Diagnostic Accuracy of an Integrated AI Tool to Estimate Gestational Age From Blind Ultrasound Sweeps

Abstract

Importance: Accurate assessment of gestational age (GA) is essential to good pregnancy care but often requires ultrasonography, which may not be available in low-resource settings. This study developed a deep learning artificial intelligence (AI) model to estimate GA from blind ultrasonography sweeps and incorporated it into the software of a low-cost, battery-powered device.

Objective: To evaluate GA estimation accuracy of an AI-enabled ultrasonography tool when used by novice users with no prior training in sonography.

Design, setting, and participants: This prospective diagnostic accuracy study enrolled 400 individuals with viable, single, nonanomalous, first-trimester pregnancies in Lusaka, Zambia, and Chapel Hill, North Carolina. Credentialed sonographers established the "ground truth" GA via transvaginal crown-rump length measurement. At random follow-up visits throughout gestation, including a primary evaluation window from 14 0/7 weeks' to 27 6/7 weeks' gestation, novice users obtained blind sweeps of the maternal abdomen using the AI-enabled device (index test) and credentialed sonographers performed fetal biometry with a high-specification machine (study standard).

Main outcomes and measures: The primary outcome was the mean absolute error (MAE) of the index test and study standard, which was calculated by comparing each method's estimate to the previously established GA and considered equivalent if the difference fell within a prespecified margin of ±2 days.

Results: In the primary evaluation window, the AI-enabled device met criteria for equivalence to the study standard, with an MAE (SE) of 3.2 (0.1) days vs 3.0 (0.1) days (difference, 0.2 days [95% CI, -0.1 to 0.5]). Additionally, the percentage of assessments within 7 days of the ground truth GA was comparable (90.7% for the index test vs 92.5% for the study standard). Performance was consistent in prespecified subgroups, including the Zambia and North Carolina cohorts and those with high body mass index.

Conclusions and relevance: Between 14 and 27 weeks' gestation, novice users with no prior training in ultrasonography estimated GA as accurately with the low-cost, point-of-care AI tool as credentialed sonographers performing standard biometry on high-specification machines. These findings have immediate implications for obstetrical care in low-resource settings, advancing the World Health Organization goal of ultrasonography estimation of GA for all pregnant people.

Trial registration: ClinicalTrials.gov Identifier: NCT05433519.

Figure 1.. Participant Flow

NC indicates North Carolina. ^aMost candidates in the “other” category were excluded because of unreliable transportation and concern by the study team that they may not be able to attend all scheduled visits. There were 2 individuals excluded for clinical reasons in North Carolina. One person had a colostomy that might interfere with the novice sweep procedure and another had large uterine myomas. ^bThe index test did not produce a gestational age (GA) estimate for 1 participant during the primary evaluation window, thus diagnostic accuracy is calculated among 399 individuals. There were no instances of failure to calculate in the secondary and tertiary windows. The reference standard produced a result for all evaluated patients in all 3 windows.

Figure 2.. Performance of Index Test vs Study Reference Standard in the Primary Evaluation Window

^aDashed horizontal lines represent expected error bounds of ultrasound biometry according to the American College of Obstetricians and Gynecologists.

Figure 3.. Performance of Index Test vs Symphysis-Fundal Height and Last Menstrual Period in the Primary Evaluation Window

^aDashed horizontal lines represent expected error bounds of ultrasound biometry according to the American College of Obstetricians and Gynecologists. ^bTwenty-three participants could not recall their last menstrual period and were excluded. ^cThree participants (2 with last menstrual period error >49 days and 1 with last menstrual period error ≤49 days) and are not represented on this plot. ^dFour participants had a nonpalpable uterine fundus and were excluded.