A Coupling System for Prediction of Physiological Parameters in an Immersed Condition

Round 1

Reviewer 1 Report

I have doubts about the validity of the results and experimental procedure. On the one hand, a study carried out only with men is today more than questionable, the gender perspective must be taken into account in research, in all areas, and even more so when the bias that is being committed is so evident. Fundamental questions about the experimental methodology are not specified, for example, has been carried out any control test?, or have tests been repeated?, how many times? In experimental research repeatability study is vital. It is not analyzed in this study.

It is difficult for me to identify the contribution or novelty of this study, since the thermal model is already defined, reference 18, and if I have not misunderstood, the achievement of experimental measurements to contrast the model is presented as the greatest contribution or innovation. Leaving aside the fact that, in my opinion, the study is small in terms of the size of the sample analyzed, and that issues in the methodology followed need to be clarified in order to validate the measures, it is easy to identify that the experimental trials, with the same 10 volunteers, have already been used and previously published in another papers. This article does not have anything new that can be considered as a different study, therefore, in my opinion there is no relevant contribution, and no new analysis or measurements are made, so I consider that it should not be published.

References from previous studies by the same work team can be seen in references 6, 10, 18, and 29. References 10 and 18 of the manuscript are duplicates, it is the same reference. Analyzing:

https://doi.org/10.1016/j.ijthermalsci.2023.108364

https://doi.org/10.1016/j.ijthermalsci.2022.108029 

It is very easy to see that practically all of the images, explanations, tables and data presented in this study have already been used in recently published publications, I see practically nothing new with respect to these articles. To the eyes of the reviewer, the contribution of the paper is minimal. There is nothing new in the methodology used, nor in the type of results obtained, nor in the analysis performed. The techniques used by the authors and analysis are known and previously published studies. Moreover, the few novel contributions in the manuscript are vague and inaccurate, manuscript needs significant extra work.

 

Author Response

Replies to academic Reviewer:

We would like to express our sincere thanks to the reviewers for the constructive and positive comments.

1. I have doubts about the validity of the results and experimental procedure. On the one hand, a study carried out only with men is today more than questionable, the gender perspective must be taken into account in research, in all areas, and even more so when the bias that is being committed is so evident. Fundamental questions about the experimental methodology are not specified, for example, has been carried out any control test?, or have tests been repeated?, how many times? In experimental research repeatability study is vital. It is not analyzed in this study.

Thanks for the comment of the reviewer.

I am glad to answer your question. At the early stage of the experiment, we had considered inviting female subjects to participate in the experiment, but due to the special nature of the experiment, the subjects not only need to wear only underwear and life jackets, but also need to be immersed in low-temperature water for more than 40 minutes, which was canceled at the later stage of the experiment in view of the inconvenient factors for female subjects in the immersion conditions. In addition, the thermal manikin "Walter", which was used as the control for the experiment, was developed with reference to Chinese adult males, and has all the physiological signs of males, so the use of all male subjects is more in line with the prediction effect of the thermal manikin "Walter" studied in this paper. The coupling system is more consistent with the prediction effect of the thermal manikin "Walter" in this study.

The study also explored the reproducibility of the experiments, for example, each subject was required to wear two different styles of life jackets in the experiment, the two experiments were used as a control group, and the results of the two experiments did not have any significant differences before being analyzed later. In addition, in order to minimize the error of a single experiment, each subject was required to conduct a pre-experiment in a climatic chamber with the same temperature environment. The experiment was started at 8:00 a.m. in the morning of each day.

2. References from previous studies by the same work team can be seen in references 6, 10, 18, and 29. References 10 and 18 of the manuscript are duplicates, it is the same reference. Analyzing:

 

https://doi.org/10.1016/j.ijthermalsci.2023.108364

 

https://doi.org/10.1016/j.ijthermalsci.2022.108029

Thanks for the comment of the reviewer.

Duplicate references have been deleted in the revised manuscript.

 

3. It is very easy to see that practically all of the images, explanations, tables and data presented in this study have already been used in recently published publications, I see practically nothing new with respect to these articles. To the eyes of the reviewer, the contribution of the paper is minimal. There is nothing new in the methodology used, nor in the type of results obtained, nor in the analysis performed. The techniques used by the authors and analysis are known and previously published studies. Moreover, the few novel contributions in the manuscript are vague and inaccurate, manuscript needs significant extra work.

Thanks for the comment of the reviewer.

In this paper, although the same subjects were used for the control, the purpose of this study is completely different from the previous publications. The main purpose of this study is to develop a coupling system based on the thermal manikin "Walter" to predict human physiological parameters underwater, which is essentially a functional upgrade of the thermal manikin "Walter". The main objective is to develop a coupling system based on the thermal manikin "Walter" to predict human physiological parameters underwater. Because the original design of the thermal manikin can only reflect the heat transfer from the body to the environment in a set situation, even if the environment changes, the output power of the manikin always stays the same. In water, which is a more complex environment than in air, the redesigned thermal manikin was able to simulate the physiological response of the human body in the water environment and adjust the heating power to maximize the reflection of the human body's physiological response in water.

Yang et al. also developed a coupling system based on the thermal manikin “Newton”, which can be applied to the prediction of human physiological parameters in high temperature environments. Based on this, we established to develop a coupled system applied in water immersion environment, I used the same idea with Yang, but there are differences in the design principle, and the final results are better than Yang's prediction.

In the revised manuscript, the published figures, descriptions, and tables are reworked, and the novelty of this paper is re-emphasized.

 

References

[1] J. Yang, W. Weng, M. Fu. A coupling system to predict the core and skin temperatures of human wearing protective clothing in hot environments. Appl. Ergon. 51 (2015) 363–369.

Author Response File: Author Response.pdf

Reviewer 2 Report

Summary of the Work

This work aims to study the performance of thermal manikin and a multi-segmented human thermal model capable of predicting human physiological parameters in heat in immersion conditions. The authors analyzed two life preservers with intrinsic thermal resistances between 0.09 and 0.41 °C·m2/W. The measured heat production, core temperature, and skin temperature data were successively compared with the predicted results.

Main Results Obtained

Based on statistical comparison analysis and paired-sample T-tests, the author determined the accuracy and dependability of the manikin coupling system and found that the coupling system predicts heat production, core temperature, average skin temperature, and, to a lesser extent, local skin temperature.

 

General Considerations

- Please specify all the acronyms when they appear first in the manuscript (e.g., in the Abstract, please specify RMSD = Root Mean Square Deviation, e SPSS = Statistical Package for the Social Sciences, etc.)

- It is not clear the definition T_predict. According to the author, T_predict is the “predicted temperature of the coupling system”. However, T_predict does not come from a model but it also comes from measurements on thermal manikin. Please clarify the definition.

- The statistical analysis carried out has not been exhaustively illustrated.

- Paired-sample t-tests are a useful tool for comparing two related groups, but this methodology has limitations. The author did not mention and justify the underlying assumptions met before interpreting the results (see the suggestions below).

- The author did not discuss the advantages of adopting the paired-sample T-tests compared with other alternative statistical tests (see the suggestions below).

- There are other gaps that need to be filled (see the section "Suggestions" below).

 

Suggestions

1) The author mentioned that the predicted results for heat production, core temperatures, skin temperature (head, back upper arm, etc.), and measured data are in agreement with a significance level of p=0.05. However, we know that when interpreting the results of statistical tests and p-values, it's important to consider a range of statistical parameters and measures to assess the reliability and correctness of the results. For completeness, the author is asked to specify the confidence intervals i.e., the range of plausible values for the parameter of interest (e.g., mean difference, odds ratio) along with an associated confidence level.

2) The author used the paired sample T-test to determine if the correlation is statistically significant (i.e., if the p-value is below your chosen significance level). However, usually, we have also to compute the correlation coefficient to quantify the strength of a relationship and then use a hypothesis test. What are the values of the correlation functions found by the author in the experiments shown in Figures 6 to Figures 9?

3) Another important statistical measure R-squared is the value (R²). I would have expected to have also the values of R-squared as it assesses the overall goodness of fit of the regression model and indicates how well the model explains the variance in the dependent variable (while p-values assess the significance of individual predictor variables in the model). Please, provide the values of R-squared, at least for data reported in Figure 6 and Figure 8.

4) We know that SPSS is a widely used statistical software program for data analysis and statistical modeling. Although SPSS is a powerful tool for many analytical tasks, it has some limitations. When dealing with big data or complex data structures we may find that SPSS has performance limitations. In this case, the thermal manikin is used to simulate the thermoregulatory response of the human body in an immersion environment. This requires, in principle, the acquisition and the manipulation of big data. The author recommends using this software to further investigate the proposed coupling system, but we know that SPSS may have serious difficulties with very large datasets, both in terms of data import and processing speed. The author is invited to discuss this important aspect.

5) Has the author used ANOVA in SPSS? Indeed, performing ANOVA as part of the data analysis workflow offers several advantages (e.g., extensive data management capabilities, integration with other analyses, etc.). If yes, have you checked the ANOVA assumptions, such as normality and homogeneity of variances? It can generate diagnostic plots and conduct statistical tests to assess the validity of these assumptions. Note that SPSS provides tools for checking ANOVA assumptions and conducting statistical tests to assess the validity of these assumptions. The author is invited to discuss this important point.

 

Concluding Remarks

The work is interesting and current. However, the most vulnerable aspect is the statistical analysis. The work does not report the values of the other fundamental parameters necessary to validate the final author's assertion according to which the results obtained for the coupling system are in agreement with the measured ones, showing good accuracy in predicting the local skin temperatures. For clarity, the author is also invited to provide his/her comments regarding the observations reported in points 4) and 5) above.

I detected a limited number of typos. However, it is advisable to double-check.

Author Response

Replies to academic Reviewer:

We would like to express our sincere thanks to the reviewers for the constructive and positive comments. Thank you very much for your comments on the statistical analysis of the paper, the questions you asked were very specialized and we responded to as many of the relevant suggestions you made as we could, please forgive us if there are any errors.

1. Please specify all the acronyms when they appear first in the manuscript (e.g., in the Abstract, please specify RMSD = Root Mean Square Deviation, e SPSS = Statistical Package for the Social Sciences, etc.)

Thanks for the comment of the reviewer.

Acronyms have been explained in the revised manuscript.

 

2. It is not clear the definition Tpredict. According to the author, Tpredict is the “predicted temperature of the coupling system”. However, Tpredict does not come from a model, but it also comes from measurements on thermal manikin. Please clarify the definition.

Thanks for the comment of the reviewer.

“T_predict is the predicted temperature of the coupling system” was changed to “T_predict is the simulated temperature of the thermal manikin.” 

 

3. The author mentioned that the predicted results for heat production, core temperatures, skin temperature (head, back upper arm, etc.), and measured data are in agreement with a significance level of p=0.05. However, we know that when interpreting the results of statistical tests and p-values, it's important to consider a range of statistical parameters and measures to assess the reliability and correctness of the results. For completeness, the author is asked to specify the confidence intervals i.e., the range of plausible values for the parameter of interest (e.g., mean difference, odds ratio) along with an associated confidence level.

Thanks for the comment of the reviewer.

Based on the experimental sample size, the study used a 95% confidence level with confidence intervals within two standard errors of the overall parameter values, and in the statistical analysis, it was found that the predicted values of the experimental measurements of the subjects' heat production, core temperature, and mean skin temperature fell within the 95% confidence intervals, and thus the fluctuations of the measurements were eventually shown as error bars (standard deviation SD) on the graph.

 

4. The author used the paired sample T-test to determine if the correlation is statistically significant (i.e., if the p-value is below your chosen significance level). However, usually, we have also to compute the correlation coefficient to quantify the strength of a relationship and then use a hypothesis test. What are the values of the correlation functions found by the author in the experiments shown in Figures 6 to Figures 9?

Thanks for the comment of the reviewer.

We are sorry that correlation functions were not calculated in our statistical analysis, this is due to the fact that correlation coefficients (R²) are used in most of the studies in the reference to characterize the correlation between predicted and experimental values, therefore the R² values for all predicted and experimental values have been supplemented in the revised manuscript.

 

5. Another important statistical measure R-squared is the value (R²). I would have expected to have also the values of R-squared as it assesses the overall goodness of fit of the regression model and indicates how well the model explains the variance in the dependent variable (while p-values assess the significance of individual predictor variables in the model). Please, provide the values of R-squared, at least for data reported in Figure 6 and Figure 8.

Thanks for the comment of the reviewer.

In this paper, the correlation between predicted and experimental values was statistically analyzed using the paired R² test of SPSS, which showed that the predicted and experimental correlation coefficients R² of heat production LP1 and LP2 were 0.952 and 0.946, the predicted and experimental correlation coefficients R² of core temperatures LP1 and LP2 were 0.984 and 0.993, and the predicted and experimental correlation coefficients R² of mean skin temperatures LP1 and LP2 were 0.974 and 0.985 (i.e., the closer the R² is to 1, the stronger the correlation), which according to the R² test can show that the predicted and experimental values have a high correlation.

 

6. We know that SPSS is a widely used statistical software program for data analysis and statistical modeling. Although SPSS is a powerful tool for many analytical tasks, it has some limitations. When dealing with big data or complex data structures we may find that SPSS has performance limitations. In this case, the thermal manikin is used to simulate the thermoregulatory response of the human body in an immersion environment. This requires, in principle, the acquisition and the manipulation of big data. The author recommends using this software to further investigate the proposed coupling system, but we know that SPSS may have serious difficulties with very large datasets, both in terms of data import and processing speed. The author is invited to discuss this important aspect.

Thanks for the comment of the reviewer.

In this paper, the paired-samples t-test of SPSS was used for statistical analysis, and the sample size for statistical analysis was only 60 per group (i.e., one set of data was obtained every minute during the one-hour period of the experiment.) Sixty samples, while not a small sample, is not a particularly large amount of computational effort for SPSS.

 

7. Has the author used ANOVA in SPSS? Indeed, performing ANOVA as part of the data analysis workflow offers several advantages (e.g., extensive data management capabilities, integration with other analyses, etc.). If yes, have you checked the ANOVA assumptions, such as normality and homogeneity of variances? It can generate diagnostic plots and conduct statistical tests to assess the validity of these assumptions. Note that SPSS provides tools for checking ANOVA assumptions and conducting statistical tests to assess the validity of these assumptions. The author is invited to discuss this important point.

Thanks for the comment of the reviewer.

This study did not use ANOVA, which is mainly because ANOVA is mainly used to compare whether there is a significant difference in the means between three and more groups of data, while t-test is mainly used to compare whether there is a significant difference in the means between two groups of data and is applicable to continuous data. According to the statistical analysis methods of previous related research papers, the study used the more appropriate paired samples t-test for analysis.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

I appreciate your response, and the effort dedicated,  but I consider that no significant changes have been made with respect to the previous version, in any of the comments provided, so I cannot modify my previous assessment.

Author Response

Detailed Response to the Reviewers：

Based on your comments and suggestions we have overhauled the structure and content of the article, mainly from the following points:

(1) In the Methods section, we have re-altered the structure, highlighted the thermal manikin experimental part, deleted part of the subject experiments, and introduced the principle and process of the coupling system in detail.

(2) In the Discussion section, we have rewritten most of the content, and the text focuses more on the discussion of the coupled system, analyzing the applicability and limitations of the coupled model. The discussion of physiological experiments on subjects was deleted.

(3) In the Conclusions section, the conclusions have been rewritten to emphasize the results and innovation of the experiment.

(4) Repeated images and text in the article were also redrawn and redacted.

 

In addition, please allow me to restate the content of the article, this study establishes a coupled system based on the human body heat model through a large number of thermal manikin experiments, the significance of this study is to improve the applicability of the human body heat model, in the human body thermal model, the heat exchange between the human body and the environment is computed by empirical modeling, which is unable to be simulated in the face of the complex environment. The thermal manikin acts as a heat flow sensor in the system, which can accurately measure the heat exchange between the human body and the environment, based on which the calculation of the human thermal model is more accurate and reliable. Warm body dummies cannot form thermal equilibrium in a strong convective environment such as water. Only under the control of the human body thermal model, the thermal manikin is able to actively regulate and thus form thermal equilibrium. The coupled system makes up for the defects of both and also utilizes the advantages of both, which is an important tool for future research on the physiological parameters of the human body in water.

 

Author Response File: Author Response.pdf

Reviewer 2 Report

The authors have answered all the questions raised in my previous report. I consider the results they have obtained promising. I therefore encourage the authors to continue their research on this matter.

Author Response

Thank you very much for your comments and suggestions, and good luck with your work!

Round 3

Reviewer 1 Report

First of all, I would like to acknowledge the effort made by the authors in the detailed and extensive answers, and the effort to include some modifications and clarifications in the document. I consider that the quality of the paper has been improved, and now it could be published.