Preference and Demand for Digital Pathology and Computer-Aided Diagnosis among Korean Pathologists: A Survey Study Focused on Prostate Needle Biopsy

Round 1

Reviewer 1 Report

This study focuses on the preference and demand for digital pathology and computer-aided diagnosis among Korean pathologists by using a survey. This is an on-time issue since more and more pathologists are facing the demand. Korean experience about this matter is certainly worth sharing. On the other hand, the survey dealt with the attitude and expectation of pathologists who tested a software program (PROMISE-P)  to diagnose prostate cancer on needle biopsy. However, the true performance of the program, especially whether it is better than human, was not analyzed. We do not know whether the program is less time-comsuming, more accurate and more efficient in comparison with traditional pathology workflow.

          Because I am not able to access the program (the link to the supplementary files is not working), I have no idea about whether the program is user-friendly or not, that certainly will affect the willingness of pathologists to adopt digital pathology in their routine work. I suggest the authors to demonstrate a few crucial screen-shots of PROMISE-P so that the readers can understand how the software program performs and achieves the goal.

Author Response

This study focuses on the preference and demand for digital pathology and computer-aided diagnosis among Korean pathologists by using a survey. This is an on-time issue since more and more pathologists are facing the demand. Korean experience about this matter is certainly worth sharing. On the other hand, the survey dealt with the attitude and expectation of pathologists who tested a software program (PROMISE-P)  to diagnose prostate cancer on needle biopsy. However, the true performance of the program, especially whether it is better than human, was not analyzed. We do not know whether the program is less time-comsuming, more accurate and more efficient in comparison with traditional pathology workflow.

-> We agree with your opinion. However, the specific numerical performance has never been disclosed to the pathologists surveyed. They responded to the survey without knowing that the program had errors or limitations. Therefore, we think it is appropriate not to present the actual performance of the program.

          Because I am not able to access the program (the link to the supplementary files is not working), I have no idea about whether the program is user-friendly or not, that certainly will affect the willingness of pathologists to adopt digital pathology in their routine work. I suggest the authors to demonstrate a few crucial screen-shots of PROMISE-P so that the readers can understand how the software program performs and achieves the goal.

-> Thank you for thoughtful recommendation. We added aditional screen-shots guideline for user as a supplementary material 2 (promise-p user guide.pdf).

Author Response File: Author Response.pdf

Reviewer 2 Report

Digital pathology has been actively implemented in recent years in many hospitals around the world. Alongside, there have been many discussions on the importance and effectiveness of digital pathology and computer-aided diagnosis. In this work, Nam et al. surveyed Korean pathologists to gain information on preferences and demands for digital pathology and computer-aided diagnosis using the example of prostate needle biopsy.

 

General comment:

This study adds to the knowledge on the needs of digital pathology and computer aided diagnosis in improving diagnostic procedure in pathology based on the example of prostate needle biopsy.

 

Minor points:

Abstract: Since a part of contacted pathologists did not take the survey, it would be more precise to say ''Korean pathologists that participated in the survey had highly positive perceptions of digital pathology and maintained a positive attitude toward the use of CAD software.''

Table 2: In the footnote, the authors say: ''The sample size for each position may not have been large enough, so careful interpretation is required.'' It would be good if the positions with sample sizes that are statistically not large enough are denoted.

Figure 1f and 1g: Does ''Normal'' reffer to ''Somewhat in need'' from Figure 1a? Maybe it could be re-phrased, so that it is more clear.

Figure 2 does not bring a lot of additional information in comparison to Figure 1e and 1g and it could be omitted, since the main information that it brings is well described in the accompanying text from the Results section. Instead of reffering to Figure 2, the authors could write ''(data not shown)'' at the end of the sentence.

Figure 4d and 4e are switched, since their contents do not correspond to the main text and the text from the figure captions.

In regard to PROMISE-P, better evaluation would have probably been achieved if only uropathologist had been surveyed and if their number was higher. This could be briefly mentioned in the penultimate or ultimate paragraph of the Discussion.

Font size on some Figures is not large enough, so it is hard to read the information. This should be improved. 

Author Response

Thank you for kind recommendation. We have changed all indications as you suggested. Also, we entrusted the English languige editing to a professional company. Please see the attached proof.

Author Response File: Author Response.pdf