Predicting Multi-Gene Mutation Based on Lung Cancer CT Images and Mut-SeResNet

Round 1

Reviewer 1 Report

In this papers authors proposed to embbed attention mechanism into a Resnet Architecture for predicting Multi Gene Mutations. The proposed method is original and it is very significant because it helps to perform diagnosis. The quality of the paper is very good, however more experiments are needed. I suggest to use another dataset and more experiments be carried out in order to have more results to be able to carry out a deeper analysis and discussion about the proposed method.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

 

1)      Change the title to "Predicting multi-gene mutation based on lung cancer ct images and Mut-SerestNet"

2)      There is no justification why author need to pro[posed such a study. Therefore, a new section "Related works" is needed to clarify challenges into existing study and highlight the main gap of these studies.

3)      The contribution of the study is not highlighted well at the end of the introduction section.

4)      What is the problem with trained dataset that made the author using data augmentation process? Highlight.

5)      According to figure 1 the output is multi-class classification which based on four classes. I wonder if there is any over-fitting or under-fitting or bias scenario is happened during the training process especially authors have used data augmentation to handle the minority of the data in distinct class. Authors are suggested to provide a confusion matrix per each class.

6)      Dataset and model settings should be highlighted into table form.

7)      Comparison is based on author selection not on the selecting studies from relevant literature or in other word state of the art studies. So revise this section with comparison with 3-4 work from recent literature.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

Title of the paper: "Lung cancer CT images for predicting multi gene mutations based on Mut-SeResNet "

 

    As a researcher working in the same field, I am impressed by the technique introduced in the paper, because it sheds new light on the earlier results of several authors and obviously can be successfully used in practice. From this point of view, the subject of the paper fits well with the scope of the journal (Applied Sciences).

 

The paper is ended with numerical simulations that corroborate the theoretical results.

This manuscript contains new ideas and good results that help other researchers.

The decision is too major revision it for publication in the " Applied Sciences".

 

Therefore, I recommend publishing this work after taking these points into account.

1-      Introduction needs to explain the main contributions of the work more clearly.

2-     The novelty of this paper is not clear. The difference between present work and previous Works should be highlighted.

3-     In the references in the Introduction section, the authors cite some works. However, they have not indicated the advantage or disadvantage and their relations to this paper. It’s a little confusing.

4-     Comparison with recent studies and methods would be appreciated.

5-      Need a detailed explanation of the preprocessing steps.

6-      Clarify the finding Error rate and accuracy in the performance analysis section.

7-      Introduce the chart for the given algorithm with a description.

8-     Authors should explain the reason why they choose these methods. What are the limitations of this work? How can the rigor of this work be demonstrated?

9-     The  authors can add the following reference to enrich the introductory section:

 

*A numerical method for solving the nonlinear equations of Emden-Fowler models, Journal of Ocean Engineering and Science, 2022. doi.org/10.1016/j.joes.2022.04.019.

*Use of optimal control in studying the dynamical behaviors of fractional financial awareness models, Soft Computing, vol. 26, pp. 3401–3409, 2022.

10- Future recommendations should be added to assist other researchers to extend the presented research analysis.

Author Response

Please see the attachment

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Authors have improved the manuscript by adding new information. Here are some suggestions to have a better version of the paper so it can be published:

1. Line 45: authors mentioned "Deep learning is a process that simulates the cognition of human things". DL does not simulates any human process, it is only an approximation. It is not clear the words "human things"

2. Line 61. The sentence "In the past, medical approach was used to predicted gene mutation, while we used deep learning methods" is not clear, I suggest to rewrite it to have a better understanding

3. Please make sure equation 4 is correct

4. Lines 129-133 has a different format than the rest of the manuscript

5. Line 135: replaces Elu by ReLU

 

Reviewer 2 Report

My opinion is still the same about following issues:

-The contribution is to describe the real novelty of the specific work in terms of presenting new system or framework or model that can be applied for real time scenario. Please have a look on following paper:

https://peerj.com/articles/cs-303.pdf.

-The confusion matrices for all classes still not mentioned into manuscript.

-More recent references need to included such as:

-https://iovs.arvojournals.org/article.aspx?articleid=2776681.

-https://www.mdpi.com/2075-4418/12/10/2472.

-https://www.hindawi.com/journals/misy/2022/7675925/.

-https://www.hindawi.com/journals/cin/2022/1307944/

 

Reviewer 3 Report

There are no comments. The revised version has now been excellent. In my opinion, the revised version is well organized, and the results are correct and interesting. I strongly recommend its publication.

 

But, Please added the Author (A.M.S. Mahdy) in [12] and Authors (A.M.S. Mahdy, Kh. Lotfy, A. A. El-Bary) in [13].