PONE-D-24-31115Table Tennis Coaching System Based on a Multimodal Large Language Model with a Table Tennis Knowledge Base

PLOS ONE

Dear Dr. Ma,

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 has been evaluated by three reviewers, and their comments are available below. The reviewers have raised a number of major concerns.

They note several shortcomings in the methodology, including details about inputs to the model and the validity of the evaluation methods. Please take care to significantly revise the Methods, since manuscripts published in PLOS ONE must be reproducible. In addition to the reviewers' concerns, please ensure you have provided sufficient information about the recruitment and demographic information of both the novice players and the experts, and also ensure that all data and code are fully available. If ethics approval was sought please provide details, or else explain why this was not required.

Additional concerns include the limited validation of the tool's functionality as a tool to actually improve table tennis players' abilities  - which renders some of the conclusions overstated - and insufficient discussion of the utility of the tool compared to similar approaches. Limitations of the approach also need to be discussed further.

Could you please carefully revise the manuscript to address all comments raised? Please note that your revisions should be substantial, and the manuscript may be rejected if the concerns are not addressed adequately.

Please submit your revised manuscript by Nov 08 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,

Marianne Clemence

Staff Editor

PLOS ONE

Journal Requirements:

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

1. 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 note that PLOS ONE has specific guidelines on code sharing for submissions in which author-generated code underpins the findings in the manuscript. In these cases, we expect all author-generated code to be made available without restrictions upon publication of the work. Please review our guidelines at https://journals.plos.org/plosone/s/materials-and-software-sharing#loc-sharing-code and ensure that your code is shared in a way that follows best practice and facilitates reproducibility and reuse.

3. You indicated that ethical approval was not necessary for your study. We understand that the framework for ethical oversight requirements for studies of this type may differ depending on the setting and we would appreciate some further clarification regarding your research. Could you please provide further details on why your study is exempt from the need for approval and confirmation from your institutional review board or research ethics committee (e.g., in the form of a letter or email correspondence) that ethics review was not necessary for this study? Please include a copy of the correspondence as an ""Other"" file.

4. 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.

5. Thank you for stating the following in your Competing Interests section: 

“The authors have no conflicts of interest to disclose”

Please complete your Competing Interests on the online submission form to state any Competing Interests. If you have no competing interests, please state "The authors have declared that no competing interests exist.", as detailed online in our guide for authors at http://journals.plos.org/plosone/s/submit-now

 This information should be included in your cover letter; we will change the online submission form on your behalf.

6. In the online submission form, you indicated that [Code is open to the public. Data can be requested via email].

All PLOS journals now require all data underlying the findings described in their manuscript to be freely available to other researchers, either 1. In a public repository, 2. Within the manuscript itself, or 3. Uploaded as supplementary information.

This policy applies to all data except where public deposition would breach compliance with the protocol approved by your research ethics board. If your data cannot be made publicly available for ethical or legal reasons (e.g., public availability would compromise patient privacy), please explain your reasons on resubmission and your exemption request will be escalated for approval.

7. Please amend the manuscript submission data (via Edit Submission) to include author Dr. Rongzhi Li.

8. We note that Figures 2,3 and 6 includes an image of a [patient / participant / in the study].

As per the PLOS ONE policy (http://journals.plos.org/plosone/s/submission-guidelines#loc-human-subjects-research) on papers that include identifying, or potentially identifying, information, the individual(s) or parent(s)/guardian(s) must be informed of the terms of the PLOS open-access (CC-BY) license and provide specific permission for publication of these details under the terms of this license. Please download the Consent Form for Publication in a PLOS Journal (http://journals.plos.org/plosone/s/file?id=8ce6/plos-consent-form-english.pdf). The signed consent form should not be submitted with the manuscript, but should be securely filed in the individual's case notes. Please amend the methods section and ethics statement of the manuscript to explicitly state that the patient/participant has provided consent for publication: “The individual in this manuscript has given written informed consent (as outlined in PLOS consent form) to publish these case details”.

If you are unable to obtain consent from the subject of the photograph, you will need to remove the figure and any other textual identifying information or case descriptions for this individual.

9. We note you have included a table to which you do not refer in the text of your manuscript. Please ensure that you refer to Table 1 in your text; if accepted, production will need this reference to link the reader to the Table.

[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: Partly

Reviewer #2: Partly

Reviewer #3: Partly

**********

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

Reviewer #1: Yes

Reviewer #2: No

Reviewer #3: No

**********

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: No

Reviewer #2: No

Reviewer #3: No

**********

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: No

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: GENERAL CONCEPT COMMENTS

Congratulations to the authors for their diligent work and valuable contribution to the field through this research endeavour. After carefully reviewing the article and considering the recommendations provided, it is evident that the manuscript presents valuable insights into the topic. The article “Table Tennis Coaching System Based on a Multimodal Large Language Model with a Table Tennis Knowledge Base” is an innovative and promising contribution to the intersection of sports and AI. The application of multimodal large language models (LLMs) for table tennis coaching has potential to revolutionize training by providing personalized, precise, and cost-effective guidance. However, there are several areas in which the article can be improved to meet the standards of scientific rigor typically applied in research, particularly in the methods section, which lacks crucial details.

SPECIFIC COMMENTS

1. INTRODUCTION

Specific textual recommendations for accuracy and precision:

(P. 8) - The introduction provides a good foundation on the benefits of table tennis, but the transition to AI application feels abrupt. A stronger link between the issues with current table tennis training methodologies and how AI offers an innovative solution would improve the flow.

• Suggestion: Expand on why current technologies (like VR or traditional video analysis) are insufficient for personalized, high-efficiency training and how multimodal LLMs can bridge this gap.

(P. 8) – Correct 10 and 11 reference formats.

(P. 8) – Correct 12 and 13 reference formats.

(P. 8) - “Children who regularly engage in table tennis training demonstrate better physical composition and fitness, compared to active children who do not participate in regular physical activities[3]” The sentence seems confusing; consider revising it for clarity by specifying the comparison group and the benefits more clearly.

(P. 8) – The acronym “AI” appears before it is introduced in the text. To improve clarity, “Artificial Intelligence (AI)” should be introduced the first time it’s mentioned.

(P. 9) – Correct 20 and 21 reference formats.

(P. 9) – Correct 23 and 24 reference formats.

(P. 9) - The discussion of technological advancements (VR, AI) could benefit from a more structured comparison of existing tools used in sports and how this system differs or improves upon them.

2. METHODS

Consider revising the text to avoid using first-person language, as scientific writing typically favors a more objective tone. For example, instead of “I positioned my smartphone,” use “The smartphone was positioned” to maintain formality and focus on the research process.

Specific textual recommendations for accuracy and precision:

(P. 9) - The method involving the placement of a single smartphone for capturing video is innovative for its simplicity. However, it may limit the system’s ability to capture complex footwork and body dynamics.

• Suggestion: Acknowledge that using only one camera may reduce the accuracy of certain movements (like footwork) and propose future improvements, such as adding multiple camera angles. This limitation should be made clear in the methodology section as well as in the limitations section.

(P. 10) – The figure 1, 2, 3, 4 must appear immediately after its introduction in the text.

(P. 10) - While the system mentions GPT-4 as the core model for providing training recommendations, there is insufficient detail about how the model was trained or adjusted for this specific application. The use of LLMs in a complex task such as table tennis coaching demands more transparency, especially around: (i) Whether the model was pre-trained or adjusted; (ii) The dataset used for fine-tuning (e.g., how much data, and from which sources?); (iii) Hyperparameters such as learning rate, batch size, and optimization algorithms used in training, similar to how traditional statistical methods require reporting of confidence intervals or normality checks; (iv) Was any data augmentation or preprocessing applied to enhance the model’s performance?.

(P. 11) - The integration of motion capture data (from OpenPose) and visual recognition data into the LLM is a significant aspect of the system, but there is no clear explanation of how this data is formatted or fed into GPT-4. The authors must explain how multimodal data is handled, much like a traditional study would report assumptions checks or data transformations before analysis.

Additional questions also rise: (i) Are the visual and motion data converted to text or encoded in another form before being processed by the model?; (ii) If the data is fed into the model as structured inputs, provide details on the data pipeline and preprocessing steps; (iii) Without this information, readers are left uncertain about the complexity of the integration between the multimodal data and the LLM.

(P. 11) - OpenPose is a robust tool for motion capture, but it is important to note that its performance may degrade in cluttered or poorly-lit environments, which could affect amateur users.

• Suggestion: Add a section on the environmental limitations of OpenPose and how it affects data accuracy. Including some recommendations on ideal lighting and camera quality for users could be helpful.

(P. 11) - The core functionality of the system relies on GPT-4 generating recommendations based on visual and motion data inputs. However, there is no mention of the prompts, or the structure of the input data used to query the language model. This is a critical detail for understanding how the model generates athlete-specific advice. The article should provide examples of the exact prompts given to GPT-4, much like one would report the mathematical formulas or procedures used in a statistical test. For instance: (i) What input does the model receive? Is it a natural language description, a set of structured variables, or something else?; (ii) Are there any predefined templates or rules for these prompts?; (iii) This omission weakens the study’s transparency, as it is crucial to understand the interaction between multimodal data and the LLM.

(P. 11) - The study should also address potential concerns about overfitting, particularly because it is difficult for AI models to generalize across varied real-world scenarios without extensive data. It’s important to understand if the system was tested on diverse datasets, akin to how normality or homogeneity tests are used in statistical research. In example: (i) Was the model validated on external datasets or players not included in the original training data? And (ii) Did the authors apply regularization techniques (e.g., dropout layers, early stopping) to prevent overfitting?

(P. 12) - The dataset is well-defined in terms of error categorization (e.g., wrist errors, footwork errors), but the process of how data were annotated by experts could be more detailed.

• Suggestion: Include details on the inter-rater reliability among the coaches who annotated the errors. How consistent were they in identifying these errors? A mention of this would improve the perceived robustness of the dataset.

(P. 12 L. 5–25) - The system’s accuracy in identifying table tennis errors is reported, but crucial details, like confidence intervals or variance in accuracy across different datasets, are missing. Just as traditional statistical analyses require detailed metrics (effect sizes, p-values), AI models must be rigorously evaluated using standard performance metrics. In example: (i) Report accuracy, precision, recall, and F1-scores for error identification; (ii) Include confidence intervals or standard deviations for these metrics to indicate the model’s robustness; (iii) Clarify whether cross-validation was used to ensure that the model is not overfitting to a particular set of data; and (iv) Include a confusion matrix to show where the model succeeds or fails in classifying various errors (e.g., racket position vs. footwork).

(P. 12 L. 10–20) - The article does not clarify whether the outputs generated by GPT-4 were evaluated for quality, accuracy, or consistency. This is essential, as one would expect an evaluation of any AI system’s output similar to how confidence intervals or p-values are reported in statistical studies. As a reviewer I have a few questions: (i) Was the quality of GPT-4’s recommendations assessed? Were there metrics or cross-validation techniques applied to ensure the model consistently provides correct or useful feedback?; (ii) Did the authors perform any error analysis to understand common mistakes made by the model (e.g., false positives or incorrect classifications)?; (iii) If experts evaluated the system, what were the specific criteria used to measure GPT-4’s performance?

(P. 12) – Table 1 should be introduced in the text before the actual table.

3. RESULTS

Specific textual recommendations for accuracy and precision:

(P. 13) – Figure 5 should be introduced in the text prior to its appearance in the document.

(P. 13) – The figure 6 must appear immediately after its introduction in the text.

(P. 13) - The system demonstrates impressive accuracies (73% for arm-related errors and 82% for racket errors). However, these percentages are not sufficient for real-world use if aiming for higher levels of training (beyond beginner-level players).

• Suggestion: Clarify that these accuracy metrics are primarily for beginners and that further refinement is needed for advanced players. Also, suggest ways to increase accuracy, such as incorporating biomechanics data or higher-resolution tracking.

(P. 14) - The comparison between GPT-4, Gemini Pro, and Qwen-VL is useful, but the decision criteria for selecting GPT-4 should be elaborated.

• Suggestion: Add more detailed metrics to the model comparison (e.g., computation time, resource consumption, scalability) and explain why GPT-4 was chosen over the others in those dimensions, not just for accuracy and false positives.

(P. 14) – Figure 7 should be introduced in the text prior to its appearance in the document.

(P. 15) – Figure 9 should be introduced in the text prior to its appearance in the document.

4. DISCUSSION

Specific textual recommendations for accuracy and precision:

(P. 16) - The cost analysis focuses primarily on beginner-level applications, but there is no mention of scaling this system for more advanced or professional users.

• Suggestion: Include a section that discusses how the system could be adapted for semi-professional or professional players, with potentially more advanced (and more expensive) hardware. This would broaden the scope of the application and make the study more impactful.

(P. 16) - The lower accuracy in recognizing footwork errors is mentioned, but more detail should be given regarding how this will be improved in future research.

• Suggestion: Propose specific technical solutions, such as integrating a multi-camera system or combining inertial measurement units (IMUs) with visual data. Without a clear plan, it may appear that this limitation will remain unaddressed.

5. CONCLUSION

Specific textual recommendations for accuracy and precision:

(Page 17) - The conclusion touches on future improvements in footwork recognition and training for high-level athletes, but these should be expanded with more concrete plans.

6. REFERENCES

In most of the references, the symbol ‘[J]’ appears. Please confirm whether this is intentional.

Reviewer #2: Summary: This work introduces an AI-driven TT coaching system using a Multimodal LLM with a table tennis knowledge base, aiming to provide precise training guidance and match strategies for TT novice players eventually improving the overall performance.

Merits:

The paper tackles an interesting research domain.

Demerits and Questions:

There is a gap between highlighting the novel contributions as the approaches have been utilized for the study. Can the authors explicitly discuss and highlight the novelty of the paper? As the authors employed existing approaches.

Please conduct analysis and provide the ANOVA/Cohen's kappa scores based on the experts rating.

The authors envision employing the proposed TT coaching system will enhance training efficiency, if so can the authors explicitly show the improvement rate before and after the training session?

We encourage the authors to release the data and code during the review process.

There is a lack of related literature comparison as to where this work fits and outperforms SOTA methods? - Please provide a tabular fashion comparison?

Please provide an ablation study or rationale for adopting the proposed pipeline? We found this part of the motivation is lacking throughout the paper?

Discuss the proposed framework's limitations, particularly scalability and real-world deployment, to provide a more balanced perspective on its applicability.

Please provide the feasibility test on how the authors plan to deploy in resource-constrained devices.

Reviewer #3: Minor formatting issues:

1.1 Inconsistent citation format: "physical condition[12,[13]" and "unaffected by external conditions[10,[11]". Please standardize the citation format throughout the paper.

1.2 A reference for OpenPose is needed.

1.3 Figure formatting issues in the main content: In sections 2.1 and 2.2, only figure captions are present without the actual figures.

1.4 “In Figure 6, it can be seen that in an 11-point

match analyzed by the AI table tennis coaching system, a total of 41 points were lost. Among these,

there were 9 errors due to arm movements and racket angle, 7 errors due to hitting position and

slow body transitions, 3 errors due to mistiming of strokes, 1 error due to lagging footwork, 2

errors due to wrist movements, and 3 errors due to high center of gravity.” This sentence should be changed to improve clarity. For example, it should clarify whether arm movements and racket angle errors appeared 9 times each or 9 times in total during the match

2 The authors state that data and code are publicly available without restrictions. However, no direct links to the code and data are provided. Including an anonymous GitHub repository would provide more details about the data and code framework for reproduction purposes and help people to understand the framework.

3 The authors mention a knowledge base that provides the theoretical foundation for MLLMs. It would be beneficial to clarify whether this knowledge base is text-based or video-based. A more detailed explanation or an example would help illustrate this clearly.

4 It would be valuable to compare the performance of state-of-the-art MLLMs on this task. A comparison between the authors' model and SOTA MLLMs would help clarify the contribution of this work.

The paper lacks a clear explanation of how visual recognition, motion capture, and large language models are integrated. How is the video data incorporated into the MLLM models? The authors use pre-trained recognition algorithms to identify players and ball trajectories, but the advantages of this approach are not clearly stated. Does this perform better than directly uploading the video to the MLLMs? A detailed results comparison would help support the authors' claims.

5 In Table 1, the meaning of "Proportion" and "Quantity" is unclear. Does this represent the test subset or the full dataset proposed by the authors? The proportions do not sum to 100%. A better clarification of the dataset would help readers understand the size of the dataset and the proportion used as the test set.

6 In Figure 7, the authors claim "Automated loss analysis, error cause investigation, and recommendations for a table tennis novice in an 11-point match." The AI-generated suggestions seem applicable to most people who make such errors. There is a concern that the generated suggestion content is based on error type rather than personalized movement videos. Is it possible to prompt the model to show different arm angles where the player makes mistakes, since Figure 3 suggests arm angles can be captured? This would make the model's output appear more personalized.

**********

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: Yes: Tatiana Peixoto Sampaio

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.

Attachments

Attachment

Submitted filename: PONE-D-24-31115.docx

Table Tennis Coaching System Based on a Multimodal Large Language Model with a Table Tennis Knowledge Base

PONE-D-24-31115R1

Dear Dr. Li,

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,

Peter Andreas Federolf

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #2: All comments have been addressed

**********

2. 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 #2: Yes

**********

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

Reviewer #2: Yes

**********

4. 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 #2: Yes

**********

5. 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 #2: Yes

**********

6. 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 #2: All my comments are adequately addressed. Thank you so much to the authors for their hard work and and for releasing the codes for the research community .

**********

7. 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 #2: Yes: Indrajeet Ghosh

**********