A Channel Correction and Spatial Attention Framework for Anterior Cruciate Ligament Tear with Ordinal Loss

Round 1

Reviewer 1 Report

Dear Authors, I consider your paper, which addresses the training problem caused by the small sample size of the dataset and unbalanced samples, very interesting, based on the triple classification of the ACL tear. It is clear the verification of the effectiveness of each module individually and together, the characteristics of which are exemplified in the graphs in the displayed tables.

I would like to congratulate for your exposition; the study, otherwise the limitations that you appropriately identified, certainly presents interesting points; the purpose is clear and respected. The information provided is sufficient and presents useful elements to encourage the development of new scientific work. The paper is well written, following established scientific assumptions. The study design is rational. The conclusions presented are meaningful and the statistical sample used is correct.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

Thank you for the opportunity to review the paper "The channel correction and spatial attention framework for anterior cruciate ligament tear with ordinal loss".

I make some suggestions for changes and improvements to the document itself.

Abstract

In the abstract the type of study carried out is not understood.

Modify the key words, as they are repeated with the title.

Introduction

Very long introduction, they should rephrase and show more clarity on the relevance of the study.

Method

In figure 2, you should caption the acronyms.

Results

4.3.2 Uniform ResNet-18, in the table with the text.

Discussion

Not very exploratory discussion considering the work developed. Need to discuss better and more clearly all aspects.

Conclusion

Very extensive conclusion. In the conclusion they should respond to the objectives they set out to study only.

The limitations of the study should be in the discussion and not in the conclusion, they should carry out this reformulation.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

This is a very interesting work. It is well written and explicit.

I think the manuscript can be accepted with only minor revisions.

This study allows obtaining corrections to the measured data in order to approximate the reality of all the parameters read, and thus diagnose and prevent the evolution of the lesion in order to increase the patient's quality of life.

Throughout the text there are some acronyms that have no indication of their meaning, like:

Line 51, AUC

Line 54, ROI.

The designation ‘Resnet18’ should be written in the same way throughout the text, please check.

In the Line 197, topic 3.1, is important beging this topic with a small text. Is not good appear the image immediately, without text.

In figure 2 is interesting to identify the four parts that describe in the text.

In figure 3 is important defined a), b), c) and d) in the figure. Same comment in figure 9.

The equations 2, 3, 4, 5 how it relates the equations to the tasks indicate in figure 4, I think it need more description. Where it identifies throughout the text.

I suggest reducing the number of significant digits in table 1, 2 and 3. throughout the text the author uses two significant figures, a maximum of 3 digits will be enough to present in table 1. The utilization of 6 significant digits seems too much to me.

Is interesting in figure 6 put the maximum values to verify the difference, because as they are very close the doubt remains...only the text complements the interpretation of the graph. So I think you should insert the maximum values.

I send my comment's in attach file.

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

Accept.