AI Estimation of Gestational Age from Blind Ultrasound Sweeps in Low-Resource Settings

Abstract

Background: Ultrasound is indispensable to gestational age estimation and thus to quality obstetrical care, yet high equipment cost and the need for trained sonographers limit its use in low-resource settings.

Methods: From September 2018 through June 2021, we recruited 4695 pregnant volunteers in North Carolina and Zambia and obtained blind ultrasound sweeps (cineloop videos) of the gravid abdomen alongside standard fetal biometry. We trained a neural network to estimate gestational age from the sweeps and, in three test data sets, assessed the performance of the artificial intelligence (AI) model and biometry against previously established gestational age.

Results: In our main test set, the mean absolute error (MAE) (±SE) was 3.9±0.12 days for the model versus 4.7±0.15 days for biometry (difference, -0.8 days; 95% confidence interval [CI], -1.1 to -0.5; P<0.001). The results were similar in North Carolina (difference, -0.6 days; 95% CI, -0.9 to -0.2) and Zambia (-1.0 days; 95% CI, -1.5 to -0.5). Findings were supported in the test set of women who conceived by in vitro fertilization (MAE of 2.8±0.28 vs. 3.6±0.53 days for the model vs. biometry; difference, -0.8 days; 95% CI, -1.7 to 0.2) and in the set of women from whom sweeps were collected by untrained users with low-cost, battery-powered devices (MAE of 4.9±0.29 vs. 5.4±0.28 days for the model vs. biometry; difference, -0.6; 95% CI, -1.3 to 0.1).

Conclusions: When provided blindly obtained ultrasound sweeps of the gravid abdomen, our AI model estimated gestational age with accuracy similar to that of trained sonographers conducting standard fetal biometry. Model performance appears to extend to blind sweeps collected by untrained providers in Zambia using low-cost devices. (Funded by the Bill and Melinda Gates Foundation.).

Figure 1.. Study Flow Chart.

After applying participant and visit-level exclusions, we created 2 training sets to develop and tune the deep learning model and 3 test sets to assess its performance. To be eligible for inclusion in a test set a participant must have at least one study with both blind sweeps and sonographer-acquired biometry available and have their “ground truth” gestational age (GA) established by a prior scan or in vitro fertilization (IVF). The IVF test set comprises all participants who conceived by IVF. The novice test set comprises all participants in whom at least one study visit included sweeps collection by a novice user on a low-cost device. (There were 8 such novices; all were nurse midwives.) The main test set was selected at random from among all remaining eligible participants Some participants apportioned to the test sets had contributed more than one study scan; in such cases we selected a single study scan at random. The training sets comprise all participants who remain after creation of the test sets and were split randomly, by participant, in a 4:1 ratio, into a main training set and a tuning set.

Figure 2.. Model Versus Biometry in the Main Test Set and IVF Test Set.

^a solid line indicates y = x, ^b dashed horizontal lines represent expected error bounds of ultrasound biometry according to the American College of Obstetricians and Gynecologists. In Zambia, “ground truth” gestational age is defined by the first ultrasound. In North Carolina it is defined by an algorithm incorporating both the last menstrual period and the first ultrasound (main test set) or by the known fertilization date (IVF test set). GA denotes gestational age and IVF in vitro fertilization.

Figure 3.. Model Versus Biometry and LMP in the Novice Test Set.

^a dashed horizontal lines represent expected error bounds of ultrasound biometry according to the American College of Obstetricians and Gynecologists. ^b data missing from 22 participants who could not recall their LMP ^c 13 studies from GA by LMP excluded from the plot because the absolute error is truncated at 49 days In Zambia, “ground truth” gestational age is defined by the first ultrasound. GA denotes gestational age and LMP last menstrual period.