Motion Recognition and an Accuracy Comparison of Left and Right Arms by EEG Signal Analysis

Round 1

Reviewer 1 Report

This study describes an analysis of EEG signals predicting EMG signals, and as such, demonstrating good accuracy in classifying whether a subject is moving the left or right arm based on EEG. The study included detailed methods on the collection of the data, the analysis methods, and the general approach. In particular, the study indicated that a Common Spatial Pattern filter was beneficial in separating components of the complex EEG signal, leading to better prediction of EMG.

 

My main concern about the paper is that the English needs to be inproved. I think this would make the paper much more clear and precise. Also, the figures had some errors in formatting or spelling and should be checked (e.g “norch” should be “notch”).

Author Response

Thank you for reviewing our paper.

I corrected some misspelled words and improved my overall English.

Author Response File: Author Response.docx

Reviewer 2 Report

This is a well-organized paper presenting a data driven approach to explore the basis for predicting wills during muscle signals and stimulation transmission by analyzing EEG and EMG signals through CSP filter.

The authors should elaborate more in Introduction the novelty of this work and the important contribution of this work to the scientific communities, particularly in clinical or entertainment settings. Spatial filtering is a very efficient solution for EEG-based BCI in general. However, a 10% improvement of accuracy after the application of the CSP filter does not seem significant. Have the authors tried other pre-processing ways like independent component analysis (ICA) to remove artifact before going through the filter. Although the EEG signal is not fixed and varies from person to person, larger sample scale is expected to prove the predicting accuracy and algorithm efficiency. The references are expected to be more updated, 14/20 references are published before 2015. For a data-driven research article of 2019, the majority of references should be after 2015 and summarize more recent research progression.

Author Response

Thank you for reviewing our paper.

The introduction of the paper has been revised. 

We tried to eliminate the disturbance as much as possible when constructing the experimental environment, but we found some problems in eliminating the variability and error of brain waves completely. In the future, we will conduct research to increase the data sample and improve the pre-processing process to improve the accuracy.

We have added and updated the latest paper to the reference section.

 

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

The introduction has been well improved and the references are much more updated.

Even though the sample size is still limited, however, considering the EEG signal is not fixed and the difficulty of removing the noise completely. I would recommend publishing this manuscript and encouraging enlarge sample size and optimize algorithm in the future study.

Author Response

Thank you very much for your comments. The addition of sample data to increase the research effect you mentioned is included in the future research plan, and is specified in the section of the paper's P16. 23~26 line, Future work.

The changes are as follows:

“Moreover, the next study will select a number of subjects (5-10: left-handed, right-handed) to increase the number of samples used for data analysis and compare the results. And if noise cancellation is applied step by step, a comparable study will be conducted to see how the resulting data will differ depending on the noise environment.”

Thank you.

Author Response File: Author Response.pdf