strong regulatory frameworks.

References

1.Deuster, E., & Khalil, A. (2026). Artificial intelligence in obstetrics: Current applications, opportunities, and clinical implementation challenges. Clinical Obstetrics and Gynecology. https://www.cochranelibrary.com/cdsr/doi/10.1002/14651858.CD006066.pub3/full

Share:X LinkedIn

Related Research

observational

40CEBM

Acta obstetricia et gynecologica Scandinavica

ChatGPT in urogynecology: Comparing large language model responses to human experts

3 Aug 2026

Systematic Review

15CEBM

Ultrasound in medicine & biology

A Systematic Review: The Application of Attention Mechanisms in Medical Ultrasound Image Processing

2 Aug 2026