Artificial Intelligence and Democratization of the Use of Lung Ultrasound in COVID-19: On the Feasibility of Automatic Calculation of Lung Ultrasound Score

Round 1

Reviewer 1 Report

An evaluation by a professional trained in pulmonary ultrasound would have been necessary to compare with the results of the UTRACOV.

Author Response

Madrid, 30^th of December of 2021

 

Dear Editor and Reviewers,

 

 

On behalf of my co-authors, I would like to thank you for the opportunity to revise our manuscript. We greatly appreciate the reviewers positive and constructive comments and suggestions regarding our manuscript.

Thank you very much for your consideration of our manuscript for publication.

 

We look forward to the outcome.

 

Yours sincerely,

Corresponding Author

 

 

 

REVIEWER 1

 

An evaluation by a professional trained in pulmonary ultrasound would have been necessary to compare with the results of the UTRACOV.

• RESPONSE: We agree. Indeed, the results of the ULTRACOV algorithm are compared with those obtained by a professional trained in pulmonary ultrasound (referenced as “the expert” in the text). The manuscript has been updated to hopefully clarify this point.

 

 

Reviewer 2 Report

Dear Authors

The article presented for review is interesting, the very ideal of using the automatic LUS-Score calculation is obviously tempting, and the description of the method used is accurate and comprehensive. However, I have some important comments as to the content and questions about the clinical relevance of your study.

 

1. There is no graphic explaining the probe application points
2. How many patients have been examined? There is information about 30 patients in the abstract and in the text, while in Figs. 3-4 there are numbers from 4 to 39. Why this discrepancy?
3. If pleural effusion is found in only one patient, it is debatable whether to include this parameter in the calculations at all.
4. Finally, I miss the reference to the clinical application of the LUS-Score. It is limited to a specific group of patients. Please define it and give your opinion on whether the group of patients assessed in your study corresponded to this population.
1. In whom can the LUS test be limited to 12 points, and in whom should we assess the entire available lung fields?
2. There is also a lack of information as to what clinical benefits (therapeutic decisions) would follow from automated testing. Could the finding of consolidation of subpleural or only "pure" interstitial lesions influence these decisions?

In a word - it is necessary to notice and describe in more detail the limitations of the automatic method.

I will be very pleased to see the completed material

Best regards

Author Response

Madrid, 30^th of December of 2021

 

Dear Editor and Reviewers,

 

 

On behalf of my co-authors, I would like to thank you for the opportunity to revise our manuscript. We greatly appreciate the reviewers positive and constructive comments and suggestions regarding our manuscript.

Thank you very much for your consideration of our manuscript for publication.

 

We look forward to the outcome.

 

Yours sincerely,

Corresponding Author

 

 

 

REVIEWER 2

 

Dear Authors

The article presented for review is interesting, the very ideal of using the automatic LUS-Score calculation is obviously tempting, and the description of the method used is accurate and comprehensive. However, I have some important comments as to the content and questions about the clinical relevance of your study.

1. There is no graphic explaining the probe application points

• RESPONSE: Thank you for the comment, it was included in the new version.

 

1. How many patients have been examined? There is information about 30 patients in the abstract and in the text, while in Figs. 3-4 there are numbers from 4 to 39. Why this discrepancy?

• RESPONSE: Thank you for the comment, it was a mistake in the figures and text. The number of patients in this study was 28. It has been corrected in the manuscript.

 

1. If pleural effusion is found in only one patient, it is debatable whether to include this parameter in the calculations at all.

• RESPONSE: It is a good point. We prefer to include it, because despite being only one video labeled as pleural effusion by the expert, it is interesting to confirm that the algorithm does not produce false positives for this indication. Following your suggestion, the sentence regarding this point was modified in the discussion to emphasize this fact.

 

1. Finally, I miss the reference to the clinical application of the LUS-Score. It is limited to a specific group of patients. Please define it and give your opinion on whether the group of patients assessed in your study corresponded to this population.

• RESPONSE: Thank you for this comment. In this pandemic context, easy to access and reliable diagnostic methods which can accurately guide management in COVID-19 are vital. Lung ultrasound has the potential to become a first-line diagnostic tool alternative to conventional chest X ray and CT scan, including the critically ill patients, where LUS-Score had been proven to be useful in diagnosis, prognosis and monitoring management. Moreover, since there is no exposure to ionizing radiation, can be considered in vulnerable population such as pregnant women and children. This automatic method might become as a quick and easy solution, in the long-waited standardization of the technique.

We added this idea to the text.

1. In whom can the LUS test be limited to 12 points, and in whom should we assess the entire available lung fields?

• RESPONSE: this is an interesting, yet to be answered, question. Our group believes that each patient might benefit most from a different approach, adapted to a flexible scanning protocol (6-8-12-14 areas) subject to the clinical scenario. However, adopting a standardized protocol (12 areas) such as the one we propose, might increase the accuracy of the technique, especially in minimally trained operators, where this automatic method might benefit most, allowing the democratization of the technique.

Following your suggestion we added a comment in the discussion.

•  

 

1. There is also a lack of information as to what clinical benefits (therapeutic decisions) would follow from automated testing. Could the finding of consolidation of subpleural or only "pure" interstitial lesions influence these decisions?

• RESPONSE: This automatic-method approach will allow, in particular settings and clinical conditions, to rule in or out quickly and accurately several diagnoses. Save time in the assessment of the lung involvement, with direct impact in the diagnosis, prognosis or monitoring of the disease.

Following your suggestion we added a comment in the discussion.

 

In a word - it is necessary to notice and describe in more detail the limitations of the automatic method.

• RESPONSE: From the technical point of view, a limitation of the automatic method is that it needs a correctly acquired video for being analyzed, which can be sometimes difficult to achieve for personnel with low experience. We are currently working in artificial intelligence algorithms that could help the physician during the scanning process, giving a real-time score of the image quality before processing it. From the clinical point of view, the limitations of the automatic method are that these automatic results, although can save time in the assessment, needs to be integrated with a standard clinical approach to optimize diagnostic accuracy.

Following your suggestion this analysis was included in the new version of the manuscript.

 

I will be very pleased to see the completed material

Best regards

 

Round 2

Reviewer 2 Report

Dear Authors

Thank you for taking into account my comments and introducing corrections and clarifications. Once again, congratulations on your interesting work and project that can really contribute to the standardization of LUS.

I believe that your article is interesting and definitely worth publishing.

I wish you good luck for the new year 2022 - professional, scientific, personal success and, above all, health and safety.