Enhanced Automatic Morphometry of Nerve Histological Sections Using Ensemble Learning

Round 1

Reviewer 1 Report

In general ensemble learning is an approach to machine learning that seeks better predictive performance by combining the predictions from multiple models. Ensemble learning includes many different approaches and techniques that are described in detail for example in: https://doi.org/10.1007/s11704-019-8208-z. According to the title, ensemble learning is the basis of the used methodology and should be one of the main elements of the article, but the authors devoted only a few lines of the article to this topic. Moreover, the authors chose two popular models used in the segmentation of medical images, implemented and averaged the results, therefore the authors' contribution is questionable.

However, the subject of the article itself is interesting, the used ensemble learning approach is relatively new, and the research results are promising. Therefore, I believe that the submitted article could have been published, but only after making the following corrections:

1)     A better introduction to the subject of ensemble learning, which underpins the methodology used, is needed – types of ensemble learning, other machine learning methods that cooperate with ensemble learning, etc. Literature references are welcome.

2)     The description of the methodology is incomplete. The authors devoted a lot of attention to the preparation of the dataset, but described the models used and the implementation of ensemble learning in a very general manner.

3)     Authors should better emphasize their contributions. It would be a good idea to develop the research methodology, e.g. to use several methods of ensemble learning - bagging, stacking, boosting and to compare them with each other.

4)     Metrics used in the evaluation of the algorithms should be defined in the article under the Results section.

Author Response

Please see the attached

Author Response File: Author Response.docx

Reviewer 2 Report

Comments for author File: Comments.docx

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

I would like to thank the authors for taking into account the comments and for the huge efforts put in improving the quality of the article. A very important element is to indicate the contribution of the authors compared to the methods used so far - improving results through the use of ensemble learning, which is visible in the Results section, and the use of innovative preprocessing methods.

 I still believe that the background and the method itself could be described in more detail, but you can see that the authors tried to take into account the comments submitted. Moreover, I propose to consider the use of other ensemble learning techniques and compare the obtained results in subsequent research.

In addition, I also recommend sending a revised version of the article with marked changes (which is possible in MS Word) or clearly indicating in the cover letter which lines have been changed, which will greatly facilitate the comparison.

Reviewer 2 Report

No more comments