Use Electroencephalogram Entropy as an Indicator to Detect Stress-Induced Sleep Alteration

Round 1

Reviewer 1 Report

In this paper authors used Shannon entropy as an indicator of stress-induced alterations into the electroencephalogram (EEG) to specifically investigate the effect of footshock stimulations on the rat sleep. To me the manuscript is technically sound in methodology and analysis.

The manuscript is overall in good shape. English language should be improved at some points throughout the manuscript, and I think that some paragraphs require further clarifications and elaborations (see specific comments).

Results from the study are novel and interesting. Figures,however, could be improved.

Below you can find some specific comments (all minor):

-In the introduction section (lines 60-78) I would expand the explanation of EEG desynchronization and Entropy as the entire study rely on such concept. Also, I would further explain the reason why Entropy could be a better approach compared to other EEG analytic methods.

-All figures caption should be integrated with more exhaustive explanation of what the figure depict. In particular, in Figure 2C it is not clear to me what the different colored squares (red, green and white) indicate about significance of A vs B. In Figure 3,4,5 A I would better specify the X axis (e.g. what frequencies are included).

-The expression “after exposing” is used throughout the abstract and manuscript. I would replace it with “after exposure”.

-The wording “The result depicts the property changes of brain waves” in the abstract should be “The result depicts the change property of brain waves”

Author Response

We are grateful to the editor and reviewers for reviewing this manuscript and providing incisive suggestions. According to the editor and reviewer's comments, this manuscript has been revised. The revised contents are written in red font and highlighted in the manuscript. The responses to each comment are marked with a gray background point-by-point.

Reviewer Comments:

Reviewer 1:

In this paper authors used Shannon entropy as an indicator of stress-induced alterations into the electroencephalogram (EEG) to specifically investigate the effect of footshock stimulations on the rat sleep. To me the manuscript is technically sound in methodology and analysis.

The manuscript is overall in good shape. English language should be improved at some points throughout the manuscript, and I think that some paragraphs require further clarifications and elaborations (see specific comments).

Results from the study are novel and interesting. Figures, however, could be improved.

Below you can find some specific comments (all minor):

-In the introduction section (lines 60-78) I would expand the explanation of EEG desynchronization and Entropy as the entire study rely on such concept. Also, I would further explain the reason why Entropy could be a better approach compared to other EEG analytic methods.

Response: We added more detailed information about why Shannon entropy was used as an indicator to detect stress-induced sleep alterations in Lines 62-83 on Page 2. However, we deleted the statement of desynchronization part because it is not quite relevant to the present study.

-All figures caption should be integrated with more exhaustive explanation of what the figure depict. In particular, in Figure 2C it is not clear to me what the different colored squares (red, green and white) indicate about significance of A vs B. In Figure 3,4,5 A I would better specify the X axis (e.g. what frequencies are included).

Response: We added detailed figure legend in each figure, especially in Figure 2C. Black dashed-line area marks 1 – 3 Hz, red dashed-line area marks 1 – 6 Hz, green dashed-line area marks 6 – 12 Hz. We are sorry for the unclear descriptions on Figures 3A, 4A, and 5A. We have revised new figures to describe our analysis pipeline. Firstly, we classified the sleep-wake stage as described in Method section 2.5.1. Then, Shannon entropy equation (section 2.5.2.3) was applied to the raw EEG (0.1 to 40 Hz) for its instant amplitude, frequency, or phase. Finally, we sorted values of every epoch of entropy (can be amplitude entropy (in Figure 3), frequency entropy (in Figure 4), or phase entropy (in Figure 5)) from NREM sleep, REM sleep, WAKE based on their sleep-wake stages. In another word we fed raw data into Shannon equation then sorted the entropy values according to the vigilance stages of rats. So, there are no specific frequencies for X axis.

The expression “after exposing” is used throughout the abstract and manuscript. I would replace it with “after exposure”.

Response: We changed “after exposing” to “after exposure” throughout the text.

The wording “The result depicts the property changes of brain waves” in the abstract should be “The result depicts the change property of brain waves”

Response: We changed the description to “The result depicts the change property of brain waves…” as reviewer’s comment in the abstract on page 1.

Author Response File: Author Response.docx

Reviewer 2 Report

The ms. is interesting and contains significant results, however I’d like to suggest some corrections.

The information about humans and animals mixed within paragraphs and sometimes it hard to infer, if the authors talk in general, about humans or rats. While EEG is a good translational tool, the meaning of EEG frequencies band is not the same for humans and animals, thus the separation of this knowledge should be made within a ms. (introduction\discussion)

Make the figure legend more detailed.

More information about where Shannon entropy was used before and why it is expected to be changes in sleep after stress. During the discussion of phase entropy authors relates it to sleep spindles, but this relation for me is not very clear.

Also: autistic spectrum disorder (ASD) or acute stress disorder (ASD) please check carefully

Author Response

We are grateful to the editor and reviewers for reviewing this manuscript and providing incisive suggestions. According to the editor and reviewer's comments, this manuscript has been revised. The revised contents are written in red font and highlighted in the manuscript. The responses to each comment are marked with a gray background point-by-point.

Reviewer Comments:

Reviewer 2:

The ms. is interesting and contains significant results, however I’d like to suggest some corrections.

The information about humans and animals mixed within paragraphs and sometimes it hard to infer, if the authors talk in general, about humans or rats. While EEG is a good translational tool, the meaning of EEG frequencies band is not the same for humans and animals, thus the separation of this knowledge should be made within a ms. (introduction\discussion)

Response: We are sorry for this confusion and we clarified the target subjects, humans or rodents, in every confusing description to identify what subjects were done in the referred study.

Make the figure legend more detailed.

Response: We added detailed figure legend in each figure.

More information about where Shannon entropy was used before and why it is expected to be changes in sleep after stress.

Response: We added more detailed information about why Shannon entropy was used as an indicator to detect stress-induced sleep alterations in Lines 62-83 on Page 2.

During the discussion of phase entropy authors relates it to sleep spindles, but this relation for me is not very clear

Response: We added descriptions in Lines 469-472 on Pages 13 and 14.

Also: autistic spectrum disorder (ASD) or acute stress disorder (ASD) please check carefully

Response: We thanks reviewer’s comment, and we corrected it in text.

Author Response File: Author Response.docx