A systematic review of automated segmentation of 3D computed-tomography scans for volumetric body composition analysis
The DOI matches its Crossref record; 15 of the 15 references it cites were found in a registry or an index.
Abstract
Automated computed tomography (CT) scan segmentation (labelling of pixels according to tissue type) is now possible. This technique is being adapted to achieve three‐dimensional (3D) segmentation of CT scans, opposed to single L3‐slice alone. This systematic review evaluates feasibility and accuracy of automated segmentation of 3D CT scans for volumetric body composition (BC) analysis, as well as current limitations and pitfalls clinicians and researchers should be aware of. OVID Medline, Embase and grey literature databases up to October 2021 were searched. Original studies investigating automated skeletal muscle, visceral and subcutaneous AT segmentation from CT were included. Seven of the 92 studies met inclusion criteria. Variation existed in expertise and numbers of humans performing ground‐truth segmentations used to train algorithms. There was heterogeneity in patient characteristics, pathology and CT phases that segmentation algorithms were developed upon. Reporting of anatomical CT coverage varied, with confusing terminology. Six studies covered volumetric regional slabs rather than the whole body. One study stated the use of whole‐body CT, but it was not clear whether this truly meant head‐to‐fingertip‐to‐toe. Two studies used conventional computer algorithms. The latter five used deep learning (DL), an artificial intelligence technique where algorithms are similarly organized to brain neuronal pathways. Six of seven reported excellent segmentation performance (Dice similarity coefficients > 0.9 per tissue). Internal testing on unseen scans was performed for only four of seven algorithms, whilst only three were tested externally. Trained DL algorithms achieved full CT segmentation in 12 to 75 s versus 25 min for non‐DL techniques. DL enables opportunistic, rapid and automated volumetric BC analysis of CT performed for clinical indications. However, most CT scans do not cover head‐to‐fingertip‐to‐toe; further research must validate using common CT regions to estimate true whole‐body BC, with direct comparison to single lumbar slice. Due to successes of DL, we expect progressive numbers of algorithms to materialize in addition to the seven discussed in this paper. Researchers and clinicians in the field of BC must therefore be aware of pitfalls. High Dice similarity coefficients do not inform the degree to which BC tissues may be under‐ or overestimated and nor does it inform on algorithm precision. Consensus is needed to define accuracy and precision standards for ground‐truth labelling. Creation of a large international, multicentre common CT dataset with BC ground‐truth labels from multiple experts could be a robust solution.
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.
Integrity signals
Registry
Cites
Statistics
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.
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?
In the citing paper
“Automated CT segmentation has increasingly been used for body-composition analysis and other quantitative imaging tasks, with growing interest in clinical implementation (13, 16–19, 33, 34).”
In the citing paper
“Previous studies have shown that multi-slice or volumetric CT analysis can provide body-composition information beyond that obtained from a single axial slice (15–19).”
In the citing paper
“Single-slice computed tomography (CT)-based body-composition analysis has provided important prognostic information, but it may not fully represent the three-dimensional (3D) distribution of thoracic muscle and intermuscular fat in patients undergoing thoracic RT (13, 15, 16).”
In the citing paper
“Recent advances in automated segmentation and 3D CT analysis have enabled efficient volumetric assessment of skeletal muscle and adipose compartments (16–19).”
In the citing paper
“Single-slice L3 measurement is convenient but a single anatomical plane is sampled, whereas volumetric analysis of the entire lumbar region theoretically captures the full distribution of muscle and adipose tissue [22,23].”
2 more claims about this paper
Create a free account to see every claim other papers make about this one.
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
The PRISMA 2020 statement: An updated guideline for reporting systematic reviews
Page, M. J., McKenzie, J. E., Bossuyt, P. M., Boutron, I., Hoffmann, T. C., Mulrow, C. D., Shamseer, L., Tetzlaff, J. M., Akl, E. A., Brennan, S. E., et al. (2021) · BMJ · doi:10.1136/bmj.n71 · Cited by 108,283
Asian Working Group for Sarcopenia: 2019 Consensus Update on Sarcopenia Diagnosis and Treatment
Chen, L.-K., Woo, J., Assantachai, P., Auyeung, T.-W., Chou, M.-Y., Iijima, K., Jang, H. C., Kang, L., Kim, M., Kim, S., et al. (2020) · Journal of the American Medical Directors Association · doi:10.1016/j.jamda.2019.12.012 · Cited by 7,070 · cited 1×
“47,48 After using linear regression to de-fi ne cut-offs, authors showed an 82% agreement between sex-speci fi c average waist SMI and a clinical diagnosis of sarcopenia.”
MethodPrado, C. M., Lieffers, J. R., McCargar, L. J., Reiman, T., Sawyer, M. B., Martin, L., & Baracos, V. E. (2008) · The Lancet Oncology · doi:10.1016/s1470-2045(08)70153-0 · Cited by 3,238
Warfield, S. K., Zou, K. H., & Wells, W. M. (2004) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2004.828354 · Cited by 2,335
Mourtzakis, M., Prado, C. M. M., Lieffers, J. R., Reiman, T., McCargar, L. J., & Baracos, V. E. (2008) · Applied Physiology, Nutrition, and Metabolism · doi:10.1139/h08-075 · Cited by 2,224
Collins, G. S., Reitsma, J. B., Altman, D. G., & Moons, K. (2015) · BMC Medicine · doi:10.1186/s12916-014-0241-z · Cited by 1,927
Shen, W., Punyanitya, M., Wang, Z., Gallagher, D., St.-Onge, M.-P., Albu, J., Heymsfield, S. B., & Heshka, S. (2004) · Journal of Applied Physiology · doi:10.1152/japplphysiol.00744.2004 · Cited by 1,692 · cited 3×
“19 This concept was then validated for CT within a cancer cohort using a single L3 axial slice.”
BackgroundMethodResultMitsiopoulos, N., Baumgartner, R. N., Heymsfield, S. B., Lyons, W., Gallagher, D., & Ross, R. (1998) · Journal of Applied Physiology · doi:10.1152/jappl.1998.85.1.115 · Cited by 1,522
The Liver Tumor Segmentation Benchmark (LiTS)
Bilic, P., Christ, P., Li, H. B., Vorontsov, E., Ben-Cohen, A., Kaissis, G., Szeskin, A., Jacobs, C., Mamani, G. E. H., Chartrand, G., et al. (2023) · Medical Image Analysis · doi:10.1016/j.media.2022.102680 · Cited by 1,412 · cited 1×
“68 As we anticipate a rising interest in automated volumetric BC analysis, the creation of a similar international dataset with pre-labelled CT scans would be impactful.”
MethodThe Liver Tumor Segmentation Benchmark (LiTS)
Bilic, P., Christ, P., Li, H. B., Vorontsov, E., Ben-Cohen, A., Kaissis, G., Szeskin, A., Jacobs, C., Mamani, G. E. H., Chartrand, G., et al. (2019) · arXiv · doi:10.48550/arxiv.1901.04056 · Cited by 1,412
Human Body Composition: Advances in Models and Methods
B. Heymsfield, S., Wang, Z., Baumgartner, R. N., & Ross, R. (1997) · Annual Review of Nutrition · doi:10.1146/annurev.nutr.17.1.527 · Cited by 527 · cited 1×
“14 It was demonstrated over 30 years ago that CT and MRI respectively could be used to quantify regional and total body SM and AT volumes.”
Backgroundvan der Werf, A., Langius, J. A. E., de van der Schueren, M. A. E., Nurmohamed, S. A., van der Pant, K. A. M. I., Blauwhoff-Buskermolen, S., & Wierdsma, N. J. (2018) · European Journal of Clinical Nutrition · doi:10.1038/s41430-017-0034-5 · Cited by 295
Body Composition Profiling in the UK Biobank Imaging Study
Linge, J., Borga, M., West, J., Tuthill, T., Miller, M. R., Dumitriu, A., Thomas, E. L., Romu, T., Tunón, P., Bell, J. D., et al. (2018) · Obesity · doi:10.1002/oby.22210 · Cited by 223
Reference Values for Skeletal Muscle Mass – Current Concepts and Methodological Considerations
Walowski, C. O., Braun, W., Maisch, M. J., Jensen, B., Peine, S., Norman, K., Müller, M. J., & Bosy-Westphal, A. (2020) · Nutrients · doi:10.3390/nu12030755 · Cited by 200 · cited 1×
“There is a considerable heterogeneity in how studies set SMI cut-offs for sarcopenia 71 including optimal strati fi cation compared with a speci fi c outcome such as overall survival 11,20 ; percentiles 72 ; standard deviations 73,74 ; or using ranges de fi ned in previous studies.”
A Clinician's Guide to Artificial Intelligence: How to Critically Appraise Machine Learning Studies
Faes, L., Liu, X., Wagner, S. K., Fu, D. J., Balaskas, K., Sim, D. A., Bachmann, L. M., Keane, P. A., & Denniston, A. K. (2020) · Translational Vision Science & Technology · doi:10.1167/tvst.9.2.7 · Cited by 162 · cited 2×
“31 The points considered are speci fi ed, along with technical terminology explanations.”
Background
Research works printed but not linked: 60
- 10.1016/j.afos.2021.03.002 · DOI
- 10.1097/SLA.0000000000002679 · DOI · Paper page
- 10.1016/j.chest.2019.04.115 · DOI · Paper page
- 10.3389/fonc.2019.01561 · DOI
- 10.1186/s12885-019-6169-0 · DOI · Paper page
- 10.1002/jcsm.13126 · DOI
- Guo Y, 2023, Association between sarcopenia and clinical outcomes in patients with hepatocellular carcinoma: an updated meta‐analysis, Scientific Reports, 13, 1
- 10.1136/ijgc-2020-001678 · DOI
- 10.1007/s00384-021-03839-4 · DOI · Paper page
- 10.1001/jamaoncol.2018.0137 · DOI · Paper page
- 10.1200/JCO.2012.45.2722 · DOI · Paper page
- 10.1016/0720-048X(93)90023-G · doi:10.1016/0720-048x(93)90023-g
- 10.1002/jcsm.12573 · DOI
- Kvist H, 1986, Adipose tissue volume determinations in women by computed tomography: technical considerations, Int J Obes (Lond), 10, 53
- 10.1093/ajcn/54.1.18 · DOI
- 10.1038/oby.2012.168 · DOI · Paper page
- MalietzisG CurrieAC AthanasiouT JohnsN AnyameneN Glynne‐JonesR et al.Influence of body composition profile on outcomes following colorectal cancer surgery.2016;103:572–580. · DOI · Paper page
- 10.1158/1078-0432.CCR-08-2242 · DOI · Paper page
- 10.1245/s10434-016-5188-1 · DOI
- 10.1002/jmri.24726 · DOI · Paper page
- 10.1371/journal.pone.0163332 · DOI · Paper page
- 10.1038/s41598-022-07556-y · DOI
- Covidence systematic review software Veritas Health Innovation Melbourne Australia. Available atwww.covidence.org
- 10.1117/1.JRS.15.041501 · doi:10.1117/1.jrs.15.041501
- 10.1016/j.ejso.2018.04.008 · doi:10.1016/j.ejso.2018.04.008
- 10.1186/s12880-015-0068-x · DOI · Paper page
- 10.1117/1.JMI.5.1.015006 · DOI · Paper page
- HuP HuoY KongD CarrJJ AbramsonRG HartleyKGet al.Automated characterization of body composition and frailty with clinically acquired CT. Computational Methods and Clinical Applications in Musculoskeletal Imaging: 5th International Workshop MSKI 2017 Held in Conjunction with MICCAI 2017 Quebec City QC Canada September 10 2017 Revised Selected Papers/Ben Glocker Jianhua Yao Toma 2018;10734:25‐35. · DOI
- 10.1007/s10278-019-00178-3 · DOI
- 10.1002/mp.14465 · DOI
- 10.1002/mp.14141 · DOI
- DaMa VincentChow KarteekPopuri Mirza FaisalBeg.Comprehensive validation of automated whole body skeletal muscle adipose tissue and bone segmentation from 3D CT images for body composition analysis: towards extended body composition. ArXiv. 2021;Preprint.
- KoitkaS KrollL MalamutmannE OezcelikA NensaF.Fully‐automated body composition analysis in routine CT imaging using 3D semantic segmentation convolutional neural networks. ArXiv 2021. · DOI · Paper page
- 10.1016/j.clnu.2021.06.025 · DOI
- Wang M, 2019, Label fusion method combining pixel greyscale probability for brain MR segmentation, Scientific Reports, 9, 1
- DingW LiL ZhuangX HuangL.Cross‐modality multi‐atlas segmentation using deep neural networks. Published online August 14 2020. · DOI
- 10.1136/bmjopen-2019-031018 · doi:10.1136/bmjopen-2019-031018
- 10.1016/j.euf.2021.03.013 · DOI
- 10.1055/s-0039-3400267 · DOI
- 10.2214/AJR.22.27749 · DOI · Paper page
- 10.21037/qims-21-370 · DOI
- 10.1111/j.1745-3984.1971.tb00916.x · DOI
- 10.1111/codi.15621 · DOI
- 10.1016/j.clnu.2022.03.016 · DOI
- 10.1016/j.jocd.2013.08.007 · DOI · Paper page
- 10.1002/crt2.53 · DOI
- 10.3389/fneur.2018.00679 · DOI · Paper page
- 10.1038/s41598-018-31911-7 · DOI · Paper page
- 10.3389/fonc.2022.772403 · DOI
- 10.1016/j.compbiomed.2021.104497 · DOI
- 10.1152/jappl.2000.89.2.599 · DOI
- 10.1016/S2589-7500(19)30123-2 · DOI · Paper page
- 10.1038/s41597-020-00715-8 · DOI
- 10.1016/j.media.2017.06.015 · DOI · Paper page
- Mesguich C, 2022, 18F‐FDG PET/CT and MRI in the management of multiple myeloma: a comparative review, Front Nucl Med, 12
- 10.1002/mp.15821 · DOI
- 10.1038/s41598-020-79471-z · DOI
- 10.1159/000480407 · DOI
- 10.1245/s10434-016-5385-y · DOI
- 10.1002/bjs.10063 · DOI · Paper page
Cited by
Wang, D., Sui, J., Chen, J., Chen, L., Yang, Q., Wang, Y., & Lin, S. (2026) · Healthcare · doi:10.3390/healthcare14172846
Sung, S., Choi, H. J., Jeong, S., Cho, W., Kim, Y. J., Cha, H., Min, C. H., & Kim, M. S. (2026) · Frontiers in Oncology · doi:10.3389/fonc.2026.1946212
Metadata
- Authors
- Mai, D. V. C., Drami, I., Pring, E. T., Gould, L. E., Lung, P., Popuri, K., Chow, V., Beg, M. F., Athanasiou, T., Jenkins, J. T., & the BiCyCLE Research Group
- Affiliations
- Department of Surgery St Mark's Academic Institute, St Mark's Hospital London UK
- Department of Computer Science Memorial University of Newfoundland St Johns Canada
- School of Engineering Science Simon Fraser University Burnaby Canada
- Department of Surgery and Cancer Imperial College London UK
- ORCID
- Mai, Dinh Van Chi
- Venue
- Journal of Cachexia, Sarcopenia and Muscle
- Publisher
- Wiley
- Published
- 2023-08-10
- Volume
- 14
- Issue
- 5
- Pages
- 1973-1986
- ISSN
- 2190-5991
- 2190-6009
- Cited by
- 78
- References
- 15
- Language
- en
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
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.