A Novel Method to Assist Clinical Management of Mild Traumatic Brain Injury by Classifying Patient Subgroups Using Wearable Sensors and Exertion Testing: A Pilot Study

Round 1

Reviewer 1 Report

Thank you for the opportunity to review this manuscript. The topic is important and there is much to be learned. The presentation and grammar to unremarkable, however, there are a few areas of concern. Below I list each area of the manuscripts with my specific feedback. 

1. Abstract

- The abstract should reiterate this is a pilot study and state a more general aim. The first two sentences are very specific and do not provide the reader with a good overview of the purpose of this pilot study (ie, why is this pilot study needed?) 

-The conclusory statement at the end is far too presumptive. The research showed it is potentially feasible to integrate wearable sensors with the BCTT, but it is still too early to assert this actual utilization in clinical decision-making. the vestibulo-ocular group had was slightly greater than 50% chance of being identified.

2. Introduction 

-The introduction is well written. It sets the need for the study up well. It provides sufficient information for the reader to understand the study, even if the reader is not familiar with machine learning. 

3. Methods

-The detail in the methods is sufficient.

-Figure 1 has several abbreviations that are not defined.

-more information regarding the ECG & acceleration data, with units, should be added to the appropriate subsection.

- more information regarding how age and gender were controlled for should be described. 

3. Results

- First sentence of the results is too long with unnecessary detail.

- I appreciate all of the visuals within the results. 

4. Discussion

- The discussion is not sufficient. It is too short and generally summarizes the findings. The future directions section serves as a better discussion.

-There should be a discussion regarding how age & gender could impact the results. Both factors have been shown to have different effects on mTBI.  

-There should be more references to support the discussion. 

-There should be a discussion regarding the 60% correctly identified in the vestibulo-ocular group. This is slightly better than chance and does not warrant concrete proof. 

-The discussion speaks of 14/17 participants; however, it should be discussed in terms of each group. 

-There is no discussion regarding how the ECG & accelerometry data had similar findings to the accelerometry-only data. This would be a significant point, particularly in a clinical setting. 

-There should be more discussion of sensitivity and specificity, along with reliability values. 

5. Future directions 

-This section is too long and reads more as the discussion section. It should more specifically identify what the next steps are. 

-There should be a discussion of repeating this study in a sufficiently powered sample. 

6. Limitations

-The sample size limitation should not be minimized with the explanation of "richness of within-subjects time series data." This study was not sufficiently powered. It cannot be stated that the findings are proof. The findings indicate the methodology is feasible. 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Very interesting and uptodate research on concussion and traumatic brain injuries. 

I only have some minor issues, please see my comments below:

-Please add some references on long term effects of TBI and mTBI to demonstrate the reader the severity of this injury. 

-Please also add on which based guidelines the diagnosis of TBI was obtained. 

Author Response

On behalf of the authors, we give thanks to the reviewer for their comments that have improved this document. Please see our responses to your comments below:

-Please add some references on long term effects of TBI and mTBI to demonstrate the reader the severity of this injury. 

Thank you for this comment, we have added the following statement to the Introduction to stress the potential of ongoing symptom burden following mTBI.

However, updated reports from large epidemiological studies of the general population have indicated that up to half of mTBI patients can experience prolonged symptom burden beyond this window [2, 9-12]. Moreover, incomplete recovery at six and 12 months has been observed in 50% of mTBI patients who presented to an emergency department within 24 hours post-injury [11, 12]. These poor outcomes, even in early presenters, suggest improved assessment and management pathways are needed to reduce the burden of mTBI. 

-Please also add on which based guidelines the diagnosis of TBI was obtained. 

Thank you for raising this detail. The following statement has been added to the Study Design and Participants section within the Methods.

Diagnosis of sport-related mTBI was made in line with the 2017 Concussion in Sport Group Consensus Statement [8]. 

Round 2

Reviewer 1 Report

The authors should be commended on the work they have done. I am satisfied with the revision and support publication.