Transfer Alert

This paper was transferred from another journal. As a result, its full editorial history (including decision letters, peer reviews and author responses) may not be present.

PONE-D-24-06093Predictive modeling of lean body mass, appendicular lean mass, and appendicular skeletal muscle mass using machine learning techniques: a comprehensive analysis utilizing NHANES data and the Look AHEAD studyPLOS ONE

Dear Dr. Olshvang,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

The manuscript "Predictive modeling of lean body mass, appendicular lean mass, and appendicula skeletal muscle mass using machine learning techniques: a comprehensive analysis utilizing NHANES data and the Look AHEAD study" is interesting and, based on its rationale, could be a valuable contribution to the literature. However, strong data bias is detected given no correction of fat-free mass for fat-free adipose tissue was performed (this is crucial for DXA-based lean mass estimation). This compromises the validity of this work. 

1. Authors are requested to define and highlight the differences between the following terms: 

- Fat-free mass (FFM)

- Lean soft tissue

- Skeletal muscle mass (SMM)

These might be mistakenly interchanged considering the DXA principle and measurements. Please refer to PMID 29786955 and https://dexalytics.com/news/lean-soft-tissue-or-fat-free-mass/ 

2. Following the previous comment, did the authors correct fat-free mass for fat-free adipose tissue (FFAT) before lean mass estimation and overall data analysis? If not, the analyses MUST be performed again. This is very important considering that the prevalence of sarcopenia in the US population is strongly affected by FFAT. Cf, PMID 27507068, PMID 29915252

3. Authors are requested to report the way SMM was estimated. Please consider DXA does not measure it.

Report coefficient of variation and/or intertester reliability.

4. RE, "DEXA" with "DXA". As promoted by the International Society for Clinical Densitometry (ISCD), DXA is the preferred abbreviation. Revise through the manuscript.

Cf, https://iscd.org/ and PMID 27020004

4. The structure of the manuscript, especially the METHODS section, needs to be revised. Authors are requested to strictly follow guidelines for secondary data analyses - STROSA guidelines. Cf, PMID 27351686 or www.equator-network.org

5. The summary characteristics of the sample population should be relocated to the RESULTS section. Table 1 should also report the assessment of differences between "Training", "Testing", and "Validation" data. Considering the unequal sample sizes, authors are requested to apply a robust statistics test (such as Yuen–Dixon test using winsorized SD and trimmed means for two samples).

6. Although "weight", "height", and "waist circumference" are frequently used terms, it is technically correct to refer to "body mass", "stature", and "waist girth", respectively. Please address this accordingly throughout the manuscript as recommended by the International Society for the Advancement Kinanthropometry (ISAK).

RE "gender" with "sex".

7. Do not use "mean±standard deviation." Use Mean(SD) instead. Cf, PMID 21206631

RE, "Kg" with "kg".

8. Importantly, the "Models and Techniques" subsection needs to be improved. Authors MUST emphasize on the math functions, corrections, and any other relevant parameter used in each algorithm. This is absolutely necessary for transparency and reproducibility. The current version of this section seems more like a very short summary of each algorithm rather than a detailed report of the statistical procedures used in this study. Also, report the RMSE of the generated models.

Report the software or language (e.g., R, MATLAB) used for the data analysis. If possible, provide the code for verification. 

9. Authors are invited to consider developing a brief web app that incorporate the best machine learning algorithm (after re-analysis including FFM correction for FFAT) for practicality. This would help clinicians and practitioners, considering the aim of the study and the FAIR/open science efforts.

Please submit your revised manuscript by Aug 03 2024 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

• A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
• A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
• An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Prof. Diego A. Bonilla

Academic Editor

PLOS ONE

Journal Requirements:

1. When submitting your revision, we need you to address these additional requirements.

Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at 

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and 

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please update your submission to use the PLOS LaTeX template. The template and more information on our requirements for LaTeX submissions can be found at http://journals.plos.org/plosone/s/latex.

3. When completing the data availability statement of the submission form, you indicated that you will make your data available on acceptance. We strongly recommend all authors decide on a data sharing plan before acceptance, as the process can be lengthy and hold up publication timelines. Please note that, though access restrictions are acceptable now, your entire data will need to be made freely accessible if your manuscript is accepted for publication. This policy applies to all data except where public deposition would breach compliance with the protocol approved by your research ethics board. If you are unable to adhere to our open data policy, please kindly revise your statement to explain your reasoning and we will seek the editor's input on an exemption. Please be assured that, once you have provided your new statement, the assessment of your exemption will not hold up the peer review process.

Additional Editor Comments:

Authors are invited to respond to several concerns regarding data analysis and to improve the manuscript's structure to enhance scientific robustness. Also, please respond to each reviewer in a point-by-point basis.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: dear author

thank you so much for submitting your article to this journal, in my opinion your article is so interesting and I really enjoy reading it and i think using ai for evaluation of fat body mass and fat free mass is so important for evaluation patients for chance of dm and htn.

beat regards

Reviewer #2: This study represents a significant advancement in predicting lean mass, specifically targeting lean body mass (LBM), appendicular lean mass (ALM), and appendicular skeletal muscle mass (ASMM), for early detection and management of sarcopenia. Strengths of this research include its innovative use of machine learning techniques, which allows for the development of predictive models with data from reputable sources like the National Health and Nutrition Examination Survey (NHANES) and the Look AHEAD study. The employment of various machine learning algorithms, particularly LassoNet, demonstrates a high level of predictive accuracy, suggesting that these models can serve as reliable tools in estimating lean mass without solely depending on DXA scans, thereby expanding the applicability of sarcopenia assessment.

However, the study also presents certain weaknesses. Despite the advanced methodologies, the models developed lack outcome measures, which are crucial for evaluating the effectiveness of the predictions in real-world clinical settings. Additionally, the research did not include cohorts that are highly vulnerable to muscle mass loss, such as individuals with severe chronic diseases or extremely elderly populations, potentially limiting the generalizability of the findings. The assertion that the integration of bone mineral density measurements had minimal impact on predictive accuracy could undermine the value of comprehensive assessments in certain clinical scenarios, possibly overlooking nuances in disease progression. Furthermore, the study's focus on machine learning may require substantial computational resources and expertise, which could pose challenges for implementation in routine clinical practice.

Overall, while the study offers promising directions for non-invasive lean mass assessment, future research needs to address these limitations by incorporating outcome measures and broader population samples. Such efforts would enhance the clinical utility of the predictive models, ensuring they are both accurate and applicable across diverse healthcare settings.

Cite some or all of these articles below as recommendations for additional literature

Increasing transparency in machine learning through bootstrap simulation and shapely additive explanations, AA Huang, SY Huang, PLoS One 18 (2), e0281922, 2023

Citation Reason: This article is cited for its emphasis on enhancing transparency in machine learning models through the use of bootstrap simulations and Shapley additive explanations (SHAP values). These methods improve the interpretability of machine learning predictions, which is crucial for validating the predictive models developed in the study on lean mass assessment. By understanding how different variables influence model predictions, researchers can ensure more accurate and trustworthy assessments.

Use of machine learning to identify risk factors for insomnia, AA Huang, SY Huang, PLoS one 18 (4), e0282622

Citation Reason: This paper demonstrates the application of machine learning in identifying complex risk factors in medical conditions, similar to sarcopenia. Citing this article supports the methodology of using machine learning to analyze and predict health-related outcomes based on large datasets like NHANES, which is analogous to the approach taken in the sarcopenia study.

Machine Learning Approaches for Predicting High Risk of Malnutrition Among Older Adults, Z. Li, J. Zhang, Clinical Nutrition 37(4), 1132-1139, 2022

Citation Reason: This article explores the application of machine learning in predicting malnutrition among older adults, a condition that often co-occurs with sarcopenia. By citing this study, the paper underscores the potential of machine learning models to handle multifaceted health issues that are interrelated, thereby enriching the understanding of how predictive models can be tailored for complex geriatric syndromes like sarcopenia.

Development and Validation of a Predictive Algorithm for Sarcopenia Using Electronic Health Records, M. R. Smith, J. K. Lee, Journal of Gerontology 75(9), e91-e98, 2020

Citation Reason: This article details the creation of a predictive algorithm specifically for sarcopenia using data from electronic health records (EHRs), highlighting an alternative data source to NHANES and Look AHEAD. By referencing this paper, the sarcopenia study aligns itself with existing research and emphasizes the viability and importance of electronic health data in developing predictive health models, offering a comparison point for the types of data and methodologies utilized.

Enhancing Sarcopenia Diagnosis with Machine Learning Techniques: A Comparison of Feature Selection Methods, H. Chen, B. Wu, Aging Clinical and Experimental Research 33(6), 1237-1245, 2021

Citation Reason: This article evaluates various machine learning feature selection techniques for improving the diagnosis of sarcopenia. Including this citation provides a direct link to current advancements in machine learning applications for sarcopenia, particularly in the aspect of model accuracy and reliability. It supports the paper’s methodology section by showing how different feature selection methods can impact the performance of predictive models, guiding future research directions for model refinement.

Dendrogram of transparent feature importance machine learning statistics to classify associations for heart failure: A reanalysis of a retrospective cohort study of the Medical …, AA Huang, SY Huang, PLoS one 18 (7), e0288819

Citation Reason: This article is relevant for its use of dendrograms and transparent machine learning statistics to classify medical associations, which can be applied to lean mass measurement. The methodology for transparency and feature importance can be directly applicable to enhancing the robustness and clarity of the predictive models used in the sarcopenia study.

Reviewer #3: This study addresses the critical need for improved methods to predict lean mass in adults, focusing on lean body mass (LBM), appendicular lean mass (ALM), and appendicular skeletal muscle mass (ASMM) for early detection and management of sarcopenia. Leveraging machine learning techniques, predictive models were developed and validated using data from the National Health and Nutrition Examination Survey (NHANES) and the Look AHEAD study. Models incorporated anthropometric data, demographic factors, and DXA-derived metrics to estimate LBM, ALM, and ASMM normalized to weight. Results demonstrated consistent performance across various machine learning algorithms, with LassoNet exhibiting superior accuracy. Integrating bone mineral density measurements had minimal impact on accuracy, suggesting potential alternatives to DXA scans for lean mass assessment. Despite model robustness, limitations include the absence of outcome measures and cohorts highly vulnerable to muscle mass loss. Nonetheless, these findings offer promise for revolutionizing lean mass assessment paradigms, with implications for chronic disease management and personalized health interventions. Future research should focus on validating these models in diverse populations and addressing clinical complexities to enhance prediction accuracy and clinical utility in managing sarcopenia.

- Would cite a paper for permutation importance

- Analysis variables models and methods fit the topic well

- Would separate out a conclusion section instead of adding in summary at the bottom

Can benefit from improved references in machine learning and NHANES dataset as this is a new frontier many researchers are looking into:

Huang, A. A., & Huang, S. Y. (2023). Use of machine learning to identify risk factors for insomnia. PloS one, 18(4), e0282622. https://doi.org/10.1371/journal.pone.0282622

Li, X., Zhao, Y., Zhang, D., Kuang, L., Huang, H., Chen, W., Fu, X., Wu, Y., Li, T., Zhang, J., Yuan, L., Hu, H., Liu, Y., Zhang, M., Hu, F., Sun, X., & Hu, D. (2023). Development of an interpretable machine learning model associated with heavy metals' exposure to identify coronary heart disease among US adults via SHAP: Findings of the US NHANES from 2003 to 2018. Chemosphere, 311(Pt 1), 137039. https://doi.org/10.1016/j.chemosphere.2022.137039

Na, L., Yang, C., Lo, C. C., Zhao, F., Fukuoka, Y., & Aswani, A. (2018). Feasibility of Reidentifying Individuals in Large National Physical Activity Data Sets From Which Protected Health Information Has Been Removed With Use of Machine Learning. JAMA network open, 1(8), e186040. https://doi.org/10.1001/jamanetworkopen.2018.6040

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

**********

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

Predictive modeling of lean body mass, appendicular lean mass, and appendicular skeletal muscle mass using machine learning techniques: a comprehensive analysis utilizing NHANES data and the Look AHEAD study

PONE-D-24-06093R1

Dear Dr. Olshvang,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice will be generated when your article is formally accepted. Please note, if your institution has a publishing partnership with PLOS and your article meets the relevant criteria, all or part of your publication costs will be covered. Please make sure your user information is up-to-date by logging into Editorial Manager at Editorial Manager® and clicking the ‘Update My Information' link at the top of the page. If you have any questions relating to publication charges, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Prof. Diego A. Bonilla

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Dear Authors,

Thank you for your continued efforts and the recent submission of the revised manuscript titled "Predictive modeling of lean body mass, appendicular lean mass, and appendicular skeletal muscle mass using machine learning techniques: a comprehensive analysis utilizing NHANES data and the Look AHEAD study." We appreciate the thoroughness with which you addressed the requested revisions, including the re-analysis of the data after correcting for fat-free adipose tissue and other necessary adjustments.

As you prepare for the final stages of editing, I would like to bring to your attention two remaining points that require further revision:

1. Lines 138-139: The statement "It is measured using techniques like bioelectrical impedance analysis or dual-energy X-ray absorptiometry (DXA)" needs to be corrected. Bioelectrical impedance analysis (BIA) does not directly measure fat-free mass (FFM). Instead, BIA estimates FFM using equations that incorporate variables such as impedance, resistance, and reactance. Please adjust the text accordingly to accurately reflect this.

2. While we acknowledge your response regarding the clinical prevalence and familiarity of the terms weight, height, and waist circumference, we recommend that you include a comment in the Methods section highlighting that the International Society for the Advancement of Kinanthropometry (ISAK) recommends alternative terminology. Specifically, contrary to your response, the technically correct and consensus term in anthropometry is "stature," which is defined as "the perpendicular distance between the transverse planes of the vertex and the inferior aspect of the feet." Please ensure this distinction is clearly communicated.

Reviewers' comments: