CoChat
Sign Up
Cited by 36Prenatal Screening and DiagnosticsArtificial Intelligence in Healthcare and EducationCongenital Anomalies and Fetal Surgery

Clinical workflow of sonographers performing fetal anomaly ultrasound scans: Deep-learning-based analysis

Open access
Drukker, L., Sharma, H., Karim, J. N., Droste, R., Noble, J. A., Papageorghiou, A. T.·2022·Ultrasound in Obstetrics & Gynecology
DOI 10.1002/uog.24975PMC PMC10107110
Open this paper in CoChat — free accountOpen PDF

Abstract

Objective: Despite decades of obstetric scanning, the field of sonographer workflow remains largely unexplored. In the second trimester, sonographers use scan guidelines to guide their acquisition of standard planes and structures; however, the scan‐acquisition order is not prescribed. Using deep‐learning‐based video analysis, the aim of this study was to develop a deeper understanding of the clinical workflow undertaken by sonographers during second‐trimester anomaly scans. Methods: We collected prospectively full‐length video recordings of routine second‐trimester anomaly scans. Important scan events in the videos were identified by detecting automatically image freeze and image/clip save. The video immediately preceding and following the important event was extracted and labeled as one of 11 commonly acquired anatomical structures. We developed and used a purposely trained and tested deep‐learning annotation model to label automatically the large number of scan events. Thus, anomaly scans were partitioned as a sequence of anatomical planes or fetal structures obtained over time. Results: A total of 496 anomaly scans performed by 14 sonographers were available for analysis. UK guidelines specify that an image or videoclip of five different anatomical regions must be stored and these were detected in the majority of scans: head/brain was detected in 97.2% of scans, coronal face view (nose/lips) in 86.1%, abdomen in 93.1%, spine in 95.0% and femur in 92.3%. Analyzing the clinical workflow, we observed that sonographers were most likely to begin their scan by capturing the head/brain (in 24.4% of scans), spine (in 23.2%) or thorax/heart (in 22.8%). The most commonly identified two‐structure transitions were: placenta/amniotic fluid to maternal anatomy, occurring in 44.5% of scans; head/brain to coronal face (nose/lips) in 42.7%; abdomen to thorax/heart in 26.1%; and three‐dimensional/four‐dimensional face to sagittal face (profile) in 23.7%. Transitions between three or more consecutive structures in sequence were uncommon (up to 13% of scans). None of the captured anomaly scans shared an entirely identical sequence. Conclusions: We present a novel evaluation of the anomaly scan acquisition process using a deep‐learning‐based analysis of ultrasound video. We note wide variation in the number and sequence of structures obtained during routine second‐trimester anomaly scans. Overall, each anomaly scan was found to be unique in its scanning sequence, suggesting that sonographers take advantage of the fetal position and acquire the standard planes according to their visibility rather than following a strict acquisition order. © 2022 The Authors. Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of International Society of Ultrasound in Obstetrics and Gynecology.

AI snapshot

Written from the paper’s full text.

TL;DR

Using deep-learning-based video analysis of 496 routine second-trimester anomaly scans, this study found wide variation in sonographer workflow, with no two scans sharing an entirely identical acquisition sequence. The findings suggest anomaly scanning is opportunistic, guided by fetal position rather than a fixed order.

Key findings

  • Among 496 scans performed by 14 sonographers, the five UK FASP mandatory structures were detected in most scans: head/brain 97.2%, coronal face (nose/lips) 86.1%, abdomen 93.1%, spine 95.0% and femur 92.3%. At least four of five mandatory structures were detected in 92.9% of scans, and all five in 72.4%.
  • Sonographers most often began by capturing head/brain (24.4%), spine (23.2%) or thorax/heart (22.8%). The most common two-structure transitions were placenta/amniotic fluid to maternal anatomy (44.5%), head/brain to coronal face (42.7%), abdomen to thorax/heart (26.1%) and 3D/4D face to sagittal face (23.7%).
  • Transitions between three or more consecutive structures were uncommon, occurring in up to 13% of scans. None of the captured anomaly scans shared an entirely identical sequence.
  • Mean anomaly scan duration was 32.1 ± 11.4 min, representing approximately 265 h of scan video, with a mean of 39 ± 13 important scan events per scan.

Methods

  • Prospective observational study of routine anomaly scans in singleton pregnancies at 19–21 weeks, conducted May 2018–February 2020 at a single UK maternity unit. Of 518 recruited women, 496 scans were included after 22 exclusions for technical recording problems.
  • Full-length scan videos were recorded via HDMI. Image freeze, image save and clip save events were detected automatically; 5-s clips around each event were extracted and labeled as one of 11 anatomical structures.
  • A deep-learning model was trained using manual labels from 62 full-length scans and then used to label all scan events automatically. Timelines of structure labels were analyzed with descriptive statistics for structure capture, order and transitions.

Limitations

  • The paper states that it could not judge whether specific sequences perform better or result in shorter scanning times, because it collected prospective observational data on routine real-world practice with minimal interference.
  • The study was conducted in a single maternity unit, which may not represent practice at other centers.
  • Automatic detection missed some structures in some scans; this may reflect incomplete scans or misclassification, including structures labeled as mixed. The average maternal BMI was 25, so generalizability to women with higher BMI remains to be established.
  • Evident from the text: the observational design cannot establish causal relationships between workflow patterns and scan quality or efficiency, and sonographers were aware that scans were being recorded, although they were not informed of the current analysis aim.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1002/uog.24975

Cites

34 of 34 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

0 of 6 cited sources checked · 0 supported · 0 flagged

  • Findingown data

    The two-structure transition head/brain to coronal face (nose/lips) occurred in 42.7% of scans.

    “head/brain to coronal face (nose/lips) in 42.7%”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: Image freeze and image/clip save events represent important scan events.
    • Rests on an assumption: The 5-s clips extracted around each important event contain the anatomical structure being assessed.
  • Findingown data

    In 359 (72.4%) scans, all five mandatory structures were detected.

    “in 359 (72.4%) scans, all five mandatory structures were detected”
    Quote in the paperSubstantial claim
    • Rests on an assumption: Image freeze and image/clip save events represent important scan events.
    • Rests on an assumption: The 5-s clips extracted around each important event contain the anatomical structure being assessed.
  • Findingown data

    Sonographers were most likely to begin their scan by capturing head/brain (24.4% of scans).

    “sonographers were most likely to begin their scan by capturing the head/brain (in 24.4% of scans)”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: Image freeze and image/clip save events represent important scan events.
    • Rests on an assumption: The 5-s clips extracted around each important event contain the anatomical structure being assessed.
  • Findingown data

    The three-structure transition placenta/amniotic fluid to maternal anatomy to placenta/amniotic fluid occurred in 53 (10.7%) scans.

    “placenta/amniotic fluid to maternal anatomy to placenta/amniotic fluid in 53 (10.7%)”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: Image freeze and image/clip save events represent important scan events.
    • Rests on an assumption: The 5-s clips extracted around each important event contain the anatomical structure being assessed.
  • Findingreasoning

    A total of 496 second-trimester anomaly scans (one per woman) were included in the analysis.

    “Therefore, 496 second‐trimester anomaly scans (one per woman) were included.”
    Quote in the paperSubstantial claim
  • May combine two claims. 7
  • Classified as probably not a substantial claim. 11
  • The quote only partly states this claim. 2

  • May combine two claims.

    Findingown data

    The structures most commonly captured multiple times were thorax/heart, spine and maternal anatomy, recorded on average seven, five and three times in each scan, respectively.

    “The structures most commonly captured multiple times were thorax/heart, spine and maternal anatomy, recorded on average seven, five and three times in each scan, respectively”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: Image freeze and image/clip save events represent important scan events.
    • Rests on an assumption: The 5-s clips extracted around each important event contain the anatomical structure being assessed.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    22 (4.2%) women were excluded because of technical problems with recordings, such as incomplete videos.

    “Of these, 22 (4.2%) were excluded because of technical problems with recordings, such as incomplete videos.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Findingown data

    The scans were performed by 14 operators, of whom 11 were accredited sonographers and three fetal medicine doctors.

    “Each scan was carried out by one of 14 operators, of whom 11 were accredited sonographers and three fetal medicine doctors”
    Quote in the paperNot a substantial claim

  • The quote only partly states this claim.

    Findingown data

    Sonographers were most likely to begin their scan by capturing thorax/heart (22.8% of scans).

    “thorax/heart (in 22.8%)”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: Image freeze and image/clip save events represent important scan events.
    • Rests on an assumption: The 5-s clips extracted around each important event contain the anatomical structure being assessed.

  • Probably not a substantial claim: may not pass S1 (asserted), S3 (load-bearing). The quote only partly states this claim.

    Findingown data

    Sonographers were most likely to begin their scan by capturing spine (23.2% of scans).

    “spine (in 23.2%)”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: Image freeze and image/clip save events represent important scan events.
    • Rests on an assumption: The 5-s clips extracted around each important event contain the anatomical structure being assessed.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    The mean gestational age at scan was 20.3 ± 0.6 weeks.

    “at a mean ± SD gestational age of 20.3 ± 0.6 weeks”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim. May combine two claims.

    Premisecitation

    Artificial neural networks are particularly good at image pattern recognition and can be utilized to improve the speed and accuracy of diagnosing patient conditions in medical fields that depend heavily on images and video, as in ultrasound.

    “Artificial neural networks are particularly good at image pattern recognition and can be utilized to improve the speed and accuracy of diagnosing patient conditions in medical fields that depend heavily on images and video, as in the case of ultrasound 11 , 12 .”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites [11] Drukker L, Noble JA, Papageorghiou AT. Introduction to artificial intelligence in ultrasound imaging in obstetrics and gynecology. Ultrasound Obstet Gynecol 2020; 56: 498– 505. 32530098 PMC7702141 source

      Not checked in the free run.

    • cites [12] Xie HN, Wang N, He M, Zhang LH, Cai HM, Xian JB, Lin MF, Zheng J, Yang YZ. Using deep‐learning algorithms to classify fetal brain ultrasound images as normal or abnormal. Ultrasound Obstet Gynecol 2020; 56: 579– 587. 31909548 source

      Not checked in the free run.

  • Classified as probably not a substantial claim.

    Premisecitation

    The five mandatory structures required by UK FASP guidelines are head/brain, coronal face view (nose/lips), abdomen, spine and femur.

    “According to the UK FASP guidelines, five mandatory structures should be captured as images or clips: head/brain, coronal face view (nose/lips), abdomen, spine and femur.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [14] Public Health England. Fetal anomaly screening programme: standards. https://www.gov.uk/government/publications/fetal‐anomaly‐screening‐programme‐standards. 36680493

      Not checked in the free run.

  • May combine two claims.

    Study design and conductreasoning

    Sonographers were aware that the scans were being recorded but had not been informed of the aim of the current analysis, making it unlikely that they acted differently from their usual practice.

    “even though sonographers were aware that the scans were being recorded, they had not been informed of the aim of the current analysis, meaning that it is unlikely that they acted differently from their usual practice while participating in this study.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Sonographers should assess, but are not obliged to store an image or clip of, sagittal face, thorax/heart, kidneys, bladder, limbs and placenta/amniotic fluid.

    “sonographers should assess, but are not obliged to store an image or a clip of, the following structures: sagittal face (fetal profile), thorax/heart, kidneys, bladder, limbs and placenta/amniotic fluid 14”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [14] Public Health England. Fetal anomaly screening programme: standards. https://www.gov.uk/government/publications/fetal‐anomaly‐screening‐programme‐standards. 36680493

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    Confusion matrices were used to compare interannotator and manual vs automatic labeling.

    “Confusion matrices were used to compare interannotator and manual vs automatic labeling.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Study design and conductreasoning

    To reduce the risk of bias, the study included 14 sonographers, and the findings were consistent, making external validity more likely.

    “To reduce the risk of bias, we included 14 sonographers, and we noted that the findings were consistent, making external validity more likely.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim. May combine two claims.

    Premisecitation

    Scanning guidelines comprise a checklist of standard planes, structures and anatomies to be surveyed, but do not dictate the order in which the checklist should be populated.

    “Such scanning guidelines comprise typically a checklist of standard planes, structures and anatomies to be surveyed, but do not dictate the order in which the checklist should be populated.”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites [9] Salomon LJ, Alfirevic Z, Berghella V, Bilardo CM, Chalouhi GE, Da Silva Costa F, Hernandez‐Andrade E, Malinger G, Munoz H, Paladini D, Prefumo F, Sotiriadis A, Toi A, Lee W. ISUOG Practice Guidelines (updated): performance of the routine mid‐trimester fetal ultrasound scan. Ultrasound Obstet Gynecol 2022; 59: 840– 856. 35592929 source

      Not checked in the free run.

    • cites [10] AIUM‐ACR‐ACOG‐SMFM‐SRU Practice Parameter for the Performance of Standard Diagnostic Obstetric Ultrasound Examinations. J Ultrasound Med 2018; 37: E13– 24. 30308091 source

      Not checked in the free run.

  • Classified as probably not a substantial claim.

    Study design and conductown data

    The deep-learning model was purposely trained and tested to label automatically the large number of scan events.

    “We developed and used a purposely trained and tested deep‐learning annotation model to label automatically the large number of scan events.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Interpretationnone given

    The study was conducted in a single maternity unit which may not represent practice at other centers.

    “the study was conducted in a single maternity unit which may not represent practice at other centers”
    Quote in the paperNot a substantial claim
15 of 15

  • The paper relies on this without support.

    Image freeze and image/clip save events represent important scan events.

    “we analyzed the full‐length anomaly scans automatically and identified sonographer actions that represent important scan events by detecting the occurrences of image freeze, image save and clip save.”

  • The paper relies on this without support.

    The 5-s clips extracted around each important event contain the anatomical structure being assessed.

    “After identifying the important scan events, we extracted 5‐s clips that occurred around the event (i.e. immediately preceding and following the event).”
2 of 2

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 in workflow analysis applications was the impossibility of the generalization of workflow sequences due to maternal−fetal factors and different skill levels of sonographers for different anatomical regions during a full routine scan.

    “Limitations in application were reported to be the impossibility of the generalization of workflow sequences due to maternal−fetal factors and different skill levels of sonographers for different anatomical regions during a full routine scan [155,158].”

    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 [158]

    The Results and Discussion (page unavailable) report that no two full scans followed an identical sequence and that fetal position and movement influence acquisition order. They do not establish that workflow sequences cannot be generalized because of maternal–fetal factors or differing sonographer skill by anatomical region; the Discussion instead calls for research to establish useful workflow patterns. (LLM, conf=0.87)

    Experimental — accuracy not yet measured

  • Sharma et al. and Drukker et al. analyzed full-length scanning videos to assess workflow by building timeline models illustrating the scanning sequence of anatomical regions over time.

    “Sharma et al. and Drukker et al. analyzed full-length scanning videos to assess workflow by building timeline models illustrating the scanning sequence of anatomical regions over time [155,158].”

    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 [158]

Cites

Flagged first, then by citations

Cited by

Sorted by citations

Related

Authors
Drukker, L., Sharma, H., Karim, J. N., Droste, R., Noble, J. A., & Papageorghiou, A. T.
Affiliations
  • Tel Aviv University (IL)
  • University of Oxford (GB)
  • John Radcliffe Hospital (GB)
Corresponding author
Papageorghiou, A. T.
ORCID
Venue
Ultrasound in Obstetrics & Gynecology
Publisher
Wiley
Published
2022-12
Volume
60
Issue
6
Pages
759-765
ISSN
  • 0960-7692 (ISSN-L)
  • 1469-0705
DOI
10.1002/uog.24975
PMC
PMC10107110
OpenAlex
W4283217943
Cited by
36
References
33
Citations by year
2026: 8 · 2025: 8 · 2024: 11 · 2023: 8 · 2022: 1
FWCI
7.68
Type
journal article
Language
en
Open access
Licence
Locations
  • Ultrasound in Obstetrics and Gynecology (publishedVersion, cc-by-nc)
  • PubMed Central (publishedVersion, cc-by-nc)
  • PubMed (publishedVersion)
  • St George's Online Research Archive (St George's University of London) (acceptedVersion)
Funders
H2020 European Research Council (ERC‐ADG‐2015 694581)
Topics
  • Health Sciences › Medicine › Pediatrics, Perinatology and Child Health › Prenatal Screening and Diagnostics
  • Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
  • Health Sciences › Medicine › Surgery › Congenital Anomalies and Fetal Surgery
Keywords
clinical workflow, ultrasound video analysis, abdominal imaging
MeSH terms
Deep Learning, Female, Fetus, Humans, Pregnancy, Pregnancy Trimester, Second, Ultrasonography, Prenatal, Workflow

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.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-10.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-10.

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