A Scoping Review of Machine-Learning Derived Radiomic Analysis of CT and PET Imaging to Investigate Atherosclerotic Cardiovascular Disease

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The paper is interesting, but some points should be clarified.

In the title the words “radiomic techniques” should include several imaging techniques, while in the abstract only CT, CTCA, PET/CT techniques are indicated. On the other hand in the Supplementay Material also Magnetic Resonance is reported. Several other techniques are indicated from line 41 to 56. In line 91 CT, MRI and PET are indicated. See also lines 99-101. In Table 1 no MRI studies seems to be reported.  The focus, the purpose and the methodology of inclusion/exclusion of the present paper are then not so clear and should be clarified. Probably also in the Title, only techniques under study in this paper should be more clearly indicated for a better understanding.  The fact that only paper where AI was used are included should be also clear in the Title, in the text and in lines 300-301.

The sentence in lines 78-79 appears also too generic.

Exclusion criteria and reasons for excluding full text-studies reported in Figure 2 (Eligibility) should be also described in detail in the text (Paragraph 2.1. Eligibility).

Lines 92-93  report that “Included studies focused on detection or prediction of atherosclerotic disease in the native carotid, coronary or lower limb peripheral arteries “. This appears as a specification of the purpose of the paper, which should probably be reported in some way in the title, to allow the title itself to become more informative and not so generic.

Table 1 and 2 are too long, with too many redundant details: should be shortened and more concise. On the other hand,  AUC and ROC values are not reported, but could be useful for comparison. Also a detailed comparative discussion of the clinical relevance of the results and limitations for each study should be reported  in the Discussion and Limitations sections.  Table 3 seems also redundant and repetitive: It could be perhaps summarized in the text and/or in the other Tables  (could Table 3 be deleted?).

Conclusions in the Abstract and in the Text indicate: “ There is a need for implementation of standardised imaging acquisition protocols, adherence to published reporting guidelines and economic evaluation.” A more widespread involvement even of small and peripheral centers, a quality analysis of the results and a comparison of advantages, if not marginal, with consolidated techniques should be probably added.

Comments on the Quality of English Language

Minor editing required.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This is a review of a pioneer technique: Radiomics involving the extraction of quantitative data from imaging features that are imperceptible to the eye. The focus is on cardiovascular disease. One important aspect of the review is about the machine learning technique. 

Methodology: Near perfect technique with detailed review of the literature. 

Critical assessment of the studies: very helpful for the readers. 

Critical assessment of the ML techniques when studying the data (from CTCA, PET scan, etc. 

Very helpful for the readers to know the current status and the problems to be solved in order to advance the technique. 

English language is perfect. The authors use British terminology (which is a little different to the American readers). However, it shows the richness of the English language

The authors should be congratulated for an outstanding review  

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

This comprehensive review by Badesha et al aims to summarize radiomic techniques used in the investigation of CVD, specifically focusing on imaging modalities such as CT, MRI, and PET. It reviews existing studies to summarize the methodologies and findings related to the use of radiomics in assessing atherosclerotic disease in the carotid and coronary arteries. The main contributions of the paper include identifying the heterogeneity in imaging acquisition, segmentation techniques, and radiomics processing among the studies, and highlighting the need for standardized protocols and reporting guidelines in this field. A strength of the paper is its comprehensive approach to synthesizing existing research, which underscores the potential of radiomics to contribute to precision/personalized medicine in CVD, despite the current variability in methods.

The field of radiomics, especially in regard to cardiovascular disease (CVD), is of great interest since CVD is the leading cause of mortality and morbidity worldwide, with a significant economic impact. Therefore, anything that facilitates easier and earlier diagnosis is highly valuable. Radiomics is such a tool that can either identify CVD early in imaging studies performed for other reasons or predict disease progression based on certain imaging parameters. Overall, the review is very well-organized, and the use of PRISMA guidelines enhances its quality.

 

I have two comments:

1) I appreciate that the limitations are clearly described in the discussion section, but a separate limitations section between the discussion and conclusion would help the reader assess these limitations at a glance.

2) Given that there are limitations, consider grading the quality of the included studies using a tool like the Newcastle-Ottawa Scale, and including a table with the results. This would provide a clearer assessment of the quality of the studies reviewed and strengthen the overall analysis.

Author Response

Please see the attachment..

Author Response File: Author Response.pdf

Reviewer 4 Report

Comments and Suggestions for Authors

Thank you for the opportunity to read this review article. Radiomics is a frequent argument of debate in scientific literature and it could be used to evaluate the CVD. The aim of this review is to summarise the existing literature on radiomic analysis techniques in CVD.

The article is well written and well organized. I have only few suggestions.

ABSTRACT

I suggest reporting the aim of this review article.

INTRODUCTION AND MandM

Well written.

RESULT

I suggest reducing the lenght of the paragraphs because they could be difficult to read. I suggest eliminating the references and reporting the most relevant explanations in the discussion.

DISCUSSION AND CONCLUSION

Well written.

TABLES AND FIGURES

I suggest adding the abbreviations under all tables and figures.

ENGLISH LANGUAGE

The english language is fine.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The Authors made the required changes.

Comments on the Quality of English Language

Minor editing required.