Development of a Web Application for the Detection of Coronary Artery Calcium from Computed Tomography

Round 1

Reviewer 1 Report

The manuscript presents a web application that semi-automatically quantifies the amount of coronary artery calcium (CAC) based on the coronary calcium score (CS) using the Agatston technique through digital image processing. To verify the correct functioning of this web application, 30 CTCS were analyzed by a cardiologist in comparison to a commercial software (OsiriX DICOM Viewer). All the scans were correctly classified according to the cardiovascular event risk group, with an average error in the calculation of CS of 1.9% and a Pearson correlation coefficient r=0.9997, with potential clinical application.

The paper is somewhat messy and should be ordered according to the different sections. Therefore, in the introduction, lines 51 to 57 should be moved to the material and methods section, following line 67. Then it should be mentioned that Pearson’s correlation coefficient was calculated for each of the assessments performed as compared to the reference OsiriX DICOM viewer software, and also that the diagnoses issued by the specialist were compared with the total CS value per scan for each of the density score method developed (Original, Gaussian and MinMode), and that the average error was calculated between each of the comparisons made. Finally, it should be mentioned here that the risk of cardiovascular event corresponding to each of the studies according to the classification given by the CS was compared to the reference.

Likewise, lines 52 to 62 of the introduction and lines 78 and 79 of material and methods, as well as table 1 should be removed.

Finally, the authors state that according to the results obtained in the comparisons of the image processing and CS calculation methods, the MinMode technique is the one that most closely resembles the results obtained with the reference commercial software. How does the MinMode method compares with the reference OsiriX DICOM viewer software for the entire sample of 80 CTCS (and not only for the presented subsample of 30 CTCS)?

Author Response

"Please see the attachment."

Author Response File: Author Response.docx

Reviewer 2 Report

Dear Editor

Thank you for the opportunity to review the paper titled: DEVELOPMENT OF A WEB APPLICATION FOR DETECTION OF CORONARY ARTERY CALCIUM FROM COMPUTED TOMOGRAPHY

 

The paper address interesting topic and the authors tried to come up with new product to measure CAC. I am not sure of the importance of such product as most of the CT machines come up with the same software. If the authors will add education and teaching material with the software, then the benefits will be higher.

-          Do the authors validate the software and compare it with available software results?

-          Can the authors define the DICOM image (DICOM tag) required information to process the images? It may be useful for the images exported without patient’s information.

-           Add a figure showing comparisons between the software and company software results.

-          The author should add clear instruction about the website, am not able to upload images due to size and type restriction. Why win rar not accepted. How many images should be uploaded to see the result?

Author Response

"Please see the attachment."

Author Response File: Author Response.docx

Reviewer 3 Report

In my own work I have become a big fan of moving things to web applications for the reasons the authors state, so I think it's good to get examples in the literature showing different kinds of applications.

It would be good if the site could provide at least one example data set built in for people to test things out. I don't normally work with dicom files so I downloaded 4 different dicom files from different sites on the internet and was able to upload all of them to the platform, but none of them displayed as images and all of them gave errors when I tried to load or save results. So I wasn't able to test the site for myself. I don't know what combination of things might have been wrong (whether I was using a wrong file type or if there was a problem with the site). But having a built-in example and maybe information about file type constraints would help.

Author Response

"Please see the attachment."

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

The final comment from reviewer 3 has not been addressed:

"the authors state that according to the results obtained in the comparisons of the image processing and CS calculation methods, the MinMode technique is the one that most closely resembles the results obtained with the reference commercial software. How does the MinMode method compares with the reference OsiriX DICOM viewer software for the entire sample of 80 CTCS (and not only for the presented subsample of 30 CTCS)?"

In addition, do the 30 CTCS finally compared with the three score calculation methods come from the same institution or from the three hospitals mentioned, if so, in what proportion?

Author Response

"Please see the attachment."

Author Response File: Author Response.docx