Next Article in Journal
Spatial Distribution Characteristics and Source Appointment of Heavy Metals in Soil in the Areas Affected by Non-Ferrous Metal Slag Field in the Dry-Hot Valley
Previous Article in Journal
Deformation Behaviour of Cold-Rolled Ni/CNT Nanocomposites
 
 
Article
Peer-Review Record

The Relationship of Behavioral, Social and Diabetes Factors with LVEF Measured Using Machine Learning Techniques

Appl. Sci. 2022, 12(19), 9474; https://doi.org/10.3390/app12199474
by Cezara-Andreea Soysaler 1,*, Cătălina Liliana Andrei 1, Octavian Ceban 2 and Crina-Julieta Sinescu 1
Reviewer 1: Anonymous
Appl. Sci. 2022, 12(19), 9474; https://doi.org/10.3390/app12199474
Submission received: 24 August 2022 / Revised: 16 September 2022 / Accepted: 17 September 2022 / Published: 21 September 2022

Round 1

Reviewer 1 Report

The manuscript entitled “applsci-1907096” dealing with The relationship of behavioral, social and diabetes factors with LVEF measured using machine learning techniques has been reviewed. The paper has been nicely written but needs significant improvement. Please follow my comments.

 

 

1.     Please add a brief statement on your methodology in the abstract.

2.     Add more detail about your work to the last paragraph of the introduction.

3.     Machine learning has great usage in different industries such as medical, mechanical and manufacturing. Add a statement about the application of machine learning in manufacturing by adding the four following papers to the introduction.

·       A review of Industry 4.0 and additive manufacturing synergy

·       Sustainable design guidelines for additive manufacturing applications

·       Relative density prediction of additively manufactured Inconel 718: a study on genetic algorithm optimized neural network models

·       Fused filament fabrication of nylon 6/66 copolymer: Parametric study comparing full factorial and Taguchi design of experiments

4.     Avoid using dot points in the “materials and methods”.

 

5.     Add more detail to the conclusion and explain how your findings can support the text.

6.     What is the future work for this research? How do authors believe that the next stage of this work is useful? Please add a sentence or two.

 

 

 

 

 

 

 

 

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 2 Report

The authors analyze a dataset related with cardiac health with machine learning methods. The approach is interesting, however there are several important issues (see comments). I think that the analysis is too simple and the paper should be enriched in order to be published.

Major comments:

1) Please describe with more detail what the LVEF variable shows, since this is the target variable.

2) Shapley values, a key component of the methodology, should be presented. The reference provided (33) is irrelevant.

3) Classficiation accuracy is not very high (59%, 74%, 79%). I would expect a model with all variables together. It might achieve better classification, thus provide more solid knowledge. Otherwise, the practical application for diagnosis and prevention, which is the aim of this paper, will  be very limited.

4) Some other classifcation models should be applied and if XGB is the best, then proceed with it for variable exploration.

5) For Logistic Regression, some results (classification accuracy etc) could be reported.

6) Fig. 1: at which level are the correlations significant (p < 0.05)?

7) It should be mentioned that the diabetes features are not independent. If a person has no diabetes, all are 0 etc.

8) Optionally, I suggest instead of print-screens in Tables 3-5, to create  tables including only the important information.

9) Discussion: a more in depth discussion is needed. What other studies have shown about these factors?

Minor comments:

1) Data: when is a patient considered obese?

2) "following table", "following image": please replace with Table X, Figure Y.

3) English language can be improved. For example the adjective "large" should be used instead of "big" in phrases such as "big impact".

 

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

The paper is ready to publish.

Author Response

Thank you!

Reviewer 2 Report

The authors replied on all points. For some suggestions, some changes have been done, making the manuscript more complete. For other suggestions, a reply has been provided, clarifying the point of view (although the answers could be more polite). However, the following points remain:

1) Regarding comment "Fig. 1: at which level are the correlations significant (p < 0.05)?"

Considering that you have 200 sample, the significance level is at |r| >= 0.14, which can be mentioned in text.

2) About comment "Discussion: a more in depth discussion is needed. What other studies have shown about these factors?"

Some content has been added, however, if possible I would expect some more references about the specific factors that were found significant.

 

 

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Back to TopTop