Making CT Dose Monitoring Meaningful: Augmenting Dose with Imaging Quality

Round 1

Reviewer 1 Report (Previous Reviewer 1)

Thank you very much for the revisions and the nicely written manuscript, which is in my opinion now ready to publish.

Author Response

Thank you for your feedback, we appreciate your time. We are glad to hear that the latest version of the manuscript met your expectations.

Author Response File: Author Response.docx

Reviewer 2 Report (New Reviewer)

1. For the Table 2, why are the Chart 11 divided into a and b? 

2. The 15 Charts were determined by the authors? or are there any literatures to support this?

3. there are only 12 professionals enrolled in this study, this number may not be enough to precisely evaluate their RDMSs. 

4. in the abstract, the author may give more descriptions to introduce their results, not just the conclusions.

Author Response

We thank the reviewer for taking the time in reviewing our paper and offering valuable comments. We have carefully addressed the reviewer insightful comments.

Please see attached document for the responses. 

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report (New Reviewer)

All of my previous concerns have been addressed

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

 

Thank you for the possibility to review this article about the new design of RDMS, which provides a visualization of radiation dose and image quality in terms of noise in one approach. This is an interesting work, even though it suffers from one major limitation:

 

Apart from age, sex, BMI, WED etc., other features influence image quality significantly: especially (motion) artifacts (breath-hold, heart motion, motion itself, overlying arms etc.). I cannot find any discussion about the influence of motion artifacts on image quality or on radiation dose optimization (different heart modes for example). I think this should be included as an additional section of this paper.

 

The figures present mainly different ways of visualization of various parameters – for me, we have a mixture of two main parts: finding a system that takes image quality into account and the way of presenting it to radiology professionals. I would recommend outlining this approach better in the introduction, material & method, and results section.

Author Response

Please find the attached document. 

Author Response File: Author Response.docx

Reviewer 2 Report

 

The authors described their radiation dose monitoring system (RDMS) designed to monitor objective image quality as well as radiation dose.

 

Image quality assessment is indispensable to optimize medical imaging, and such an integrated system can be a powerful tool for opimization.

However, the manuscript appears to be ambiguous and subjective in general, and its study purpose is unclear. I feel that the manuscript is classified as an educational material rather than an original research manuscript.

The Results section indicates that the main component of their study is to evaluate the usefulness of their RDMS by subjective scoring by 12 radiology professionals. However, the Results are presented mainly in Supplementary Materials and are not analyzed or discussed sufficiently.　Moreover, the description of the evaluation procedures is limited. Instead, they explain general matters of radiation dose monitoring, not related to the evaluation, in the Materials and Methods.

The authors mention justification. Please note that justification requires information about clinical indications and is beyond the scope of the RDMS.

2.2.1. Dose metrics

“SSDE still does not account for body shape, organ location, and tissue composition”

SSDE considers attenuation severities in different tissues. Moreover, the writing may implicate that they calculate effective dose and risk index considering the body shape and organ location of each patient although effective dose calculation is commonly based on a mathematical phantom. Does their system use CT data of each patient to assess the body shape and organ location?

2.2.2. Image quality metrics

Although this subsection is in the Materials and Methods, no explanation is provided about the methods of calculating the image quality metrics.

2.3.1. RDMS Design

“These data are collected from the following: picture archiving and communication system (PACS), radiology information system (RIS), and individual imaging systems.”

Figure 1 indicates that data are transferred from imaging modalities and RIS to PACS and from PACS to the database.

Figure 2 gives no name to the second layer. Analysis would be an important role of this layer but is not mentioned in the text.

2.3.1. RDMS Interface

The number 2.3.1. is repeated.

The evaluation method is described in the second paragraph of this subsection. If this is a research manuscript, the description should be the main component of the Materials and Methods. However, actual procedures are not well stated.

Because the authors employed subjective assessment, the characteristics of observers are essential. They explain the observers as “12 radiology professionals including four radiologists, four technologists, and four medical physicists familiar with the needs and potential utility of RDMS.” I do not think that familiarity with the needs and potential utility of RDMS qualifies the observers to evaluate the usefulness of RDMS. Information about experiences in radiation dose management is required.

“The participants were also provided free-text fields for comments on how the charts can possibly be improved.”

Corresponding results are not given. Suggestions for improvement would not be important in a research article, and instead, the basis of judging the usefulness should be informative.

Table 2 provides a list of evaluated charts together with targeted questions. The descriptions do not well represent the charts provided as supplementary materials, and targeted questions should also be reconsidered.

Although the authors emphasize the weakness of CTDIvol in 2.2.1., they evaluated CTDIvol as dose metrics and did not evaluate SSDE and RI at all.

Evaluation of the usefulness of image quality metrics in RDMS should be the key of this manuscript; however, image noise appears in charts 8, 9, and 13 among 15 charts, and neither noise texture nor detectability index is included in the charts.

3. Results

Table S1 presents the main results and should be presented not as a supplementary material but in the main body of the manuscript.

Descriptions of actual results are limited in the text.

The median effective dose is very low in Figure 3 (S6) and Figure S7 although CTDIvol is usual in Figure S7. The effective dose is higher in Figure S12 despite a narrower scan range.

They divide data based on the manufacturer. However, because a manufacturer provides scanners of different models, a division based on the scanner is preferable.

Explanation about the blue and orange bars is required in Figures 6 (S10), 7 (S4), and S5.

The pitch factor is surprisingly high. Additionally, one examination was performed with a pitch of zero. Does this mean non-spiral imaging?

The liver dose in Figures S11a and S11b is much lower than CTDIvol.

Figure S11b indicates a combination of exceptionally high liver dose and low mAs. A noise index of 280 is not real. The value presented as mAs, a product of mA and s, may be a product of mA and ms.

Data for abdominopelvic CT in Figure S12 should be for pelvic CT.

4. Discussion

The main new knowledge obtained through the evaluation may be different scores between different professionals. The differences in scores are discussed in the second paragraph but are not well linked with the results. For example, why did technologists give a high score to Figure S14 and low scores to Figures S4 and S5?

5. Conclusion

The two former sentences among the three seem a priori and are not supported by the results.

Author Response

Please find the attached document 

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Thank your for the revision of your manuscript according to my recommendations.

Author Response

Thank your for your consideration and your time. 

Reviewer 2 Report

Some mistakes have been corrected from the previous manuscript; however, the framework has not been revised.
The authors evaluated the interface of their radiation dose monitoring system only. The scope of the study is limited, and related descriptions are insufficient. Most descriptions are irrelevant to their results.

Author Response

Thank you for your time in the review process. We are sorry that the new version of the manuscript did not address all your concerns. About the specific comment above, the manuscript focused both on designing a new approach to quality monitoring in radiology by also including image quality, and on the most effective way to report and visualize such data. The simultaneous assessment of radiation dose and image quality in a single tool, to our knowledge, is a very novel topic that has never been presented before. Therefore, the first part of our study was focused on the interpretation of the data related with CT practice. We believe that this is essential in the design and the implementation of a performance monitoring system.  

In this scenario, the heterogeneity of the CT data was a constant point of debate in the radiology community after and before the introduction of RDMS. Achieving CT data monitoring harmonization is crucial to enable comparison between studies and institutions and we believe it is absolutely relevant in a study that describes a new approach to performance monitoring.  

After addressing the rationale behind the design and the implementation of the described RDMS, the manuscript then focused on a preliminary evaluation performed by radiology professionals among radiologists, technologists, and medical physicists representing the primary users of the system. Hence, we believe that the reported results are consistent with the second purpose of the study which was to evaluate the usefulness of visualizing the data reported by the system in assessing quality and radiation dose in clinical practice.