Active Semi-Supervised Learning via Bayesian Experimental Design for Lung Cancer Classification Using Low Dose Computed Tomography Scans

Round 1

Reviewer 1 Report

Please mention the obtained statistics of this work in the abstract and conclusion.

I would like to recommend below article to be discussed in Introduction or related work section.

WCS Low, JH Chuah, CATH Tee, S Anis, MA Shoaib, A Faisal, A Khalil, Azira Khalil, Khin Wee Lai, An overview of deep learning techniques on chest X-ray and CT scan identification of COVID-19, Computational and Mathematical Methods in Medicine 2021, 1-17.

Author Response

We added obtained statistics of this work in both abstract and conclusion sessions.

We added the recommended citation. We made improvements in writing the manuscript by rephrasing sentences and giving better explanations in all sessions.

Author Response File: Author Response.pdf

Reviewer 2 Report

I have read with interest this paper. Methodology is correct and the text is well written. I only suggest to improve references considering the following ones:

- Azour L, Hu Y, Ko JP, Chen B, Knoll F, Alpert JB, Brusca-Augello G, Mason DM, Wickstrom ML, Kwon YJF, Babb J, Liang Z, Moore WH. Deep Learning Denoising of Low-Dose Computed Tomography Chest Images: A Quantitative and Qualitative Image Analysis. J Comput Assist Tomogr. 2023 Jan 27. doi: 10.1097/RCT.0000000000001405. Epub ahead of print. PMID: 36790870.

- Li Y, Liu J, Yang X, Wang A, Zang C, Wang L, He C, Lin L, Qing H, Ren J, Zhou P. An ordinal radiomic model to predict the differentiation grade of invasive non-mucinous pulmonary adenocarcinoma based on low-dose computed tomography in lung cancer screening. Eur Radiol. 2023 Feb 15. doi: 10.1007/s00330-023-09453-y. Epub ahead of print. PMID: 36790469.

- Aquila I, Sicilia F, Ricci P, Antonio Sacco M, Manno M, Gratteri S. Role of post-mortem multi-slice computed tomography in the evaluation of single gunshot injuries. Med Leg J. 2019 Dec;87(4):204-210. doi: 10.1177/0025817219848264. Epub 2019 Sep 28. PMID: 31564213.

Congratulations to the authors.

Author Response

The recommended references have been added in the paper. Please refer the attached PDF with all the changes.

Author Response File: Author Response.pdf

Reviewer 3 Report

I find this article interesting and informative. The work is well written and represents the area of the problem.

Please, take into account:

1) Figures 2 and 3 are poorly presented; a clear comment is needed; also a better resolution and a larger size are needed.

2) Figure 4 is informative, but poorly readable, other options for presenting information should be considered. You should also number the stages so that their discussion is more comparable to the flowchart.

6) Figure 5 is also not informative and needs a clear comment or the application of a blue stripe over a yellow one.

Author Response

We enlarged the images of the nodules in figure 2 and 3.

We redo the figure 4 for better visualization and presentation.

We added clear comments for Figure 5. We could not redo the figure 5 since it requires rerunning the training experiments. However, we added explanations and interpretation in the writing.

Author Response File: Author Response.pdf

Reviewer 4 Report

The manuscript is well planned and interesting. However, it was found that similar results were published in authors' another article "Nguyen, P., Chapman, D., Menon, S., Morris, M., & Yesha, Y. (2020, March). Active semi-supervised expectation maximization learning for lung cancer detection from Computerized Tomography (CT) images with minimally label training data. In Medical Imaging 2020: Computer-Aided Diagnosis (Vol. 11314, pp. 553-564). SPIE. Although given paper was published as proceeding, it was published as article and has doi number (https://doi.org/10.1117/12.2549655). The editor should consider this publication or authors should explain the difference of this manuscript from given article.

Author Response

Please see the attachment for the response to the comment below.

Author Response File: Author Response.pdf

Round 2

Reviewer 4 Report

The manuscript can be accepted in its current form.