CoChat

Smartphone camera based assessment of adiposity: a validation study

Open access

Published version of a preprint

The DOI matches its Crossref record; 52 of the 52 references it cites were found in a registry or an index.

Abstract

Abstract Body composition is a key component of health in both individuals and populations, and excess adiposity is associated with an increased risk of developing chronic diseases. Body mass index (BMI) and other clinical or commercially available tools for quantifying body fat (BF) such as DXA, MRI, CT, and photonic scanners (3DPS) are often inaccurate, cost prohibitive, or cumbersome to use. The aim of the current study was to evaluate the performance of a novel automated computer vision method, visual body composition (VBC), that uses two-dimensional photographs captured via a conventional smartphone camera to estimate percentage total body fat (%BF). The VBC algorithm is based on a state-of-the-art convolutional neural network (CNN). The hypothesis is that VBC yields better accuracy than other consumer-grade fat measurements devices. 134 healthy adults ranging in age (21–76 years), sex (61.2% women), race (60.4% White; 23.9% Black), and body mass index (BMI, 18.5–51.6 kg/m 2 ) were evaluated at two clinical sites ( N = 64 at MGH, N = 70 at PBRC). Each participant had %BF measured with VBC, three consumer and two professional bioimpedance analysis (BIA) systems. The PBRC participants also had air displacement plethysmography (ADP) measured. %BF measured by dual-energy x-ray absorptiometry (DXA) was set as the reference against which all other %BF measurements were compared. To test our scientific hypothesis we run multiple, pair-wise Wilcoxon signed rank tests where we compare each competing measurement tool (VBC, BIA, …) with respect to the same ground-truth (DXA). Relative to DXA, VBC had the lowest mean absolute error and standard deviation (2.16 ± 1.54%) compared to all of the other evaluated methods ( p < 0.05 for all comparisons). %BF measured by VBC also had good concordance with DXA (Lin’s concordance correlation coefficient, CCC: all 0.96; women 0.93; men 0.94), whereas BMI had very poor concordance (CCC: all 0.45; women 0.40; men 0.74). Bland-Altman analysis of VBC revealed the tightest limits of agreement (LOA) and absence of significant bias relative to DXA (bias −0.42%, R 2 = 0.03; p = 0.062; LOA −5.5% to +4.7%), whereas all other evaluated methods had significant ( p < 0.01) bias and wider limits of agreement. Bias in Bland-Altman analyses is defined as the discordance between the y = 0 axis and the regressed line computed from the data in the plot. In this first validation study of a novel, accessible, and easy-to-use system, VBC body fat estimates were accurate and without significant bias compared to DXA as the reference; VBC performance exceeded those of all other BIA and ADP methods evaluated. The wide availability of smartphones suggests that the VBC method for evaluating %BF could play an important role in quantifying adiposity levels in a wide range of settings. Trial registration : ClinicalTrials.gov Identifier: NCT04854421.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1038/s41746-022-00628-3

Cites

52 of 52 references verified0 citation errors · 0 not checkable

Statistics

Figures

Fig. 1 Consort diagram.
Fig. 1 Consort diagram.CC BY · PubMed Central
Fig. 2 Accuracy of different measurement methods. Mean absolute errors (MAE) of the various methods evaluated with DXA as the reference a. MAE of various methods in comparison to DXA stratified by sex, BMI, and ethnicity b. The colors are interpolated linearly from green (low error) to red (high error). We defined an acceptable error range as ≤3% (dark green). This value was selected as the mean MAE for ADP and the best of the pBIA devices. Light green, white, and red shadings indicate errors outside of this range. *p < 0.05 in comparison to VBC-DXA MAE.
Fig. 2 Accuracy of different measurement methods. Mean absolute errors (MAE) of the various methods evaluated with DXA as the reference a. MAE of various methods in comparison to DXA stratified by sex, BMI, and ethnicity b. The colors are interpolated linearly from green (low error) to red (high error). We defined an acceptable error range as ≤3% (dark green). This value was selected as the mean MAE for ADP and the best of the pBIA devices. Light green, white, and red shadings indicate errors outside of this range. *p < 0.05 in comparison to VBC-DXA MAE.CC BY · PubMed Central
Fig. 3 Comparison of %BF between VBC and DXA. Correlation between %BF by VBC and DXA, for a men and b women. The dashed line is identity and the solid line is the automatically fitted regression line. The two lines are very close to one another, and the correlations in both figures are significant at p < 0.0001. Bland-Altman analysis of the difference between %BF by VBC and DXA, for c men, d women, and e all. The horizontal black lines are at the mean ± 1.96 SD and the dashed gray lines are the fitted regression lines described by the equation in the panel.
Fig. 3 Comparison of %BF between VBC and DXA. Correlation between %BF by VBC and DXA, for a men and b women. The dashed line is identity and the solid line is the automatically fitted regression line. The two lines are very close to one another, and the correlations in both figures are significant at p < 0.0001. Bland-Altman analysis of the difference between %BF by VBC and DXA, for c men, d women, and e all. The horizontal black lines are at the mean ± 1.96 SD and the dashed gray lines are the fitted regression lines described by the equation in the panel.CC BY · PubMed Central
Fig. 4 Bland-Altman analyses of the differences between %BF by DXA and the six methods evaluated for estimation of %BF. a cBIA 1; b cBIA 2; c cBIA 3; d pBIA 1; e pBIA 2; f ADP. The horizontal black lines are at the mean ± 1.96 SD and the dashed gray lines are the fitted regression lines.
Fig. 4 Bland-Altman analyses of the differences between %BF by DXA and the six methods evaluated for estimation of %BF. a cBIA 1; b cBIA 2; c cBIA 3; d pBIA 1; e pBIA 2; f ADP. The horizontal black lines are at the mean ± 1.96 SD and the dashed gray lines are the fitted regression lines.CC BY · PubMed Central
Fig. 5 Bland-Altman analysis of repeated VBC measurements to assess technical error.
Fig. 5 Bland-Altman analysis of repeated VBC measurements to assess technical error.CC BY · PubMed Central
Fig. 6 Example smartphone images. The VBC algorithm requires images of the user’s front and back while wearing minimal clothing and holding an “A” pose as inputs to the model.
Fig. 6 Example smartphone images. The VBC algorithm requires images of the user’s front and back while wearing minimal clothing and holding an “A” pose as inputs to the model.CC BY · PubMed Central
Fig. 7 Different phases of our algorithm. Block diagrams illustrating the data flow for different phases of our algorithm: training a, testing b, and live use c. DXA images are never used.
Fig. 7 Different phases of our algorithm. Block diagrams illustrating the data flow for different phases of our algorithm: training a, testing b, and live use c. DXA images are never used.CC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

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?

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

Cites

Show all 52 references

Research works printed but not linked: 3

  1. Sommer, I. et al. The performance of anthropometric tools to determine obesity: a systematic review and meta‑analysis. Sci. Rep. 10, 12699 (2020). · DOI · Paper page
  2. Krizhevsky, A., Sutskever, I. & Hinton, G. ImageNet classification with deep convolutional neural network. Adv. Neural Inf. Process. Syst. 25, 1097–1105 (2012).
  3. Siri, W. E. 1961 Body composition from fluid spaces and density: analysis of methods. In: Brozek J., Henschel A. (eds) Techniques for Measuring Body Composition. National Academy of Sciences/National Research Council, Washington, DC, pp. 223–224.

Cited by

See all 75

Metadata

Authors
Majmudar, M. D., Chandra, S., Yakkala, K., Kennedy, S., Agrawal, A., Sippel, M., Ramu, P., Chaudhri, A., Smith, B., Criminisi, A., Heymsfield, S. B., & Stanford, F. C.
Affiliations
  • Amazon (United States) (US)
  • Pennington Biomedical Research Center (US)
  • Harvard University (US)
Corresponding author
Majmudar, Maulik D.
Venue
npj Digital Medicine
Published
2022-06-29
Volume
5
Issue
1
Pages
79-79
ISSN
2398-6352 (ISSN-L)
OpenAlex
W4283698854
Cited by
75
References
52
Citations by year
2026: 12 · 2025: 18 · 2024: 22 · 2023: 15 · 2022: 7
FWCI
9.60
Type
journal article
Language
en
Open access
Licence
Locations
  • npj Digital Medicine (publishedVersion, cc-by)
  • medRxiv (acceptedVersion)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion)
Funders
National Institute of Diabetes and Digestive and Kidney Diseases (P30DK072476, P30DK040561)
Topics
  • Health Sciences › Medicine › Physiology (medical) › Body Composition Measurement Techniques
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
  • Health Sciences › Medicine › Surgery › Body Contouring and Surgery
Keywords
adiposity assessment, body fat percentage, smartphone camera, computer vision, convolutional neural networks, body composition measurement, bioelectrical impedance analysis, air displacement plethysmography, dual-energy X-ray absorptiometry, body mass index, measurement accuracy, healthy adults, sex differences

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

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Smartphone camera based assessment of adiposity: a validation study. https://app.cochat.ai/papers/10.1038/s41746-022-00628-3
The paper
Majmudar, M. D., Chandra, S., Yakkala, K., Kennedy, S., Agrawal, A., Sippel, M., Ramu, P., Chaudhri, A., Smith, B., Criminisi, A., Heymsfield, S. B., & Stanford, F. C. (2022). Smartphone camera based assessment of adiposity: a validation study. npj Digital Medicine. https://doi.org/10.1038/s41746-022-00628-3

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.