Is the Neuromuscular Organization of Throwing Unchanged in Virtual Reality? Implications for Upper Limb Rehabilitation

Round 1

Reviewer 1 Report

Comparing real movements and virtual-reality ones of humans is valuable to enable virtual-reality rehabilitation/training methods for muscle recovery, as described by authors. This work specifically investigates and proves the similarity between real and virtual-reality throwing movements. The experiment design, data processing/organization, and manuscript content organization are reasonable to show the conclusion. I think that this work can be considered for publishing.   

Author Response

Please see the attached file.

Author Response File: Author Response.pdf

Reviewer 2 Report

This manuscript presents an experiment to compare virtual reality and reality in terms of neuromuscular organization during upper limb rehabilitation. The experiment is explained clearly. Most of the paper is focused on data analysis. The results show similarity between real and virtual rehabilitation activities. 

The researchers did not consider the effect of the weight of the ball on the muscle activities. Will the result be different when comparing throwing a real heavy ball with a virtual ball? As discussed in section 4 and section 1, some related works have been done. Can the researchers explain why the results do not agree with [27] but do agree with [55]? Some content in the discussion section 4 can probably be moved to section 1 to stress the novelty of the paper and make the discussion more coherent. A lot of statistical models and skills are used in section 2.3, 2.4, 2.5, and section 3, but the presentation needs to be improved to make it easier to follow. What is the muscle synergy? What is the number of muscle synergies? What is an overall model? In equation (1), is EMG the 12 by 20,000 matrix? If m is the number of muscles, shouldn't it be 11? It is unclear how to compute Wi and Ci. Also, what is the reconstructed EMG matrix? What is Pearson's coefficient? What is NNMF? What is Shapiro-Wilk test? What is ANOVAs? What is Bonferroni's test? It is unclear how to get the D, ND, and Mode values in Tables 1 to 4? What do they stand for? It is highly recommended to add some qualitative explanation after each section of statistical analysis.

Author Response

Please see the attached file.

Author Response File: Author Response.pdf

Reviewer 3 Report

Authors have undertaken an interesting problem of comparing the throwing task that is performed in real and virtual environment. In particular the Authors concentrate mainly on the measurement of the associated muscle activity by measuring upper limb EMG signals. Possible applications are: rehabilitation and sports.

The number of point I would like to raise are the following:

1) Throughout the text the term “muscle synergy” is recalled many times, whereas this term is just shortly explained in 3-4. For ‘Electronics’ journal readers this may be a vague concept without any knowledge why this is used and why this is important. I suggest to include a simple schematic explaining this concept (as in some of the cited references). In particular time variant and time-invariant synergies should be better defined.

2) I would expect also a better justification for the use of the Non-Negative Matrix Factorisation algorithm. This algorithm is just mentioned in the paragraph preceding Eq. (1) without any rationale why exactly this algorithm is the preferred computational method for analysis of EMGs in particular. Also, it is not obvious why Wi terms should be time independent and Ci are time dependent and perhaps their relation to muscle synergies be better explained. More elaboration on that is required.

3) An alternative approach might be discussed in which kinematics derived e.g. from inertial sensors might be also inclused in the study, theese are much simpler to implement than many electrodes that need to be glued to the skin

Detailed comments:

What do authors mean by: “Subsequently, each EMG signal was resampled at 1,000 samples”, i.e. was the sampling rate increased or decreased or was different for each recording? Please be more specific. What is the size and weight of the ball? Probably the weight should be very small … In caption of Fig. 3 instead of ‘probes’ I would rather use ’electrodes’ English needs to be corrected in a number of text fragments, e.g. : “…have been already been demonstrated…” (line 346)

 

 

Author Response

Please see the attached file.

Author Response File: Author Response.pdf

Reviewer 4 Report

Dear authors,

please find my questions and concerns below:

Presented work does not deal with virtual reality. What is the difference between the target printed and displayed on the screen I do not se any connection of Your work with rehabilitation EMG data are treated in a not proper manner! What does mean/explain: i. "grinding methods" used in EMG data postprocessing?  ii. the continuous residual component (cut off at 10Hz) iii. A notch 50Hz, what about informative signals at this frequency? iv. Why You did: "Then the signal was rectified, and an additional 4th order Butterworth filter was applied in a step-by-step mode with a cut-off frequency of 6 Hz to extract the envelope". v. What kind of method was applied to obtain the envelope and what time window was used to do this? What was the reason of used time window range? vi.  What kind of method was used to "Subsequently, each EMG signal was resampled at 1,000 samples.", and why? vii. What does mean "Finally, the 20 signals of each repetition related to the task were concatenated one after the other, attaining for each subject and each task" This description is very confusing to understand the way of the Authors's consideration. 5. According to the description, skin was not prepared correctly to the sEMG,  6. The target (barrel) have different size in case of "real" and "virtual" experiment 7. The results are described in an unsatisfactory manner, the authors do not explain why they chose this type of "statistical" processing 8. To many self-cutations, especially in the most important parts!

Author Response

Please see the attached file.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

The authors have addressed all of my concerns. I do not have other questions.

Reviewer 3 Report

Authors have addressed all my comments, I eccept the manuscript in the present form for publication