The DOI matches its Crossref record; the references it cites are not checked yet.
Automated Fetal Head Detection and Circumference Estimation from Free-Hand Ultrasound Sweeps Using Deep Learning in Resource-Limited Countries
Open accessAbstract
Ultrasound imaging remains out of reach for most pregnant women in developing countries because it requires a trained sonographer to acquire and interpret the images. We address this problem by presenting a system that can automatically estimate the fetal head circumference (HC) from data obtained with use of the obstetric sweep protocol (OSP). The OSP consists of multiple pre-defined sweeps with the ultrasound transducer over the abdomen of the pregnant woman. The OSP can be taught within a day to any health care worker without prior knowledge of ultrasound. An experienced sonographer acquired both the standard plane-to obtain the reference HC-and the OSP from 183 pregnant women in St. Luke's Hospital, Wolisso, Ethiopia. The OSP data, which will most likely not contain the standard plane, was used to automatically estimate HC using two fully convolutional neural networks. First, a VGG-Net-inspired network was trained to automatically detect the frames that contained the fetal head. Second, a U-net-inspired network was trained to automatically measure the HC for all frames in which the first network detected a fetal head. The HC was estimated from these frame measurements, and the curve of Hadlock was used to determine gestational age (GA). The results indicated that most automatically estimated GAs fell within the P2.5-P97.5 interval of the Hadlock curve compared with the GAs obtained from the reference HC, so it is possible to automatically estimate GA from OSP data. Our method therefore has potential application for providing maternal care in resource-constrained countries.
Overview
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
Verification
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?
A limitation of AI applications for fetal biometry is slow processing times.
“slow processing times [35,43]”
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 [35]
The “Deployment on low-cost hardware” section says frame classification is fast enough for real-time computation on the study’s training hardware, rather than identifying slow AI processing as a limitation. The “Improvements” section identifies a different delay: the ultrasound device took one minute to save each sweep. (LLM, conf=0.84)
Checked by a reasoning model — may be wrong.
Experimental — accuracy not yet measuredVan den Heuvel (2019) found that it is possible to automatically estimate gestational age, determine fetal presentation and detect twin pregnancies using the same acquisition protocol.
“Van den Heuvel (2019) found that it is possible to automatically estimate gestational age, determine fetal presentation and detect twin pregnancies using the same acquisition protocol.”
From Automatic Placenta Localization From Ultrasound Imaging in a Resource-Limited Setting Using a Predefined Ultrasound Acquisition Protocol and Deep Learning (2022) · cites this paper as Van den Heuvel (2019)
The Gestational age estimation section reports that gestational age can be estimated automatically from head-circumference measurements using the obstetric sweep protocol. The Data section says twins and two fetuses in shoulder presentation were excluded from the head-circumference evaluation; it does not report automatic detection of twin pregnancies or determination of fetal presentation. (LLM, conf=0.90)
Experimental — accuracy not yet measuredAI usage in standard plane acquisition has the benefit of applicability for medically underserved areas.
“applicability for medically underserved areas [17,35]”
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 [35]
The Introduction says the automated system was designed for deployment on a low-cost laptop or tablet to facilitate use in developing countries. But the paper does not demonstrate AI-assisted standard-plane acquisition: it estimates fetal head circumference from free-hand sweeps, and the Improvements section says an experienced sonographer was required to ensure the standard plane was recorded in this study. Page numbers are not provided in our copy. (LLM, conf=0.83)
Experimental — accuracy not yet measuredA limitation of AI applications for fetal biometry is lack of training for algorithms with pathological cases.
“lack of training for algorithms with pathological cases [28,35]”
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 [35]
The Improvements section (page unavailable) warns that pathology could cause the system to measure fetal head circumference incorrectly. It does not say that this risk results from a lack of training on pathological cases. The Data section describes excluding an anencephalic fetus from test-set evaluation, not from training. (LLM, conf=0.78)
Experimental — accuracy not yet measuredAI usage in standard plane acquisition has the benefit of feasibility of biometric assessment by non-experts.
“the feasibility of biometric assessment by non-experts [35,39,43]”
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 [35]
The Introduction says free-hand sweeps can be taught to health care workers without ultrasound experience and that automated head-circumference estimation could remove the need for a trained sonographer. But the Improvements section says an experienced sonographer acquired the study data and that assessment using data acquired by untrained midwives remains for a future study. The paper therefore supports the potential for non-expert assessment, not its demonstrated feasibility in non-expert hands. (LLM, conf=0.85)
Experimental — accuracy not yet measured
AI is of public health relevance by reducing costs and the need for sonography experts.
“It, therefore, is of public health relevance, by reducing costs and the need for sonography experts [35,39].”
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 [35]
The Introduction (page unavailable) says the proposed system could remove the need for a trained sonographer to acquire and interpret images for fetal head-circumference measurement. The Deployment on low-cost hardware section (page unavailable) says low-cost implementation could make the system affordable in developing countries. The paper presents these as potential benefits, not demonstrated reductions in costs or sonographer use. (LLM, conf=0.86)
Experimental — accuracy not yet measuredBecause most studies use data from healthy subjects or healthy fetuses, miscalculation or misdiagnosis may occur in case of pathologies.
“As most of the studies are performed with data obtained from healthy subjects or healthy fetuses, miscalculation or misdiagnosis may occur in case of pathologies [35,95,145].”
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 [35]
The Improvements section (page unavailable) says pathology could cause the system to measure fetal head circumference incorrectly. The Data section (page unavailable) also reports excluding anencephalic fetuses from parts of the analysis. The paper does not establish that most studies use healthy subjects or fetuses, or specifically discuss misdiagnosis. (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.
Citations
The citation lists haven’t been fetched yet
References 57 · Cited by 91
Research works printed but not linked: 1
- The Theano Development Team. Theano: A Python framework for fast computation of mathematical expressions. arXiv preprint arXiv:1605.02688, 2016. · arXiv
Metadata
- Authors
- van den Heuvel, T. L. A., Petros, H., Santini, S., de Korte, C. L., & van Ginneken, B.
- Venue
- Ultrasound in Medicine & Biology
- Publisher
- Elsevier BV
- Published
- 2019-03
- Volume
- 45
- Issue
- 3
- Pages
- 773-785
- ISSN
- 0301-5629 (ISSN-L)
- OpenAlex
- W2904169708
- Cited by
- 91
- References
- 57
- Citations by year
- 2026: 6 · 2025: 15 · 2024: 9 · 2023: 11 · 2022: 19 · 2021: 15 · 2020: 11 · 2019: 5
- FWCI
- 3.87
- Type
- journal article
- Language
- en
- Open access
- bronze
- doi.org
- Locations
- Ultrasound in Medicine & Biology (publishedVersion)
- PubMed (publishedVersion)
- University of Twente Research Information (publishedVersion)
- Publikationsdatenbank der Fraunhofer-Gesellschaft (Fraunhofer-Gesellschaft) (submittedVersion)
- Fraunhofer-Publica (Fraunhofer-Gesellschaft) (submittedVersion)
- Radboud Repository (Radboud University) (submittedVersion)
- Radboud Repository (Radboud University) (submittedVersion, other-oa)
- University of Twente Research Information (submittedVersion)
- Funders
- Life Sciences & Health for Development Fund (LSH14ET04)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Medical Image Segmentation Techniques
- Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
- Health Sciences › Medicine › Pediatrics, Perinatology and Child Health › Prenatal Screening and Diagnostics
- Keywords
- gestational age estimation, Hadlock formula, fully convolutional network, U-Net, VGG, resource-limited settings, maternal care, prenatal ultrasound
- MeSH terms
- Deep Learning, Adult, Developing Countries, Female, Head, Humans, Pregnancy, Reproducibility of Results, Ultrasonography, Prenatal
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
