AI-Enhanced Analysis to Investigate the Feasibility of EMG Signals for Prosthetic Hand Force Control Incorporating Anthropometric Measures
Round 1
Reviewer 1 Report
Comments and Suggestions for Authorsthanks a lot for the opportunity to score this work. the work is interesting although the results are expected. Overall the major flaw of this work is the scarse adherence to classical canons for wrtiting a scientific article. Second, the reader does not catch which is the real novelty of data obtained. I think that a large work should be done to shorten and refine the work in some parts, avoiding speculations and focusing on fundamental data and results.
Major:
1) The introduction should be shortened on key terms adopted (from line 117 to 152).
2) The introduction should be shortened on contribution of the study (this part is more appropriate in the discussion section) line 177-199
3) Methods: overall this section needs editing and adjutement to canonical description as generally reported in scientific papers.
4) It is not clear if a standardized method has been used to position the EMG electrodes
5) Why hand grip was not performed in standardized canonical position?
6) Discussion lacks of citations; in the current form it appears as a mere speculation
Conclusions should be shortened. Now this paragraph is redundant.
Author Response
Point-by-point response to Comments and Suggestions for Authors (response sheet pdf and revised manuscript are attached)
Authors would like to thank the Editor for the timely review and the reviewers for their positive comments and constructive feedback on the manuscript. Authors understand and respect the sincere efforts and time devoted by reviewers to provide precise and succinct comments. Authors have carefully reviewed the suggestions and tried our best to address all the concerns. Authors found that these comments have helped us to improve further and strengthen the manuscript. Point by point response for each comment is provided below:
Comment: thanks a lot for the opportunity to score this work. the work is interesting although the results are expected. Overall the major flaw of this work is the scarse adherence to classical canons for wrtiting a scientific article. Second, the reader does not catch which is the real novelty of data obtained. I think that a large work should be done to shorten and refine the work in some parts, avoiding speculations and focusing on fundamental data and results.
Response: Thank you for your valuable feedback and for reviewing our work. Authors have carefully revised the manuscript to address your concerns. The following key improvements have been made:
- Adherence to Scientific Writing Standards: The manuscript has been thoroughly refined to align with classical scientific writing principles. Authors have reorganized sections for clarity, ensured precise language, and removed speculative content to maintain focus on core findings.
- Highlighting Novelty: The revised manuscript now clearly emphasizes the novelty of our work, particularly the integration of physiological factors such as body composition into EMG-based force control models and the robust application of XGBoost for precise feature analysis.
- Conciseness and Focus: Authors have streamlined the content by shortening redundant sections and focusing on the fundamental data, results, and their implications. This ensures a concise presentation without compromising the scientific rigor.
Authors believe these revisions enhance the quality and readability of the manuscript, addressing the issues you raised. Thank you again for your constructive input.
Comments 1: The introduction should be shortened on key terms adopted (from line 117 to 152).
Response 1 : Thanks for your suggestions. The section on key terms has been streamlined to focus on the essential information. The descriptions of MVC, muscle palpation, and BIA have been made more concise, offering the reader a clear and brief introduction to each term. Now the information is short and is in the single paragraph.
Change in the updated manuscript – (Section 1, Line 119-135, on page 3 and page 4)
Comments 2: The introduction should be shortened on contribution of the study (this part is more appropriate in the discussion section) line 177-199.
Response 2: Thank you for your valuable feedback. Based on your suggestion, authors have revised the introduction to focus more on the study's primary objectives. Earlier the information was repeated and given in two long paragraphs. Now authors have shortened the section in single paragraph as bullet points in the revised manuscript that describes the contributions and moved the more detailed discussion to the appropriate section, which authors believe better aligns with the structure of the paper. Authors hope this revision makes the content clearer and more concise for the reader.
Change in the updated manuscript – (Line 136-156). It is now combined with main contribution part and explanation of key terms for better readability and clear understanding.
Comments 3: Methods: overall this section needs editing and adjustment to canonical description as generally reported in scientific papers.?
Response 3: Thank you for the observation. Authors have thoroughly reviewed and revised the Methods section to ensure that it adheres to the canonical description generally reported in scientific papers. Authors have made changes to improve the clarity, concision, and organization of the section. Authors have associated the text with the research work, defined parameters, added machine-learning as separate sub-chapter. Authors believe that it now meets the standard expected in scientific publications.
.
Change in the updated manuscript – (Line in red-colour). The text is also consisted at various places.
Comments 4: It is not clear if a standardized method has been used to position the EMG electrodes
Response 4: The placement of the EMG electrodes followed a standardized approach as outlined by the SENIAM guidelines, which specify key aspects such as electrode type, precise placement, orientation, and the distance between the centres of the bipolar electrodes (2 cm in this case for the muscle group). This adherence to SENIAM guidelines ensures reproducibility and accuracy in EMG signal capture. For accurate localization of the flexor digitorum superficialis muscle, the muscle palpation technique, commonly used in medical practice, was employed. This method was carefully studied and validated to ensure correct electrode placement.
Change in the updated manuscript – Details of the muscle palpation technique are provided in both the Methodology and Introduction sections (Lines – (122-125), (253-255), (254-257)), with references to the SENIAM guidelines and palpation methodology included in the paper.
Comments 5: Why hand grip was not performed in standardized canonical position?
Response 5: In this study, the participant's hand was positioned in a fully extended, relaxed posture to minimize muscle involvement and reduce interference from muscles not directly engaged in the gripping task. This position was chosen to ensure that only the muscles involved in the grip force were actively engaged, providing a more accurate measurement of grip strength. The standing position was selected to further minimize any activation of upper body muscles, keeping the forearm and hand in a neutral state.
Authors also considered the reproducibility of this position across subjects. The extended hand position was standardized for all participants, with clear guidance provided to ensure proper placement. Variations in arm's length or wrist mobility were accounted for by instructing subjects to maintain a relaxed posture and avoid compensatory muscle activation.
While this position may differ from the typical, standardized grip posture used in some grip force experiments, it was chosen intentionally to minimize unwanted muscle activation, which was a key goal of the study. This methodology allowed us to isolate the specific muscles involved in grip force without introducing complications from stabilizing or unrelated muscle actions.
Given the context of this experiment, the decision to use this position was a deliberate one aimed at minimizing muscle interference. However, authors recognize that future studies could investigate whether a more standardized hand position might yield comparable or even enhanced results, particularly when considering generalizability to real-world applications.
Change in the revised manuscript – (Line 274-279). The following information has been added.
To ensure grip strength was primarily attributed to the muscles directly involved in the task, a fully extended, relaxed hand position was adopted to minimize interference from non-target muscles. Standardizing the standing position, with the hand and forearm in a neutral orientation, helped minimize variations due to individual body differences. While this posture deviates from more common grip positions, it was chosen specifically to minimize extraneous muscle activation.
Comments 6: Discussion lacks of citations; in the current form it appears as a mere speculation
Response 6: Thank you for highlighting this area for improvement. Authors appreciate the suggestion to enhance the discussion section by incorporating additional citations, as it will lend further support to our findings and contextualize them within existing literature. Authors have now added references to various studies that address EMG-based prosthetic control, the impact of body composition on EMG signal variability, and EMG applicability across diverse user populations. Authors hope these additions provide helpful context and alignment with current research in prosthetic control.
Change in the updated manuscript – (Line 543-588).
Comments 7: Conclusions should be shortened. Now this paragraph is redundant
Response 7: Thank you for your comment. In response to your suggestion, authors have shortened the conclusion to make it more concise while retaining the key findings of the study. Authors believe this revision better aligns with the overall focus of the paper and ensures clarity for the reader. The updated conclusion now highlights the core outcomes without unnecessary detail, and authors hope it provides a clearer summary of the study's contributions. Authors appreciate your input, as it has helped us streamline the paper and enhance its readability.
Change in the updated manuscript – (Line 589-601).
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsDear colleagues, please find the comments in the attached document.
Good luck!
Comments for author File: Comments.pdf
The English could be improved to more clearly express the research
Author Response
Point-by-point response to Comments and Suggestions for Authors (response sheet pdf and revised manuscript are attached)
Authors would like to thank the Editor for the timely review and the reviewers for their positive comments and constructive feedback on the manuscript. Authors understand and respect the sincere efforts and time devoted by reviewers to provide precise and succinct comments. Authors have carefully reviewed the suggestions and tried our best to address all the concerns. Authors found that these comments have helped us to improve further and strengthen the manuscript. Point by point response for each comment is provided below:
Comments 1: Please pay attention to the following typing issues:
- the way you have inserted the citations in the text
- lines 116-117, I think it is a typo
- from page 6, line 240, you have changed the font size
Response 1 (a): Thank you for pointing this out. The citation style has been updated to align with the journal's guidelines, ensuring clarity and consistency throughout.
Response 1 (b): Thanks for your suggestions. The typo on lines 116-118 has been corrected for better readability and to avoid confusion.
Change in the updated manuscript – (Line 117-119). Sentence has been updated as “Some Key terms like Maximum Voluntary Contraction, Muscle palpation, Bioelectrical Impedance Analysis (BIA), XGBoost used in this study are explained below to make it accessible and familiar for an adequate understanding of the study conducted.”
Response 1 (c): Thanks for your suggestions. The font size discrepancy on page 6, line 240, was due to an unintentional adjustment to 9pt. This has been corrected, and the text now adheres to the journal's standard 10pt size. Please refer lines 200 onwards in Section 2,2.
Comments 2: Consider restructuring the paragraph between lines 162 - 177. A list format might make it easier for readers to navigate and understand the content.
Response 2: Thank you for the suggestion. The paragraph has been restructured into bullet points to enhance readability and improve the flow of information, making it easier for readers to follow and understand the content.
Change in the updated manuscript – (Lines 136-156) in Section 1. It is now combined with main contribution part and explanation of key terms for better readability and clear understanding.
Comments 3: Page 6, 2. Materials and Methods - over what period was the study conducted?
Response 3: Thank you for the observation. The timeframe of the study has now been clarified in the manuscript, specifying that the study was conducted between January 2023 and May 2024.
Change in the updated manuscript – (Line 201 in Section 2.2). Above information has been added.
Comments 4: The paragraph between lines 263 - 279—It would be helpful if you could list the information so that the reader can more easily skim through it.
Response 4: Thank you for the suggestion. The paragraph has been restructured into bullet points to enhance readability and improve the flow of information. This adjustment aligns with a similar change made earlier, following the reviewer's helpful suggestion to present key information in a format that readers can easily skim and understand.
Change in the revised manuscript – (Line 228-247 in Section 2, Page 7)
Comments 5: Line 280: "First, the Measurement of the body parameters of the subject...." - What do you want to convey with this sentence? Perhaps you could rephrase it.
Response 5: Thank you for the comment. The intent was to outline the procedural steps in the methodology by using terms like 'firstly' and 'secondly' for each stage. However, as you rightly noted, the original phrasing was somewhat unclear. This sentence has been revised to enhance clarity, aiming to make the procedural steps more straightforward and accessible for the reader.
Change in the updated manuscript – (Line 248-249). The sentence is revised as “The initial step in the procedure involves the acquisition of body parameter measurements.”
Comments 6: Line 301:" Then 301 the subject gets in the standing position with their arm straight and at a neutral position" - a good idea would be, somewhere in the study, to specify what this "neutral position" is. The text is not very clear.
Response 6: Thank you for your suggestion. The neutral position refers to the subject assuming a standing position with their arm fully extended, aiming for a neutral orientation with minimal flexion, rotation (including pronation and supination), and deviation of the hand and wrist. Efforts were made to minimize any such movements to maintain a consistent position, ensuring the best possible accuracy in the measurements. This revision has been incorporated into the manuscript to provide the reader with a clearer understanding of the neutral position.
Change in the updated manuscript – (Line 272-275 on Page 8). The sentence is revised as, “Then the subject assumes a standing position with their arm fully extended, aiming for a neutral orientation with minimal flexion, rotation (including pronation and supination), or deviation of the hand and wrist.”
Comments 7: Page 9, table 1. Why did you find it necessary to make a mean and a standard deviation? You have a study with 37 people. I understand that you measure the strength of MVC, 50% of MVC, and 25% of MVC, but with what data do you average? The same observation is for Table 2. When such data is collected, it explains why you choose to represent by mean. Why is standard deviation needed? You have not pursued an interpretation from a statistical point of view.
Response 7: Thank you for raising this important point. In this study, data was collected 15 times per participant—5 repetitions each at 100%, 50%, and 25% of maximum voluntary contraction (MVC). While all 15 data points per subject were used in the analysis to ensure statistical rigor, the table presents only a summary in the form of mean and standard deviation for each MVC level to maintain readability. To summarize this extensive data set in a readable format, Authors presented the mean and standard deviation of each MVC level across the 5 repetitions in Tables 1 and 2. This approach provides a representative snapshot for readers without overwhelming them with raw data points, which would be challenging to display in a table format. The mean provides a representative trend for each level, while standard deviation is essential to show the natural variability across the repetitions, offering insight into the stability of each level’s data.
Now authors have combined Table 1 and Table 2 to make it compact and save space. Thanks for your feedback.
Please refer Table in revised manuscript on Page 9.
Comments 8: Page 11, The machine learning model - try to dedicate a sub-chapter to this development because, from your title, you should highlight your results based on this AI model. It would also be better to describe the information in lines 379-389. In those lines, some information is given in a general way, but not the information related to the algorithm you used for this experiment.
Response 8: Thank you for your suggestion. Authors now dedicated sub-chapter 2.3 to the development of the machine learning model. Authors have also revised the manuscript to include a more detailed description of the machine learning model, including the specific algorithm used, its parameters, and the training process.
Regarding the information in lines 379-389, authors have provided more specific details about the algorithm used in the experiment, including relevant hyperparameters. Authors hope that the information is presented in a clear and concise manner, allowing readers to fully understand the methodology used.
Change in the updated manuscript – (Lines, 310 and 355-369). Refer new sub-chapter 2.3 “Machine Learning-based Analysis to Investigate the Feasibility of EMG Signals for Pros-thetic Hand Force Control”
Comments 9: Subchapter 3.1 EMG vs. Force. You have some correlation graphs. Did you calculate the correlation coefficient? Moreover, if you have and if not, it would be good to insert intext to validate the interpretation of the data.
Response 9: Thanks for the suggestion. Now correlation coefficient is added to figure 8 for different force levels. Also, text is added to describe the correlation coefficient, so that information regarding the interpretation should be clear.
Change in the updated manuscript – (Line 375-383) and Fig 8(a) on Page 12.
Comments 10: Page 14, line 414 - where are those 1000 cases from? It is not clear what type of cases they are. Are there cases in which you measured MCV and EMG on the same person or all persons, or are they learning cases of the algorithm?
Response 10: Thank you for the comment. The effectiveness of the model was evaluated using Repeated-K-Fold cross-validation, where the available data was partitioned into training and validation sets. This process was iterated 1000 times, with the data automatically split into 5 folds and the model trained and evaluated at each iteration.
Change in the updated manuscript – (Line 393-399 on page 13).
Comments 11: Page 14, line 424: "...Studies indicate that when..." - What studies are Authors talking about?
Response 11: Thank you for highlighting this. The studies referenced here are validated research works discussed within the context of our study, which establish the relationship between body fat, muscle mass, and surface EMG (sEMG) signal quality. These sources are now cited in the revised manuscript to substantiate our approach in comparing body parameters—such as fat level and muscle mass—to EMG values, supporting our analysis of data reliability and the influence of physiological factors on sEMG measurement
Change in the updated manuscript – (Line 407-414 on Page 14). The sentence is revised as follows:
“Studies indicate that when recording signals from individuals with high body fat levels, the distance between the muscle and the surface electrodes attached to the skin increases, leading to a decrease in EMG values[45–48].”
Comments 12: On page 15, you have included graphs 10 and 11. A statistician can interpret what is there, but a non-statistically literate reader will not realise the correlations between the data. To strengthen the results, it would be good to introduce the calculated correlation coefficients.
Response 12: Thanks for the suggestion. Now correlation coefficient is added to each of the sub-figures 10 and 11. Also, text is also added to describe the correlation coefficient, so that information regarding the interpretation should be clear.
Change in the updated manuscript – (Line 421-437 on page 14 and 15).
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Comments and Suggestions for AuthorsThe work is now acceptable
Reviewer 2 Report
Comments and Suggestions for AuthorsThe article looks fine now.
