Artificial Intelligence for Multiclass Rhythm Analysis for Out-of-Hospital Cardiac Arrest During Mechanical Cardiopulmonary Resuscitation

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This paper proposes an AI solution for multi-class cardiac rhythm classification in the context of mechanical cardiopulmonary resuscitation (CPR). The submission is worthy of publication. However, there are still some shortcomings:

(1)The introduction could more clearly point out the research gap in multi-class rhythm classification in the current mechanical CPR scenario, highlighting the innovation of this study.

(2)It would be beneficial to include a confusion matrix or ROC curve to visually display the distribution of classification errors across categories.

(3)It should be explained why ResNet was chosen over other networks (such as Transformer) as a comparison model to demonstrate the rationale behind the model selection.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

- The authors need to clearly provide a statement concerning the ethics approval and committee. 

- The authors need to provide a clear and complete detail on computational experiments and performance validations. The manuscript still lacks the repeatability. 

- From the computational experiments and results, the performances obtained and presented in this study have not yet been proved to be the best performance achievable because the parameters applied to the computational experiments are limited. 

- The authors should consider to have a comparative study by clearly comparing and discussing the performance to the relevant studies. 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The revised manuscript is now in an acceptable format.