Development of a Machine Learning Model for Sonographic Assessment of Gestational Age
Open accessAbstract
Importance: Fetal ultrasonography is essential for confirmation of gestational age (GA), and accurate GA assessment is important for providing appropriate care throughout pregnancy and for identifying complications, including fetal growth disorders. Derivation of GA from manual fetal biometry measurements (ie, head, abdomen, and femur) is operator dependent and time-consuming. Objective: To develop artificial intelligence (AI) models to estimate GA with higher accuracy and reliability, leveraging standard biometry images and fly-to ultrasonography videos. Design, Setting, and Participants: To improve GA estimates, this diagnostic study used AI to interpret standard plane ultrasonography images and fly-to ultrasonography videos, which are 5- to 10-second videos that can be automatically recorded as part of the standard of care before the still image is captured. Three AI models were developed and validated: (1) an image model using standard plane images, (2) a video model using fly-to videos, and (3) an ensemble model (combining both image and video models). The models were trained and evaluated on data from the Fetal Age Machine Learning Initiative (FAMLI) cohort, which included participants from 2 study sites at Chapel Hill, North Carolina (US), and Lusaka, Zambia. Participants were eligible to be part of this study if they received routine antenatal care at 1 of these sites, were aged 18 years or older, had a viable intrauterine singleton pregnancy, and could provide written consent. They were not eligible if they had known uterine or fetal abnormality, or had any other conditions that would make participation unsafe or complicate interpretation. Data analysis was performed from January to July 2022. Main Outcomes and Measures: The primary analysis outcome for GA was the mean difference in absolute error between the GA model estimate and the clinical standard estimate, with the ground truth GA extrapolated from the initial GA estimated at an initial examination. Results: Of the total cohort of 3842 participants, data were calculated for a test set of 404 participants with a mean (SD) age of 28.8 (5.6) years at enrollment. All models were statistically superior to standard fetal biometry-based GA estimates derived from images captured by expert sonographers. The ensemble model had the lowest mean absolute error compared with the clinical standard fetal biometry (mean [SD] difference, -1.51 [3.96] days; 95% CI, -1.90 to -1.10 days). All 3 models outperformed standard biometry by a more substantial margin on fetuses that were predicted to be small for their GA. Conclusions and Relevance: These findings suggest that AI models have the potential to empower trained operators to estimate GA with higher accuracy.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The estimation of GA is one of the important indications for obstetric US in early pregnancy that helps to adjust maternity care and identify complications such as prematurity or fetal growth disorders.
“The estimation of GA is one of the important indications for obstetric US in early pregnancy that helps to adjust maternity care and identify complications such as prematurity or fetal growth disorders [144].”
From Evolving the Era of 5D Ultrasound? A Systematic Literature Review on the Applications for Artificial Intelligence Ultrasound Imaging in Obstetrics and Gynecology (2023) · cites this paper as [144]
The Introduction says fetal ultrasonography provides gestational-age estimates and evaluation for fetal growth disorders to guide maternal-fetal care. The Discussion links accurate gestational-age estimation to clinical decisions, but emphasizes later pregnancy; it mentions prematurity as a contributor to low birth weight, not as a complication identified by early-pregnancy ultrasound. The early-pregnancy and prematurity portions of the claim are therefore not supported. (LLM, conf=0.86)
Experimental — accuracy not yet measuredA benefit of GA AI models was the possibility for application in low-resource countries.
“The benefits of the GA AI models were the possibility for application in low-resource countries [144,145,148,149,152]”
From Evolving the Era of 5D Ultrasound? A Systematic Literature Review on the Applications for Artificial Intelligence Ultrasound Imaging in Obstetrics and Gynecology (2023) · cites this paper as [144]
The Results (AI GA Estimation Using Ultrasonography Images and Fly-To Videos; page unavailable) report that the models performed well across the US and Zambia, and the Conclusions describe potential incorporation into routine clinical workflow. This supports possible application in Zambia, but the paper does not establish applicability across low-resource countries generally; its Limitations note that broader validation is needed. (LLM, conf=0.81)
Experimental — accuracy not yet measuredThe purpose of the FAMLI studies was the successful establishment of an AI algorithm for GA estimation from simplified blind US sweeps of US novices in low-resource countries.
“The purpose of these studies, which were based on US data from the US and Zambia, was the successful establishment of an AI algorithm for GA estimation from simplified blind US sweeps of US novices in low-resource countries [144,145,152,153].”
From Evolving the Era of 5D Ultrasound? A Systematic Literature Review on the Applications for Artificial Intelligence Ultrasound Imaging in Obstetrics and Gynecology (2023) · cites this paper as [144]
The Methods (Algorithm Development; page not provided) describe FAMLI data from the US and Zambia and AI models for gestational-age estimation; the video training data included blind sweeps. But the paper says the scans were acquired by trained sonographers, novice studies were excluded, and its video model predicted from fly-to videos. It does not establish the claimed model based on sweeps by novices in low-resource countries. (LLM, conf=0.89)
Experimental — accuracy not yet measuredAn important limitation of GA AI models was application in very early stages of pregnancy.
“An important limitation of the AI models was described to be application in very early [144,147]”
From Evolving the Era of 5D Ultrasound? A Systematic Literature Review on the Applications for Artificial Intelligence Ultrasound Imaging in Obstetrics and Gynecology (2023) · cites this paper as [144]
The Discussion > Limitations section (page unavailable) identifies a small first-trimester sample and says more first-trimester examinations are needed to confirm the findings. That supports a limitation in the evidence for early pregnancy, but not a demonstrated inability to apply the models at very early stages. (LLM, conf=0.86)
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
MobileNetV2: Inverted Residuals and Linear Bottlenecks
Sandler, M., Howard, A., Zhu, M., Zhmoginov, A., & Chen, L.-C. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00474 · Cited by 28,755 · see citing works
Quo Vadis, Action Recognition? A New Model and the Kinetics Dataset
Carreira, J., & Zisserman, A. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.502 · Cited by 9,570 · see citing works
4 million neonatal deaths: When? Where? Why?
Lawn, J. E., Cousens, S., & Zupan, J. (2005) · The Lancet · doi:10.1016/s0140-6736(05)71048-5 · Cited by 3,670 · see citing works
Simple and Scalable Predictive Uncertainty Estimation using Deep Ensembles
Lakshminarayanan, B., Pritzel, A., & Blundell, C. (2016) · arXiv · doi:10.48550/arxiv.1612.01474 · Cited by 3,666 · see citing works
Estimation of fetal weight with the use of head, body, and femur measurements—A prospective study
Hadlock, F. P., Harrist, R. B., Sharman, R. S., Deter, R. L., & Park, S. K. (1985) · American Journal of Obstetrics and Gynecology · doi:10.1016/0002-9378(85)90298-4 · Cited by 2,682 · see citing works
Estimating the mean and variance of the target probability distribution
Nix, D. A., & Weigend, A. S. (1994) · doi:10.1109/icnn.1994.374138 · Cited by 720 · see citing works
Racial/ethnic standards for fetal growth: the NICHD Fetal Growth Studies
Buck Louis, G. M., Grewal, J., Albert, P. S., Sciscione, A., Wing, D. A., Grobman, W. A., Newman, R. B., Wapner, R., D’Alton, M. E., Skupski, D., et al. (2015) · American Journal of Obstetrics and Gynecology · doi:10.1016/j.ajog.2015.08.032 · Cited by 532 · see citing works
AI Estimation of Gestational Age from Blind Ultrasound Sweeps in Low-Resource Settings
Pokaprakarn, T., Prieto, J. C., Price, J. T., Kasaro, M. P., Sindano, N., Shah, H. R., Peterson, M., Akapelwa, M. M., Kapilya, F. M., Sebastião, Y. V., et al. (2022) · NEJM Evidence · doi:10.1056/evidoa2100058 · Cited by 80 · see citing works
Use of femur length/abdominal circumference ratio in detecting the macrosomic fetus.
Hadlock, F. P., Harrist, R. B., Fearneyhough, T. C., et al. (1985) · Radiology · doi:10.1148/radiology.154.2.3880915 · Cited by 69 · see citing works
Kim, B., Kim, K. C., Park, Y., Kwon, J.-Y., Jang, J., & Seo, J. K. (2018) · Physiological Measurement · doi:10.1088/1361-6579/aae255 · Cited by 62 · see citing works
Genetic and Environmental Influences on Fetal Growth Vary during Sensitive Periods in Pregnancy
Workalemahu, T., Grantz, K. L., Grewal, J., Zhang, C., Louis, G. M. B., & Ayele, F. T. (2018) · Scientific Reports · doi:10.1038/s41598-018-25706-z · Cited by 61 · see citing works
Automatic evaluation of fetal head biometry from ultrasound images using machine learning
Kim, H. P., Lee, S. M., Kwon, J.-Y., Park, Y., Kim, K. C., & Seo, J. K. (2019) · Physiological Measurement · doi:10.1088/1361-6579/ab21ac · Cited by 61 · see citing works
Ravishankar, H., Prabhu, S. M., Vaidya, V., & Singhal, N. (2016) · IEEE International Symposium on Biomedical Imaging (ISBI) · doi:10.1109/isbi.2016.7493382 · Cited by 58 · see citing works
Deep learning fetal ultrasound video model match human observers in biometric measurements
Płotka, S., Klasa, A., Lisowska, A., Seliga-Siwecka, J., Lipa, M., Trzciński, T., & Sitek, A. (2022) · Physics in Medicine & Biology · doi:10.1088/1361-6560/ac4d85 · Cited by 57 · see citing works
Källén, K. (2004) · Ultrasound in Obstetrics and Gynecology · doi:10.1002/uog.1082 · Cited by 51 · see citing works
Gomes, R. G., Vwalika, B., Lee, C., Willis, A., Sieniek, M., Price, J. T., Chen, C., Kasaro, M. P., Taylor, J. A., Stringer, E. M., et al. (2022) · Communications Medicine · doi:10.1038/s43856-022-00194-5 · Cited by 47 · see citing works
Self-citationFetalNet: Multi-task Deep Learning Framework for Fetal Ultrasound Biometric Measurements
Płotka, S., Wlodarczyk, T. W., Klasa, A., Lipa, M., Sitek, A., & Trzcinski, T. P. (2021) · Communications in computer and information science · doi:10.1007/978-3-030-92310-5_30 · Cited by 41 · see citing works
AutoFB: Automating Fetal Biometry Estimation from Standard Ultrasound Planes
Bano, S., Dromey, B., Vasconcelos, F., Napolitano, R., David, A. L., Peebles, D., & Stoyanov, D. V. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-87234-2_22 · Cited by 39 · see citing works
Morken, N., Klungsøyr, K., & Skjærven, R. (2014) · BMC Pregnancy and Childbirth · doi:10.1186/1471-2393-14-172 · Cited by 38 · see citing works
Deb, S., Mohammed, M. S., Dhingra, U., Dutta, A., Ali, S. M., Dixit, P., Juma, M. H., Hassan, M. J., Sazawal, S., Nisar, M. I., et al. (2020) · The Lancet Global Health · doi:10.1016/s2214-109x(20)30034-6 · Cited by 38 · see citing works
Fetal biometry by an inexperienced operator using two‐ and three‐dimensional ultrasound
Yang, F., Leung, K. Y., Lee, Y. P., Chan, H. Y., & Tang, M. H. Y. (2010) · Ultrasound in Obstetrics and Gynecology · doi:10.1002/uog.7600 · Cited by 34 · see citing works
Automated Fetal Head Classification and Segmentation Using Ultrasound Video
Rasheed, K., Junejo, F., Malik, A. I., & Saqib, M. (2021) · IEEE Access · doi:10.1109/access.2021.3131518 · Cited by 30 · see citing works
Prediction of Large‐for‐Gestational‐Age Neonates by Different Growth Standards
Duncan, J. R., Odibo, L., Hoover, E. A., & Odibo, A. O. (2020) · Journal of Ultrasound in Medicine · doi:10.1002/jum.15470 · Cited by 13 · see citing works
Skelton, E., Matthew, J., Li, Y., Khanal, B., Cerrolaza Martinez, J. J., Toussaint, N., Gupta, C., Knight, C., Kainz, B., Hajnal, J. V., et al. (2021) · Radiography · doi:10.1016/j.radi.2020.11.006 · Cited by 11 · see citing works
Cawyer, C. R., Anderson, S. B., Szychowski, J. M., Skupski, D. W., & Owen, J. (2019) · American Journal of Perinatology · doi:10.1055/s-0039-1681055 · Cited by 3 · see citing works
Flagged first, then by citations
Cited by
Sorted by citations
Jost, E., Kosian, P., Jimenez Cruz, J., Albarqouni, S., Gembruch, U., Strizek, B., & Recker, F. (2023) · Journal of Clinical Medicine · doi:10.3390/jcm12216833 · Cited by 44 · see citing works
“The purpose of these studies, which were based on US data from the US and Zambia, was the successful establishment of an AI algorithm for GA estimation from simplified blind US sweeps of US novices in low-resource countries [144,145,152,153].”
BackgroundInfluential
- Authors
- Lee, C., Willis, A., Chen, C., Sieniek, M., Watters, A., Stetson, B., Uddin, A., Wong, J., Pilgrim, R., Chou, K., Tse, D., Shetty, S., & Gomes, R. G.
- Affiliations
- Google Health, Palo Alto, California
- Northwestern University (US)
- Corresponding author
- Lee, Chace
- ORCID
- Venue
- JAMA Network Open
- Publisher
- American Medical Association (AMA)
- Published
- 2023-01-04
- Volume
- 6
- Issue
- 1
- Pages
- e2248685
- ISSN
- 2574-3805 (ISSN-L)
- PMC
- PMC9857195
- OpenAlex
- W4313482842
- Cited by
- 42
- References
- 25
- Citations by year
- 2026: 8 · 2025: 21 · 2024: 8 · 2023: 5
- FWCI
- 1.67
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- cc-by-nc-nd
- Locations
- JAMA Network Open (publishedVersion, cc-by-nc-nd)
- PubMed Central (publishedVersion, cc-by-nc-nd)
- PubMed (publishedVersion)
- Funders
- Bill and Melinda Gates Foundation (INV-003266, OPP1191684)
- Topics
- Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
- Health Sciences › Medicine › Pediatrics, Perinatology and Child Health › Prenatal Screening and Diagnostics
- Health Sciences › Health Professions › Health Information Management › Artificial Intelligence in Healthcare
- Keywords
- gestational age estimation, prenatal ultrasound, fetal biometry, artificial intelligence, ensemble learning, prenatal care, fetal growth disorders, South America, Latin America, Africa, Asia, Middle East, North America, resource-limited settings
- MeSH terms
- Machine Learning, Artificial Intelligence, Female, Gestational Age, Humans, Pregnancy, Ultrasonography, Zambia, Reproducibility of Results
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
