Coming Back to the Basics. Comment on Cangir et al. A CT-Based Radiomic Signature for the Differentiation of Pulmonary Hamartomas from Carcinoid Tumors. Diagnostics 2022, 12, 416

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

1. The author needs to analyze the rationale behind the effectiveness of the ΔHU value.
2. The author needs to elucidate the differences and advantages between their proposed new method and the methods mentioned in the original article.

Comments on the Quality of English Language

This comment provides a clear background statement, employs formal language, and demonstrates good readability.

Author Response

Hi,
We appreciate your advice. We've made the corrections as requested.

the advantages are included in the conclusions
"i) simple method to make diagnosis ii) useful in those spoke centers that do not have easy access to DECT or processing software for perfusion analysis or radiomics.

the rationale behind the ΔHU value is wash-out analysis.

Reviewer 2 Report

Comments and Suggestions for Authors

The authors have compared discriminative features in classifying lesions into HAs and NENs published in 2022 in the journal "Diagnostics" against their study using other techniques. They discuss whether using complex radiomics features improves the performance of strict and simple old methods.

I think it reflects the situation in many areas of day-to-day healthcare and the barrier against computer vision and artificial intelligence approaches in a sensible scope such as the clinical. I find the document interesting because the authors aim to open the dialogue.

Since my expertise is not clinical, some questions arose about the authors' claims. These issues are probably due to my complete ignorance of clinical protocols.

1. Almost all works require a continuous need to learn new skills and technologies based on computer vision, and artificial intelligence are emerging as another tool for healthcare practitioners.
2. As far as I understood from the text, the results presented in the authors' previous work need the acquisition of a new image (CECT). What implications does this new test have for the patient? I miss a discussion about the benefits against the drawbacks of both approaches.

Author Response

hi
In contrast to radiomics, our approach is more straightforward, immediate, and easily implementable, even in facilities that lack specialized software. The primary benefit lies in the absence of the need for biopsies, cost-effectiveness, and the ease of execution. I hope I've been exhaustive

Reviewer 3 Report

Comments and Suggestions for Authors

1.       Acronyms such as MLP, RF, PCT needs to be defined.

2.       PCp and Dp are subject to blood flow as well, what were the measures taken to confirm that significant differences were NENs or HAs related?

Author Response

We are thankful for your advice. We have completed the required corrections as requested.

Wash-out (PCp-Dp) is present in cancer (NEN). In HAs the the enhancement will tend to increase in the late/delayed phase and therefore will not present wash-out.

Reviewer 4 Report

Comments and Suggestions for Authors

Dear authors,

I enjoyed reading your paper that reminds us that basic CT signs are often of value to characterize lung nodules instead of complex methods.

As minor comments I suggest the following amendments

1. Please spell out MLP,  RF and AUC at first use

2. Add HU values for NEN and HA as follows : "HU values were significantly different between NENs (add here mean, SD and ranges) and HAs (add here mean, SD and ranges) in the PCp (P<0.001)."

3. Before the sentence that starts with "In the era of radiomics, dual-energy CT (DECT) and artificial intelligence, when radiologists are facing the difficulties of evolving....." please add a short sentence such as  "Although artificial intelligence has demonstrated capabilities and can help the radiologist in the detection of lung nodule [add a ref. here], one must no forget the basic.

4. I suggest to add the following reference

de Margerie-Mellon C, Chassagnon G. Artificial intelligence: A critical review of applications for lung nodule and lung cancer. Diagn Interv Imaging 2023 ;104(1):11-17. doi: 10.1016/j.diii.2022.11.007.

Thank you

 

Author Response

Hi,
We are thankful for your advice. We have completed the required corrections as requested.

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

Thank you for your reply. I reiterate that I cannot assess whether the article should be published.