Evaluating AI-Assisted Blind-Sweep Ultrasound for Pregnancy Screening by Trained Non-Specialist Health Workers in Rural Democratic Republic of Congo

What this Trial Is About

Researchers are evaluating the accuracy and feasibility of an AI-assisted blind-sweep obstetric ultrasound called SPAQ E-con AI, used by trained non-specialist health workers for antenatal screening in rural Democratic Republic of the Congo. This multi-center prospective study includes about 1,430 pregnant women and aims to compare AI measurements of gestational age and fetal conditions to manual measurements by expert readers. The study also includes an adaptive model update plan following regulatory guidance. Participants undergo ultrasound scans using a smartphone-based 9-sweep technique operated by trained non-specialists after short training. The AI estimates gestational age, fetal presentation, and placenta location. The study includes natural enrollment of antenatal care attendees and enriched groups for placenta praevia, fetal malpresentation, and early pregnancy cases. Data collection spans up to 12 months, with interim analyses and adaptive updates planned. During the study, participants receive routine antenatal care while undergoing ultrasound screening. Researchers measure gestational age accuracy in second and third trimesters, AI confidence calibration, and secondary outcomes like sensitivity for placenta praevia, inter-rater reliability, and acceptability of the intervention. They also track protocol adherence, screening coverage, and sustainability of AI use. The study lasts through completion up to June 2027, with ongoing data collection and monitoring.

AI Blind-Sweep Ultrasound for Antenatal Screening by Non-Specialist Health Workers in Rural DR Congo