Designing a Remote Photoplethysmography-Based Heart Rate Estimation Algorithm During a Treadmill Exercise

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

1. The sentence mentions, "Even though such dataset, described in [walking rppg dataset citation number]," should it include a reference link and the date the dataset was downloaded?

2.The text mentions, "Utilizing the coordinate information corresponding to the movement obtained through the aforementioned procedure, a spectrogram in the time-frequency domain is generated in Figure 1. In this spectrogram, the red solid line represents the frequency with the maximum amplitude at each time instance. A comparison with the heart rate label data in Figure 1 reveals a notable similarity between the estimation and the true label." Does the term "spectrogram" actually refer to Figure 2? Additionally, the steps from Figure 1 to Figure 2 are incomplete here, making it difficult for the reader to understand how to implement them.

3. Table 1. Pearson correlation coefficient value between the main frequency components of facial coordinate movements and heart rate labels." In terms of correlation at different speeds, it can be observed that the correlations are all quite high. Why, then, does the average reach as high as 0.74?

4. In the sentence, "The coordinates extracted from each facial image frame were preprocessed using whitening and the computation of first derivatives. Whitening was applied to eliminate correlations among the data dimensions and to standardize the feature set, thereby enhancing the effectiveness of subsequent analyses," it should clarify which specific landmark is being processed.

5.There seems to be an issue with the correspondence of figure numbers in the paper. Could you please verify this in detail?

6.The text states, "Our algorithm processes facial landmark data within a 6-second window, updating this window one frame at a time. To estimate heart rate, the algorithm performs a Fast Fourier Transform (FFT) on the data within the current window and identifies the dominant frequency—the frequency with the highest amplitude." Is this method novel? Or are there any studies that support this approach, requiring citation of relevant references?

7. The article presents a "Total Comparison" and should include references to studies that use other methods for comparison.

Author Response

Thank you for your insightful comments. We’ve prepared our response to the revisions in a Word file and are sending it your way. We appreciate your patience.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

First, I suggest that a native English speaker proofread the paper.

The proposed method should be greatly expanded. Also, present more in-depth the limitations of the proposed method.

Why was the scaling factor set to 0,3?

Did the authors do some feature analysis? Maybe feature importance check?

The rPPG algorithm should be explained in detail.

Provide environment settings info - on which computers the code was run, in which programming language proposed method is written, etc. 

Provide statistical analysis of the results. A better discussion of the results is needed.

In general, the idea is well presented, but it needs further clarification based on the aforementioned comments.

Used references are relevant and up-to-date.

Author Response

Thank you for your insightful comments. We have prepared our response to the revisions in a Word file and are sending it to you. We appreciate your understanding.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

The manuscript electronics-3431316 deals with the methodology to analyze video face data to recognize the hear rate. The manuscript is within the scope of the journal. The topic itself is significant from the point of view of social health and may support the AI assisted recognition of possible health problems based on image processing. Presented original data could be published in the journal, however the manuscript needs additional work to be done. Therefore, I suggest major revision.

The abstract of the manuscript is to general – it does not provide information what is the improvement proposed by the Authors and what is its quantitative effect on the results. Also, the literature review is very general – it does not build a detailed background to understand the difference of the proposed method in relation to other algorithms because other algorithms are not summarized with sufficient details. In my opinion, a short list of references does not provide enough information to understand the state of the art in the area. Additionally, the block diagrams would be very beneficial to clarify the whole research methodology as well as the proposed algorithm. The advantages and disadvantages of the proposed methiod are not sufficiently analyzed.

Selected detailed comments:

Line 114 – Pearson coefficient reported as 0.71, while in table 1 the reported average value is 0.74

Line 146 – why was the value of scaling factor set to 0.3?

Lines 151-153 should be erased.

Lines 1547-158 – Authors state “Our algorithm demonstrated higher accuracy compared to the rPPG algorithm.” It is unclear what was the background of the evaluation. How was its evaluated – is it the difference between rPPG and described algorithm or are the results related to another result?

Line 166 – repetition of “algorithm provides accurate estimates.”

I insist the proofreading – I found many editiorial problems in the manuscript.

Author Response

Thank you very much for your insightful comments. We have carefully prepared our response to the revisions in a Word file and are sending it to you now. We sincerely appreciate your understanding and support.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Authors revised the dradft according reviwer's comments.

Author Response

Revision note

 

1. Authors revised the dradft according reviewer's comments.

 

• Thank you for your feedback. We have carefully revised the draft in accordance with your comments. We appreciate your insights, which have helped us enhance the clarity and quality of our manuscript.

 

 

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The paper is improved and can be accepted.

Author Response

1. The paper is improved and can be accepted.

 

• Thank you for your positive feedback on our paper. We are delighted to hear that you find the revisions satisfactory and that the manuscript can be accepted. Your insights have been invaluable in improving the quality of our work.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

Thank you for revision of the manuscript. In my opinion, it is significantly improved and the content is clarified. However, the current abstract version is too long and contains references, which should be avoided. I suggest to make the abstract more concise.   

Author Response

The original manuscript has been evaluated by the reviewers, requesting consideration of the following contents in summary:

• Revision of the content that includes citation numbers in the abstract
• Revise the content of the abstract more concisely.

Our responses are listed in the attached word file. Thank you again for your consideration of this paper.

Author Response File: Author Response.pdf