CoChat
Sign Up
Cited by 18Body Composition Measurement TechniquesMedical Image Segmentation TechniquesMedical Imaging and Analysis

Fully automated waist circumference measurement on abdominal CT: Comparison with manual measurements and potential value for identifying overweight and obesity as an adjunct output of CT scan

Open access
Joo, Ijin, Kwak, Min-Sun, Park, Dae Hyun, Yoon, Soon Ho·2021·PLOS ONE
DOI 10.1371/journal.pone.0254704PMC PMC8289071
Open this paper in CoChat — free accountOpen PDF

Abstract

Objective: Waist circumference (WC) is a widely accepted anthropometric parameter of central obesity. We investigated a fully automated body segmentation algorithm for measuring WC on abdominal computed tomography (CT) in comparison to manual WC measurements (WC-manual) and evaluated the performance of CT-measured WC for identifying overweight/obesity. Materials and methods: This retrospective study included consecutive adults who underwent both abdominal CT scans and manual WC measurements at a health check-up between January 2013 and November 2019. Mid-waist WCs were automatically measured on noncontrast axial CT images using a deep learning-based body segmentation algorithm. The associations between CT-measured WC and WC-manual was assessed by Pearson correlation analysis and their agreement was assessed through Bland-Altman analysis. The performance of these WC measurements for identifying overweight/obesity (i.e., body mass index [BMI] ≥25 kg/m 2 ) was evaluated using receiver operating characteristics (ROC) curve analysis. Results: Among 763 subjects whose abdominal CT scans were analyzed using a fully automated body segmentation algorithm, CT-measured WCs were successfully obtained in 757 adults (326 women; mean age, 54.3 years; 64 women and 182 men with overweight/obesity). CT-measured WC was strongly correlated with WC-manual (r = 0.919, p < 0.001), and showed a mean difference of 6.1 cm with limits of agreement between -1.8 cm and 14.0 cm in comparison to WC-manual. For identifying overweight/obesity, CT-measured WC showed excellent performance, with areas under the ROC curve (AUCs) of 0.960 (95% CI, 0.933–0.979) in women and 0.909 (95% CI, 0.878–0.935) in men, which were comparable to WC-manual (AUCs of 0.965 [95% CI, 0.938–0.982] and 0.916 [95% CI, 0.886–0.941]; p = 0.735 and 0.437, respectively). Conclusion: CT-measured WC using a fully automated body segmentation algorithm was closely correlated with manually-measured WC. While radiation issue may limit its general use, it can serve as an adjunctive output of abdominal CT scans to identify overweight/obesity.

AI snapshot

Written from the paper’s full text.

TL;DR

A fully automated deep learning body segmentation algorithm measured waist circumference (WC) on noncontrast abdominal CT with 99.2% technical efficacy in 757 adults. CT-measured WC correlated strongly with manual WC and identified overweight/obesity (BMI ≥25 kg/m²) comparably to manual WC, though with a systematic difference limiting direct interchangeability.

Key findings

  • CT-measured WC was successfully obtained in 757/763 subjects (99.2%; 99.5% among those with sufficient FOV) and correlated strongly with manual WC: r = 0.919 (p < 0.001), ICC = 0.954.
  • Agreement was limited by a mean difference of 6.1 cm (SD 4.0 cm; limits of agreement −1.8 to 14.0 cm) versus manual WC.
  • For identifying overweight/obesity (BMI ≥25 kg/m²), CT-measured WC had AUCs of 0.960 (95% CI 0.933–0.979) in women and 0.909 (95% CI 0.878–0.935) in men, comparable to manual WC (0.965 and 0.916; p = 0.735 and 0.437).
  • Optimal CT-measured WC cut-offs were 91.7 cm in women and 96.7 cm in men, with sensitivity/specificity of 87.5%/92.7% and 83.5%/83.9%, respectively.

Methods

  • Retrospective single-center study of consecutive adults who underwent abdominal CT and manual WC measurement at health check-ups between January 2013 and November 2019; exclusions were age <18 years or interval >4 weeks between manual WC/BMI and CT.
  • Noncontrast axial CT images were analyzed with a commercially available deep learning body segmentation program (modified 3D U-Net; DeepCatch) to automatically identify the mid-waist level and extract the body perimeter as CT-measured WC; manual WC was measured by trained examiners using the WHO protocol.
  • Pearson correlation, Bland-Altman analysis, ICC, ROC/AUC comparison, Youden index for cut-offs, and McNemar test were used.

Limitations

  • Stated: retrospective, single-center health check-up population; generalizability needs testing in other populations, CT machines, and scan protocols.
  • Stated: performance was evaluated for overweight/obesity but not obesity or severe obesity because fewer than 5% of the sample had obesity; metabolic conditions and clinical outcomes were not evaluated.
  • Evident from the text: radiation exposure may limit general use of CT for WC measurement, and the systematic difference between CT-measured and manual WC means the methods are not directly interchangeable without conversion.

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

Integrity signals

Registry

Crossref · 10.1371/journal.pone.0254704

Cites

22 of 22 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

15 of 25 cited sources checked · 9 supported · 0 flagged · accuracy not yet measured

  • Interpretationreasoning

    The systematic difference between CT-measured waist circumference and manually-measured waist circumference can be explained by differences in patient position, breath control, and measurement principle.

    “This discrepancy can be explained by differences in patient position (supine versus standing), breath control (expiration versus inspiration), and the measurement principle (perimeter including skin surface irregularities versus tape measurement).”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [20] A Waninge, 2010, Measuring waist circumference in disabled adults, Res Dev Disabil source

      The Discussion reports a systematic difference between standing and supine waist measurements, supporting patient position as a relevant distinction. The Measurements sections say the study used a tape measure and averaged readings taken during inhalation and exhalation. The paper does not compare CT with manual measurements or explain a discrepancy through breath control or skin-surface irregularities. (LLM, conf=0.86)

      Experimental — accuracy not yet measured
  • Interpretationreasoning

    Serially-obtained CT-measured waist circumference may also provide prognostic information in obese patients, although further validation is required.

    “serially-obtained CT-measured WC may also provide prognostic information in obese patients, although further validation is required.”
    Quote in the paperNot a substantial claim2 of 2 sources checked
    • cites [21] AE Rothberg, 2017, Impact of weight loss on waist circumference and the components of the metabolic syndrome, BMJ Open Diabetes Res Care source

      The Conclusions (page unavailable) say that monitoring changes in waist circumference during weight-loss interventions may provide prognostic information about metabolic risk in obese patients. However, the Research design and methods section (page unavailable) says waist circumference was measured with a nylon tape, not CT; the paper does not validate the claim for CT-measured waist circumference. (LLM, conf=0.90)

      Experimental — accuracy not yet measured
    • cites [22] AA Mulligan, 2019, Changes in waist circumference and risk of all-cause and CVD mortality: results from the European Prospective Investigation into Cancer in Norfolk (EPIC-Norfolk) cohort study, BMC Cardiovasc Disord source

      The Conclusion (page unavailable) reports that increases in waist circumference over approximately four years were associated with later mortality, supporting the possibility that repeated measurements have prognostic value. But the Methods say nurses measured waist circumference with a tape, not CT, and the conclusion concerns a population-based cohort of middle-aged and elderly adults rather than obese patients specifically. The paper does not establish the claim's CT-specific application or its need for further validation. (LLM, conf=0.89)

      Experimental — accuracy not yet measured
  • Premisecitation

    Waist circumference provides independent or additive value to body mass index for predicting obesity-related health risks.

    “but also provides independent or additive value to BMI for predicting obesity-related health risks [7–10].”
    Quote in the paperSubstantial claim3 of 4 sources checked
    • cites [7] EJ Jacobs, 2010, Waist circumference and all-cause mortality in a large US cohort, Arch Intern Med source
    • cites [8] YH Kim, 2019, Waist Circumference and All-Cause Mortality Independent of Body Mass Index in Korean Population from the National Health Insurance Health Checkup 2009(-)2015, J Clin Med. source
    • cites [9] JR Cerhan, 2014, A pooled analysis of waist circumference and mortality in 650,000 adults, Mayo Clin Proc source
    • cites [10] C Zhang, 2008, Abdominal obesity and the risk of all-cause, cardiovascular, and cancer mortality: sixteen years of follow-up in US women, Circulation source
  • Premisecitation

    Body mass index does not take into account the distribution of body fat, limiting its representation of abdominal adiposity.

    “since BMI does not take into account the distribution of body fat, it has limitations in representing abdominal adiposity, which is an important aspect of body composition that would be linked to metabolic diseases [4].”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [4] CM Lee, 2008, Indices of abdominal obesity are better discriminators of cardiovascular risk factors than BMI: a meta-analysis, J Clin Epidemiol source
  • Findingown data

    In men, the area under the ROC curve of CT-measured waist circumference was not significantly different from that of manually-measured waist circumference (p = 0.437).

    “p = 0.735 and 0.437, respectively”
    Quote in the paperSubstantial claim
  • Classified as probably not a substantial claim. 9
  • May combine two claims. 14
  • The quote only partly states this claim. 1

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

    Premisecitation

    Waist circumference is a widely accepted anthropometric parameter of central obesity and a globally accepted screening tool for central obesity.

    “Waist circumference (WC) is a widely accepted anthropometric parameter of central obesity.”
    • Also stated in: § Introduction · plain
      “WC is globally accepted as a screening tool for central obesity [5].”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [5] R Ross, 2020, Waist circumference as a vital sign in clinical practice: a Consensus Statement from the IAS and ICCR Working Group on Visceral Obesity, Nat Rev Endocrinol source

  • May combine two claims.

    Interpretationcitation

    CT can provide free additional information regarding waist circumference, a well-established anthropometric parameter of central obesity and a diagnostic criterion for metabolic syndrome.

    “CT can provide a free additional information regarding WC, a well-established anthropometric parameter of central obesity and a diagnostic criterion for metabolic syndrome [5, 19].”
    Quote in the paperSubstantial claim1 of 2 sources checked
    • cites [5] R Ross, 2020, Waist circumference as a vital sign in clinical practice: a Consensus Statement from the IAS and ICCR Working Group on Visceral Obesity, Nat Rev Endocrinol source

      The Abstract describes waist circumference as a simple measure of abdominal adiposity, and the Measurement of waist circumference section says it can screen for metabolic syndrome. The paper does not establish in these passages that waist circumference is a diagnostic criterion for the syndrome, or that CT provides a free, additional waist-circumference measurement; its measurement discussion concerns direct and self-measurement. (LLM, conf=0.85)

      Experimental — accuracy not yet measured
    • cites [19] KG Alberti, 2005, Group IDFETFC. The metabolic syndrome—a new worldwide definition, Lancet source
    • Rests on an assumption: High volume CT studies are performed in health check-ups and disease monitoring.

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

    Premisecitation

    Waist circumference is closely correlated with body mass index.

    “WC is not only closely correlated with BMI [6]”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [6] A Sonmez, 2013, Waist circumference cutoff points to predict obesity, metabolic syndrome, and cardiovascular risk in Turkish adults, Int J Endocrinol source

      The Methods and Results describe waist-circumference cutoffs used to predict overweight and obesity, supporting a relationship with BMI-based categories. Our copy does not report a correlation between continuous WC and BMI values or establish how close that correlation is, so it does not fully support “closely correlated.” (LLM, conf=0.78)

      Experimental — accuracy not yet measured

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

    Premisecitation

    Waist circumference is associated with all-cause or cardiovascular mortality.

    “growing evidence suggests that WC is associated with all-cause [7–9] or cardiovascular mortality [10].”
    Quote in the paperSubstantial claim3 of 4 sources checked
    • cites [7] EJ Jacobs, 2010, Waist circumference and all-cause mortality in a large US cohort, Arch Intern Med source
    • cites [8] YH Kim, 2019, Waist Circumference and All-Cause Mortality Independent of Body Mass Index in Korean Population from the National Health Insurance Health Checkup 2009(-)2015, J Clin Med. source
    • cites [9] JR Cerhan, 2014, A pooled analysis of waist circumference and mortality in 650,000 adults, Mayo Clin Proc source
    • cites [10] C Zhang, 2008, Abdominal obesity and the risk of all-cause, cardiovascular, and cancer mortality: sixteen years of follow-up in US women, Circulation source

  • May combine two claims.

    Premisecitation

    Previous studies tested the feasibility of obtaining waist circumference measurements from CT images by drawing a region of interest or by step-by-step image analysis, but those methods require time-consuming human intervention.

    “Several previous studies [12, 13] have tested the feasibility of obtaining WC measurements from CT images by drawing a region of interest or by step-by-step image analysis; however, those methods require time-consuming human intervention.”
    Quote in the paperSubstantial claim1 of 2 sources checked
    • cites [12] SL Gomez-Perez, 2016, Measuring Abdominal Circumference and Skeletal Muscle From a Single Cross-Sectional Computed Tomography Image: A Step-by-Step Guide for Clinicians Using National Institutes of Health ImageJ, JPEN J Parenter Enteral Nutr. source

      The Background describes CT-derived abdominal circumference as a proxy for waist circumference and identifies analysis time and training as barriers. The Methods give a step-by-step ImageJ procedure requiring a user to select and measure the abdominal perimeter. This paper is a tutorial rather than a study testing feasibility, and its waist-circumference step uses a tracing tool rather than manual drawing of a region of interest. (LLM, conf=0.84)

      Experimental — accuracy not yet measured
    • cites [13] A Ciudin, 2014, Measurement of waist circumference for retrospective studies—prospective validation of use of CT images to assess abdominal circumference, Endocrinol Nutr. source

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

    Premisecitation

    Deep learning-based algorithms have increasingly been investigated for automated segmentation of body composition from medical imaging.

    “In recent years, deep learning-based algorithms have increasingly been investigated for automated segmentation of body composition from medical imaging [14, 15].”
    Quote in the paperNot a substantial claim1 of 2 sources checked
    • cites [14] AD Weston, 2019, Automated Abdominal Segmentation of CT Scans for Body Composition Analysis Using Deep Learning, Radiology source
    • cites [15] Y Wang, 2017, A two-step convolutional neural network based computer-aided detection scheme for automatically segmenting adipose tissue volume depicting on CT images, Comput Methods Programs Biomed source

  • The quote only partly states this claim.

    Findingown data

    For identifying overweight/obesity (BMI ≥25 kg/m2), CT-measured waist circumference showed an area under the ROC curve of 0.960 (95% CI 0.933-0.979) in women.

    “with AUCs of 0.960 in women”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Both the CT-measured waist circumference cut-off and the manually-measured waist circumference cut-off for overweight/obesity did not result in any false positives in underweight subjects or any false negatives in obesity subjects.

    “both the CT-measured WC cut-off and WC-manual cut-off for overweight/obesity did not result in any false positives in underweight subjects or any false negatives in obesity subjects.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The optimal cut-off value of CT-measured waist circumference for overweight/obesity in women was ≥91.7 cm, with sensitivity 87.5% and specificity 92.7%.

    “the optimal cut-off values for CT-measured WC for overweight/obesity in women and men were ≥91.7 cm and ≥96.7 cm, respectively, resulting in sensitivity and specificity values of 87.5% (56/64) and 92.7% (243/262); and 83.5% (152/182) and 83.9% (209/249), respectively.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The optimal cut-off value of CT-measured waist circumference for overweight/obesity in men was ≥96.7 cm, with sensitivity 83.5% and specificity 83.9%.

    “the optimal cut-off values for CT-measured WC for overweight/obesity in women and men were ≥91.7 cm and ≥96.7 cm, respectively, resulting in sensitivity and specificity values of 87.5% (56/64) and 92.7% (243/262); and 83.5% (152/182) and 83.9% (209/249), respectively.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The optimal cut-offs for CT-measured waist circumference were 91.7 cm in women and 96.7 cm in men, which were 6.7 cm higher than the KSSO criteria for manually-measured waist circumference in both women and men.

    “the optimal cut-offs for CT-measured WC extracted from our study sample were 91.7 cm in women and 96.7 cm in men, which were substantially higher (by 6.7 cm in both women and men) than the KSSO criteria of WC-manual for Koreans (85 cm in women and 90 cm in men) [18].”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [18] SY Lee, 2007, Appropriate waist circumference cutoff points for central obesity in Korean adults, Diabetes Res Clin Pract source

  • May combine two claims.

    Interpretationown data

    While radiation issue may limit its general use, CT-measured waist circumference can serve as an adjunctive output of abdominal CT scans to identify overweight/obesity.

    “it can serve as an adjunctive output of abdominal CT scans to identify overweight/obesity.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: Radiation from CT may limit the general use of CT to measure waist circumference.
    • Rests on an assumption: The midpoint between the lowest rib margin and the highest iliac crest margin on CT corresponds to the manual waist circumference measurement location.

  • May combine two claims.

    Premisecitation

    Manual waist circumference measurement is prone to the risk of person-to-person spread of infectious diseases and is considered intimate in some cultures.

    “since WC usually involves a manual procedure performed by medical personnel, it is prone to the risk of person-to-person spread of infectious diseases, such as coronavirus disease 2019, and is considered intimate in some cultures [11].”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [11] AJ Dunkley, 2009, Waist circumference measurement: knowledge, attitudes and barriers in patients and practitioners in a multi-ethnic population, Fam Pract source

  • Classified as probably not a substantial claim.

    Study design and conductcitation

    Manual waist circumference measurements were performed by well-trained examiners according to the World Health Organization STEPwise approach to surveillance protocol.

    “WC was measured by placing an inelastic tape measure horizontally at the level of the mid‐point between the lower costal margin and the iliac crest at the end of a quiet expiration by well‐trained examiners according to the World Health Organization STEPwise approach to surveillance protocol [16].”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [16] Organization WH. Waist circumference and waist-hip ratio: report of a WHO expert consultation, Geneva, 8–11 December 2008. 2011.

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

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

    Study design and conductnone given

    CT measurement of waist circumference was considered technically successful if the z-axis range of the waist was correctly identified and all segmented waist circumferences in the range fit the abdominal skin surface well.

    “The CT measurement of WC for each subject was considered technically successful if the z-axis range of the waist (from the level of the lowest rib margin to the highest iliac crest margin) was correctly identified and all segmented WCs in the range fit the abdominal skin surface well.”
    Quote in the paperSubstantial claim

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

    Premisecitation

    Body mass index is widely used to define overweight and obesity and is a risk factor for cardiovascular disease and type 2 diabetes.

    “Body mass index (BMI), a value derived from body weight and height, is widely used to define overweight and obesity and is a risk factor for cardiovascular disease and type 2 diabetes [2, 3].”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [2] PW Wilson, 2002, Overweight and obesity as determinants of cardiovascular risk: the Framingham experience, Arch Intern Med source
    • cites [3] KM Narayan, 2007, Effect of BMI on lifetime risk for diabetes in the U.S, Diabetes Care source

  • Classified as probably not a substantial claim.

    Study design and conductown data

    The technical efficacy of waist circumference measurements on CT was assessed by a radiologist blinded to clinical parameters and CT-measured waist circumference values.

    “The technical efficacy of WC measurements on CT was assessed by a visual assessment of automatically-generated body segmentation results by a radiologist (I.J. with 12 years of experience in abdominal imaging) who was blinded to the clinical parameters and CT-measured WC values of each subject.”
    Quote in the paperSubstantial claim

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

    Study design and conductnone given

    The study was a retrospective study of subjects who received routine health check-ups at a single center.

    “First, this was a retrospective study of subjects who received routine health check-ups at a single center.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Study design and conductown data

    The study evaluated the performance of waist circumference for overweight/obesity but not for obesity or severe obesity because only small numbers of subjects with obesity (less than 5% of the study sample) were included.

    “Second, we evaluated the performance of WC for overweight/obesity, but not for obesity or severe obesity because only small numbers of subjects with obesity (less than 5% of the study sample) were included.”
    Quote in the paperSubstantial claim
19 of 19

  • The paper relies on this without support.

    Radiation from CT may limit the general use of CT to measure waist circumference.

    “While radiation issue may limit the general use of CT to measure WC”

  • The paper relies on this without support.

    High volume CT studies are performed in health check-ups and disease monitoring.

    “given the high volume CT studies performed in a health check-up as well as disease monitoring”

  • The paper relies on this without support.

    The midpoint between the lowest rib margin and the highest iliac crest margin on CT corresponds to the manual waist circumference measurement location.

    “This anatomical location was selected to reflect the principle of manual WC measurements used in this study.”
3 of 3

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

  • Men's limits of agreement do not contain the reported mean difference

    Major

    Limits of agreement must be centered on the mean difference. The midpoint of the reported limits is 0.5 cm, not 6.8 cm; the upper limit appears to be missing a digit.

    Reported 6.8 · recomputed 0.5

  • Overweight/obesity count conflicts with sex-specific counts

    Minor

    The reported 64 women and 182 men total 246, not 242. The stated 32.5% of 757 also corresponds to 246 subjects.

    Reported 242 · recomputed 246

2 of 2

Cites

Flagged first, then by citations

Cited by

Still being checked.

Related

Authors
Joo, I., Kwak, M.-S., Park, D. H., & Yoon, S. H.
Affiliations
  • Seoul National University Hospital (KR)
  • Seoul National University (KR)
Corresponding author
Yoon, Soon Ho
ORCID
Venue
PLOS ONE
Publisher
Public Library of Science (PLoS)
Published
2021-07-19
Volume
16
Issue
7
Pages
e0254704
ISSN
1932-6203 (ISSN-L)
DOI
10.1371/journal.pone.0254704
PMC
PMC8289071
OpenAlex
W3185123299
Cited by
18
References
22
Citations by year
2026: 2 · 2025: 4 · 2024: 3 · 2023: 4 · 2022: 3 · 2021: 1
FWCI
1.44
Type
journal article
Language
en
Open access
Licence
Locations
  • PLoS ONE (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • PubMed Central (submittedVersion, cc-by)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
medical ip co. ltd
Topics
  • Health Sciences › Medicine › Physiology › Body Composition Measurement Techniques
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Medical Image Segmentation Techniques
  • Physical Sciences › Engineering › Biomedical Engineering › Medical Imaging and Analysis
Keywords
waist circumference, deep learning, abdominal CT, abdominal obesity, overweight, obesity screening, anthropometry, body mass index, measurement agreement, radiation dose reduction
MeSH terms
Abdomen, Adult, Aged, Aged, 80 and over, Anthropometry, Female, Humans, Male, Middle Aged, Obesity, Body Mass Index, Overweight, Waist Circumference, Young Adult, Four-Dimensional Computed Tomography

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 2026-10-20.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-20.

Cited sources are checked by gpt-6-sol.

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