A mobile-optimized artificial intelligence system for gestational age and fetal malpresentation assessment

Abstract

Background Fetal ultrasound is an important component of antenatal care, but shortage of adequately trained healthcare workers has limited its adoption in low-to-middle-income countries. This study investigated the use of artificial intelligence for fetal ultrasound in under-resourced settings. Methods Blind sweep ultrasounds, consisting of six freehand ultrasound sweeps, were collected by sonographers in the USA and Zambia, and novice operators in Zambia. We developed artificial intelligence (AI) models that used blind sweeps to predict gestational age (GA) and fetal malpresentation. AI GA estimates and standard fetal biometry estimates were compared to a previously established ground truth, and evaluated for difference in absolute error. Fetal malpresentation (non-cephalic vs cephalic) was compared to sonographer assessment. On-device AI model run-times were benchmarked on Android mobile phones. Results Here we show that GA estimation accuracy of the AI model is non-inferior to standard fetal biometry estimates (error difference −1.4 ± 4.5 days, 95% CI −1.8, −0.9, n = 406). Non-inferiority is maintained when blind sweeps are acquired by novice operators performing only two of six sweep motion types. Fetal malpresentation AUC-ROC is 0.977 (95% CI, 0.949, 1.00, n = 613), sonographers and novices have similar AUC-ROC. Software run-times on mobile phones for both diagnostic models are less than 3 s after completion of a sweep. Conclusions The gestational age model is non-inferior to the clinical standard and the fetal malpresentation model has high AUC-ROCs across operators and devices. Our AI models are able to run on-device, without internet connectivity, and provide feedback scores to assist in upleveling the capabilities of lightly trained ultrasound operators in low resource settings.

Development of a Machine Learning Model for Sonographic Assessment of Gestational Age

“…An independent GA and variance score was predicted for each image. Our video model was assembled from an inflated 3-dimensional (I3D) convolutional model 13 and a convolutional recurrent model proposed for blindsweep GA prediction, 10 which used a MobileNetV2 14 feature extractor. The model architecture is depicted in eFigure 3 in Supplement 1.…”

“…We have recently shown that GA model estimation using ultrasonography videos of predefined sweeps was noninferior to standard fetal biometry estimates. 10 In this study, we further extend the use of ultrasonography videos by developing 3 end-to-end AI models: (1) an image model using fetal ultrasonography images captured by sonographers during biometry measurements; (2) a video model using fly-to videos, which are defined as 5 to 10 seconds of video immediately before image capture; and (3) an ensemble model using both images and fly-to videos. All data were collected retrospectively during standard biometry measurements.…”

“…Recent studies show state-of-art accuracy for fetal biometry estimation, where a classification model is used to identify standard ultrasonography planes and a segmentation model is used for fetal biometry estimation from standard ultrasonography planes, and those works focus on fetal biometry estimation to match sonographer performance on biometry measurement, whereas our work focuses on GA estimation without requiring measurement acquisition or use of regression formulas, which gives the opportunity for the model to improve GA estimation accuracy of sonographer assessment. We have recently shown that GA model estimation using ultrasonography videos of predefined sweeps was noninferior to standard fetal biometry estimates …”

Artificial intelligence as a teaching tool for gynaecological ultrasound: A systematic search and scoping review

“…This then allows for the acquisition of ultrasound data from which basic measurements can be performed, all with little or no formal training in ultrasound scanning. 9,10 Some systems are also able to provide feedback to operators as to the quality of the image they are obtaining and, in some cases, provide feedback to improve the image in real time. [10][11][12] Whilst none of these systems replace supervised scanning by becoming as proficient as a diagnostic ultrasound specialist, it is likely that AI holds potential to play a future role in the teaching of certain skills required to perform TVUS.…”

“…9,10 Some systems are also able to provide feedback to operators as to the quality of the image they are obtaining and, in some cases, provide feedback to improve the image in real time. [10][11][12] Whilst none of these systems replace supervised scanning by becoming as proficient as a diagnostic ultrasound specialist, it is likely that AI holds potential to play a future role in the teaching of certain skills required to perform TVUS. If such tools could be utilised, this may have the potential to close some of the skill gaps related to TVUS and expand access to higher quality ultrasound.…”