Validation of Different Filters for Center of Pressure Measurements by a Cross-Section Study

Round 1

Reviewer 1 Report

This study investigated the methodical comparison between different digital filters, measurement frequencies and times and their effects on the CoP process. Results showed that with different dominant frequencies in the spectrum of the group of test persons certain filter types are required for the processing of CoP data. This is a study assessing frequencies and times and their effects on the CoP process, which is an interesting topic in COP research. The paper is relevant to the readership of Technologies.

However, this manuscript has serious methodological deficiencies.

 

General comments

How did the authors determine the sample appropriate size?

 

Though studies focused on postural sway (COP), however, neurological disorders, balance disorder, and vestibular function disorder is another important contributor to postural control. In this study, no exclusion or inclusion criteria about these function or disorders was stated.

 

 

You did not elaborate on the comment about the correction of the alpha level (0.05) for the number of tests being performed on the data.

 

In analysis, the methodology with regard to the dependent variable is totally opaque to me. In this respect one also wonders why the authors did not choose for an obvious age group × 2 (Eyes open and closed) × the sampling frequency × measurement times (30 s, 60 s and 300 s length) ANOVA instead of the present piecemeal approach.

 

 

Line 326-327: What is the physiological basis for no significant fatigue condition? Discussion in neurophysiological explanation is inadequate.

 

"In comparison with the results of Carpenter et al. and Le Clari, it can be concluded that measurement times of less than 30 seconds are unsuitable for validly recording CoP processes." This is not sufficient.

 

 

The whole discussion provided by the authors is highly speculative since the crucial comparison between groups in conditions sampling frequency is missing.

 

"In addition to the validity, it could also be proven that the reliability also increases significantly with increasing measuring time. " What is the point here?

 

 

Line 469-471: remove "Against this background and on the basis of the results of this study, it can be concluded that the measurement time for a study should not be less than 30s, but can be interpolated in a window between 30s and 120s." As this study was not longitudinal.

Author Response

Thank you very much for yours excellent editing work for our manuscript.

 

We are truly grateful to yours and the reviewers’ good comments and thoughtful suggestions. Based on these comments and suggestions, we have made careful modifications on the original manuscript. All changes made to the text are in red color. “Response to Reviewer and Editor Comments” is attached below. We hope the revised manuscript will meet your journal’s standard.

 

Thanks again for your excellent work.

How did the authors determine the sample appropriate size?

 

For publication, no explicit case number planning was carried out. The analyzed data were taken from a large multicenter study and were collected for a different primary study objective. The analysis presented here is a side result.

The number of subjects should represent a balanced ratio of female to male, as well as a broad age range. No preselection was made and all subjects' results were included

 

Though studies focused on postural sway (COP), however, neurological disorders, balance disorder, and vestibular function disorder is another important contributor to postural control. In this study, no exclusion or inclusion criteria about these function or disorders was stated.

No subjects with corresponding illnesses were included. The method was sharpened by this.

You did not elaborate on the comment about the correction of the alpha level (0.05) for the number of tests being performed on the data.

We have not calculated statitik that requires the alphalevel.

 

In analysis, the methodology with regard to the dependent variable is totally opaque to me. In this respect one also wonders why the authors did not choose for an obvious age group × 2 (Eyes open and closed) × the sampling frequency × measurement times (30 s, 60 s and 300 s length) ANOVA instead of the present piecemeal approach.

The point you mentioned is very interesting. However, this is not part of the question of this publication. This is considered in a separate paper and an anova with correspondingly many variables would go beyond the scope here. 

Line 326-327: What is the physiological basis for no significant fatigue condition? Discussion in neurophysiological explanation is inadequate.

 

This was addressed by van der Kooij [1] and Collins[16] and was therefore not further explored by us. However, a supplementary note has been included in the publication

"In comparison with the results of Carpenter et al. and Le Clari, it can be concluded that measurement times of less than 30 seconds are unsuitable for validly recording CoP processes." This is not sufficient.

The conclusion is based on the results of previous studies (see Carpenter et al. and Le Clari), which have provided reliable evidence. A new proof in the context of this publication was not provided

The whole discussion provided by the authors is highly speculative since the crucial comparison between groups in conditions sampling frequency is missing.

This is an interesting approach. However, this does not correspond to the objective.  Due to the length and the short time for the changes we would want to consider this separately.

 

We would like to comment on the argument that the discussion is speculative. From our point of view, the results are based on a very large database and the deductions made can be substantiated in the principles by other investigations.

 

"In addition to the validity, it could also be proven that the reliability also increases significantly with increasing measuring time. " What is the point here?

We have sharpened the text passage in its statement

 

Line 469-471: remove "Against this background and on the basis of the results of this study, it can be concluded that the measurement time for a study should not be less than 30s, but can be interpolated in a window between 30s and 120s." As this study was not longitudinal.

               That is correct, we have put the wording into concrete terms.

Reviewer 2 Report

The paper presents a wonderful insight into the measurement of CoP and how it should be handled and processed. The findings in this paper are useful and will form a great reference tool for future CoP measurements.

 

Couple of suggestions to improve:

Diagrams/Images to illustrate the method. Make CoP data collected available via an online database. State approving ethics body. Explain how experiments were conducted among the aged population? What safety measure etc. 120 s of data collection - how can authors ensure that data collected is not affected by fatigue?

Author Response

Thank you very much for yours excellent editing work for our manuscript.

 

We are truly grateful to yours and the reviewers’ good comments and thoughtful suggestions. Based on these comments and suggestions, we have made careful modifications on the original manuscript. All changes made to the text are in red color. “Response to Reviewer and Editor Comments” is attached below. We hope the revised manuscript will meet your journal’s standard.

 

Thanks again for your excellent work.

 

The paper presents a wonderful insight into the measurement of CoP and how it should be handled and processed. The findings in this paper are useful and will form a great reference tool for future CoP measurements.

Diagrams/Images to illustrate the method. Make CoP data collected available via an online database.

Thank you for pointing out that we have never practiced anything like this before. We will discuss this and, if necessary, publish the dataset.

 

State approving ethics body. Explain how experiments were conducted among the aged population?

For securing in the one-leg stand (especially with older test participants), a volunteer was located at the side outside the field of vision of the test subjects in order to be able to support them.

 

What safety measure etc. 120 s of data collection - how can authors ensure that data collected is not affected by fatigue?

van der Kooij [1] postulates after their results from long-term measurements that fatigue has only a very small influence on the result. Based on this we have chosen the measurement time of 120s.

A corresponding clarification was made in the Method.

Reviewer 3 Report

The paper is technically sound. The coverage of the topic is sufficiently comprehensive and balanced. The technical depth of the paper is appropriate for the generally knowledgeable individual Working in the Field. 

 

This is my advices to the authors:

In the abstract the authors should tell the news made in the paper.It is not method used, because they write << methodical comparison between different digital filters,  measurement frequencies and times and their effects on the CoP process based on a healthy reference group.>>(see line 23-24 abstract)The news is database e labview. The Authors should describe the database used and LabView in abstract.In input how many channels are there?In the study of measurement time and frequency, are the intervals superimposed?See line 81 <<Subsequent analysis was performed for the defined intervals - 15 s, 30 s, 60 s and 120 s.>>Review the line 142 <<of the one-leg stand>>In figure 5 the legend (<<Kistler eyes open, Kistler eyes closed>>)is  is different from the text << for the right and left standing leg.>> Then review the caption of the figure.

I think a block diagram would be needed to present the steps that the authors will make in the article.

Very good is the Discussion.

The References are old (1996,1983, 2001). From 20013 to 2018 no one published on this? Why?

Author Response

Thank you very much for yours excellent editing work for our manuscript.

 

We are truly grateful to yours and the reviewers’ good comments and thoughtful suggestions. Based on these comments and suggestions, we have made careful modifications on the original manuscript. All changes made to the text are in red color. “Response to Reviewer and Editor Comments” is attached below. We hope the revised manuscript will meet your journal’s standard.

 

Thanks again for your excellent work.

In the abstract the authors should tell the news made in the paper.

It is not method used, because they write << methodical comparison between different digital filters,  measurement frequencies and times and their effects on the CoP process based on a healthy reference group.

>>(see line 23-24 abstract)The news is database e labview. The Authors should describe the database used and LabView in abstract.

We have supplemented the abstract with your recommendation

 In input how many channels are there?

What do you mean? At the A/D converter or the data base? A/D converter used 4 channels (each force sensor)

 In the study of measurement time and frequency, are the intervals superimposed?

No, we measured a 120s track per test band and test condition and took the shorter time intervals from this. This allowed us to dispense with the superimposition of many individual measurements and keep the time burden for the test participants short.

For the evaluation, all individual measurements were then combined according to the questions and analysis procedures. See line 81 <<Subsequent analysis was performed for the defined intervals - 15 s, 30 s, 60 s and 120 s.

>>Review the line 142 <<of the one-leg stand>>

It’s corrected

In figure 5 the legend (<<Kistler eyes open, Kistler eyes closed>>)is is different from the text << for the right and left standing leg.>> Then review the caption of the figure.#

            Thanks for the hint we have adapted the graphic

I think a block diagram would be needed to present the steps that the authors will make in the article.

            this is a good idea but the paper is already very long and such a presentation would take about one page. That's why we've decided to distance ourselves from such a presentation for the time being.

Very good is the Discussion.

            thanks a lot

The References are old (1996,1983, 2001). From 20013 to 2018 no one published on this? Why?

The discussion also includes references from Schuber 2012, Koltermann 2018 and van der Kooij 2011. Our research showed that relatively little has been published on this topic in recent years. Especially the publication of Kapteyn1983 is one of the first on the topic and a much quoted standard we didn't want to miss here.

Round 2

Reviewer 1 Report

The authors have not addressed each comment that was raised after the submission. This manuscript has serious methodological deficiencies.

 

 

"For publication, no explicit case number planning was carried out. The analyzed data were taken from a large multicenter study and were collected for a different primary study objective. The analysis presented here is a side result. " The clarification is insufficient. It looks like the authors used entirely the data from previous study. The issues in the design of the cross-sectional studies are choosing the representative sample, sample size and appropriate data collection. Besides, the cross-sectional studies of side result should refrain from conclusions. Please provide all parameters for the sample size calculation in the Methods.

 

"The point you mentioned is very interesting. However, this is not part of the question of this publication. This is considered in a separate paper and an anova with correspondingly many variables would go beyond the scope here." Statistical reporting is imperfect. The rationale of "considered in a separate paper" is not clear. Please performe all statistical for the ANOVA in the Methods.

 

"This was addressed by van der Kooij [1] and Collins[16] and was therefore not further explored by us. However, a supplementary note has been included in the publication." I am concerned that the results from this study have been over-stated in the discussion and think the language in this section needs to be modified. It was not at all clear what does it mean more specifically and which of the information presented in the manuscript support the explanation.

 

"The conclusion is based on the results of previous studies (see Carpenter et al. and Le Clari), which have provided reliable evidence. A new proof in the context of this publication was not provided." The clarification is insufficient. It is surprise that authors omitted ANOVA analysis of each time period, which seems to be very interesting and include more information and also possible influence of age. The novelty of presented study should be more clarified and explained.

 

"This is an interesting approach. However, this does not correspond to the objective. Due to the length and the short time for the changes we would want to consider this separately." The clarification is insufficient. I would like to suggest including some explanation, why the authors did choose the sampling frequency, and what they expected with respect to age-related changes in balance control.

Author Response

see attached.

Author Response File: Author Response.pdf

Reviewer 3 Report

The paper is technically sound. The coverage of the topic is sufficiently comprehensive and balanced. The technical depth of the paper is appropriate for the generally knowledgeable individual Working in the Field.

In the abstract the authors show the goal of the paper.  The title is good.  English language and style are clear.

Figures and tables have sufficient resolution. They show the results clearly.

The results are properly commented. The bibliography is enough.

The authors have responded well to the auditor's questions and They have agreed the suggested changes.

In this new form the article can be accepted.

Author Response

Thank you!