Antenatal Care, Diagnostic Imaging, Gestational Age, Maternal Health, Placenta Praevia
Conditions
Keywords
blind sweep ultrasound, artificial intelligence, point-of-care ultrasound, task-shifting, DRC
Brief summary
FS2 evaluates the diagnostic accuracy and implementation feasibility of an AI-assisted blind-sweep obstetric ultrasound (SPAQ E-con AI), operated by trained non-specialist health workers, for antenatal screening in rural Democratic Republic of the Congo. Primary outcomes are gestational age mean absolute error (Trimester 2 and Trimester 3) with 95% confidence intervals and AI confidence calibration. The reference standard is manual measurement by a reference reader (early ultrasound first; manual BPD if unavailable; last menstrual period is not used). Planned enrollment is approximately 1,380: Batch 1 (closed, 80 enrolled) and Batch 2 (approximately 1,300), within an IRB-approved ceiling of 3,000. Early termination is permitted upon achievement of pre-specified analysis-plan thresholds. The study is a multi-center prospective Hybrid Type 1 Effectiveness-Implementation design and includes a pre-specified adaptive model-update (Batch 2 cut) plan following FDA PCCP and STARD-AI guidance.
Detailed description
Study conduct is organised in Batches. Batch 1 (Centre de Sante CBCO, Kenge health zone, Kwango province) enrolled 80 participants, of whom 78 had complete datasets, and is closed. Batch 2 is planned at approximately 1,300 participants. The confirmatory analysis cut is approximately 130 to 160 cases acquired with a model frozen before collection, comprising a core set of 80 to 100, an early-gestation enrichment of 25 to 30, and an upper-gestational-age enrichment of 25 to 30. Enrolment may stop early once the pre-specified confidence-interval criteria for the two primary gestational-age outcomes are met. The registered enrolment ceiling of 3,000 is unchanged. Sites are facilities of the Kenge health zone, Kwango province, within the perimeter already approved in the protocol (up to 6 facilities plus 2 to 5 village outreach sites). Eligibility follows the protocol: pregnant women aged 18 years or older. Multiple pregnancies are eligible. Two sub-cohorts address the bounds of the gestational-age range. The early sub-cohort (crown-rump length 7 to 84 mm, approximately under 14 weeks) draws on the same source population as Cohort B-2 and is recruited through community health worker (RECO) village outreach and neighbouring health facilities; it is reported descriptively. The upper-bound sub-cohort (35 weeks or more) is a separate validation subset, recruited in parallel and analysed separately; sensitivity and specificity are reported, and positive and negative predictive values are not reported because prevalence in this subset is artificial. The registered Trimester 3 estimation window (28 to 36 weeks) is unchanged. In Batch 2, point estimates are reported for 28 to 34 weeks, and 35 to 36 weeks is handled as "to be confirmed / refer". This is an operational reporting restriction, not a change to the registered window. The AI model is identified and frozen before each Batch and recorded in a model and application version registry under a pre-determined change control plan. Results are reported by frozen version. AI inference is not used for clinical decision-making.
Interventions
DIAGNOSTIC_TESTSPAQ E-con AI blind-sweep ultrasound
Smartphone-based 9-sweep obstetric ultrasound with AI estimation of gestational age, fetal presentation, and placenta location, operated by trained non-specialist health workers after 30-60 minutes of training. The AI model is frozen before each Batch and managed in a version registry under a pre-determined change control plan (PCCP); AI inference is not used for clinical decision-making.
Sponsors
SOIK Corporation Sarl
Study design
Observational model
COHORT
Time perspective
PROSPECTIVE
Eligibility
Sex/Gender
FEMALE
Age
18 Years to No maximum
Healthy volunteers
Yes
Inclusion criteria
* Pregnant women aged 18 years or older * Identified at a participating facility routine ANC visit or RECO village outreach within the Kenge health zone catchment * Written informed consent obtained
Exclusion criteria
* Emergency presentation * Duplicate re-registration of an already-enrolled woman * Lacking capacity to consent * Planned relocation outside Kwango province during the study period * Refusal of consent
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Gestational age mean absolute error (MAE), Trimester 2 (14-27 weeks) | Through study completion, up to 12 months | MAE in days vs reference standard, with 95% CI; target MAE ≤7 days and upper 95% CI ≤10 days |
| Gestational age mean absolute error (MAE), Trimester 3 (28-36 weeks) | Through study completion, up to 12 months | MAE in days vs reference standard, with 95% CI; target MAE ≤10 days and upper 95% CI ≤14 days |
| AI confidence calibration - Expected Calibration Error (ECE) | At Batch 1 closure, up to 8 weeks | Expected Calibration Error from the reliability diagram of AI confidence versus reference-standard agreement; target \<=0.05 |
| AI confidence calibration - Brier score | At Batch 1 closure, up to 8 weeks | Brier score of AI confidence versus reference-standard agreement; target \<=0.20 |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Trimester 1 gestational age MAE (CRL and GS models) | Through study completion, up to 12 months | Integrated Cohort A T1 + Cohort B-2, N=30; CRL model MAE ≤5 days or GS model MAE ≤7 days |
| Placenta praevia and fetal malpresentation sensitivity and specificity | Through study completion, up to 12 months | Observed sensitivity/specificity with 95% CI (Clopper-Pearson) on enriched validation subset (Cohort B-1, target N=20-25); reported as observed values |
| Inter-rater reliability of gestational age - Intraclass Correlation Coefficient (ICC) | During the inter-rater assessment window, up to 1 week | ICC(2,1) between two independent readers for gestational age in days |
| Inter-rater reliability of placenta praevia classification - Cohen's kappa | During the inter-rater assessment window, up to 1 week | Cohen's kappa for placenta praevia (yes/no) between two independent readers |
| Inter-rater reliability of fetal presentation classification - weighted kappa | During the inter-rater assessment window, up to 1 week | Linear weighted kappa for fetal presentation category between two independent readers |
| Acceptability - Acceptability of Intervention Measure (AIM) score | Through study completion, up to 12 months | Mean AIM score (4-item, 5-point Likert) among health workers |
| Feasibility - ultrasound throughput (scans per device per day) | Through study completion, up to 12 months | Number of scans completed per device per day; target \>=15 |
| Fidelity - protocol adherence rate (5-item checklist) | Through study completion, up to 12 months | Percentage adherence on a 5-item fidelity checklist; target \>=80% |
| Penetration - proportion of eligible women screened | Through study completion, up to 12 months | Percentage of eligible women in the catchment screened with AI ultrasound |
| Sustainability - intention to continue use | At study completion, up to 12 months | Percentage of staff reporting intention to continue use at the end-of-study interview |
Countries
Democratic Republic of the Congo
Contacts
CONTACTKuniyuki Furuta
PRINCIPAL_INVESTIGATORKuniyuki Furuta
SOIK Corporation
Outcome results
None listed