Canonical text (the exact bytes fingerprinted)

TRUVACE RECORD VERSION
record: TRV-2026-2154
version: 1
kind: certified
reason: Certified into the record
timestamp: 2026-10-22T10:41:45.493698Z
status: published
lens: p_space
sector: health
headline: AI-Powered CURA Application for Identifying At-Risk Pregnancies in Obstetrics Management (CURAte): protocol for a randomised controlled trial
dek: Background: Early identification of high-risk pregnancies is a cornerstone of obstetric care, yet antenatal risk assessment in routine practice remains imprecise and highly variable. Evidence from contemporary cohort studies suggests that only a minority of pregnancies that subsequently experience adverse maternal or fetal/neonatal outcomes are identified as high risk in the first trimester, with reported detection rates in the range of 20%–30%. Despite rapid advances in machine-learning (ML) methods for clinica…
gain_title: (none)
problem_title: Routine antenatal risk assessment is imprecise: only about 20%–30% of pregnancies with later adverse outcomes are flagged as high risk in the first trimester, and no ML approach has been tested against standard care in an RCT.
trace_subject: (none)
gain_reading: (none)
gain_evidence: (none)
problem_reading: Routine antenatal risk assessment is imprecise: only about 20%–30% of pregnancies with later adverse outcomes are flagged as high risk in the first trimester, and no ML approach has been tested against standard care in an RCT.
problem_evidence: only a minority of pregnancies that subsequently experience adverse maternal or fetal/neonatal outcomes are identified as high risk in the first trimester | no ML-based approach for general early antenatal risk categorisation has yet been evaluated against standard care within a randomised controlled trial
quick_read: The BMJ Open article is a protocol for the CURAte trial. It will compare an AI-assisted tool that sorts pregnancies into high-risk or low-risk groups against standard antenatal care with no AI categorisation shown to clinicians. The authors say existing first-trimester risk identification catches only a minority of pregnancies that go on to have adverse outcomes.

The work matters because it would be the first randomised test of this kind of machine-learning approach in general early antenatal risk categorisation, according to the authors. No results exist yet. The stated hypotheses are that the tool will improve categorisation accuracy and reduce the composite rate of adverse maternal and fetal or neonatal outcomes, and these remain untested.
limitation: This is a protocol only, so no outcome data exist; the hypothesised gains in accuracy and reduced adverse outcomes are untested, and the tool is intended only as decision support.
tag: Evidence-backed problem
key_points: The trial will randomise 1700 participants 1:1 using block randomisation, stratified by ethnicity and parity. | The AI assessment uses maternal sociodemographic, clinical and obstetrical data at a first-trimester visit and again between 31 and 33 weeks' gestation. | Clinicians see 'high-risk' or 'low-risk' in the intervention arm and 'undisclosed' in the control arm, then follow a predefined management protocol. | The primary composite outcome is taken from medical records 6 weeks after pregnancy ends and analysed on an intention-to-treat basis.
rundown: Participants are blinded to allocation, while attending clinicians see the AI category. For pregnancies categorised as high-risk, clinicians work through predefined referral criteria for the maternal-fetal medicine-led high-risk clinic and complete specified investigations and interventions before referral.

Ethical approval came from the National Healthcare Group Domain Specific Review Board, and written informed consent will be obtained from all participants. The trial is registered as NCT06974188, and there is no interim analysis of the primary outcome.

The authors describe the tool as clinical decision support that does not replace clinician judgement. Subgroup analyses will look for differing effects across key demographic and obstetric groups.
sources:
- peer_reviewed | BMJ Open | https://doi.org/10.1136/bmjopen-2026-124115 | 2026-10-20
prev: 0000000000000000000000000000000000000000000000000000000000000000

sha256

89152b0eda93c2c7b692cbf4f396247f7a833e4207dca1d6aab632ba1f91cb8f

previous

0000000000000000000000000000000000000000000000000000000000000000