Multi-View 3D Transesophageal Echocardiography Registration and Volume Compounding for Mitral Valve Procedure Planning

Round 1

Reviewer 1 Report

The manuscript demonstrates new technique and algorithm for reconstruction of mitral valve shapes from a set of ultrasound (echo-) images. Length measurement error between final reconstructed ultrasound and ground-truth CT images for two porcine valves considered as accordance criterion. Obtained results allow concluding high quality of proposed technique for ultrasound images reconstruction. Clinical ultrasound investigations may benefit from this work. However, following remarks can be done according to the text of the manuscript.

1. Table 1. Could you clarify what is exactly means “Chordae Measurement Absolute difference” and how it was calculated? As I understand it is a difference between chordae lengths measured from ultrasound and reference CT images. Why the error of this measurement is so big? Please, add some information when describes this table (Line 176) Moreover, it will be really useful to add in this table an information about relative difference of chordae length and its error or its absolute length for better understanding of proposed method accuracy. Besides, it is better to present results in Table 1 as 0.7 ± 0.6 for Valve 1 and 0.6 ± 0.6 for Valve 2 accordingly to measurements and errors theoretical background. Please, correct this values throughout the text (Line 19, Table 1, Line 256 …)
2. Figures 3 would benefit from colour scale added.
3. Line 185 and Figure 8. For the future works I would recommend to the authors to consider a possibility of using quantitative criterion for image quality comparison. Standard deviation or Euclidean distance can be considered as simplest criterion. Calculated quantitative criterion will better support a statement about higher quality of the reconstructed images.

In conclusion, I would say that the manuscript describes useful research and can be recommended for publication after the listed corrections.

Author Response

Thank you for your feedback. Your suggested revisions have been addressed as follows:

1. Additional descriptions of the chordae length measurements and the meaning of “Chordae Measurement Absolute difference” were added to the text in section 3.1 (line 176). Additionally, the absolute length of the chordae has also been included in the text to provide context for the measurement error. Results have been changed to be reported as 0.7 ± 0.6 for Valve 1 and 0.6 ± 0.6 for Valve 2 throughout the text.
2. A colour scale has been added to Figure 3
3. Future works will include quantitative criteria for image quality assessment. A brief description has been added describing the potential criterion to be used (line 249). We plan on doing a future study with a larger set of patient images where we will include image quality metrics such as contrast to noise ratio and image sharpness (variance of a Laplacian of Gaussian filter).

Reviewer 2 Report

Dear Authors 

the paper is well written and can be considered for publication in applied sciences. However, before acceptance, some topics have to be addressed, in particular regarding ultrasound applications for the oral tissues.

1) Please discuss the utility of ultrasound in evaluating the healing of soft tissues around extraoral bone grating site. Please cite PubMed ID31140209

2) Please discuss the utility of ultrasound in management of oral malformations regarding the temporomandibular joint, please cite PubMed ID26147813

3) Please discuss the utility of ultrasound in evaluating potential implants of stem cells in the oral cavity. Please cite PubMed ID33386051

Author Response

With all due respect, we do not feel that this review applies to our paper. We believe there has been a mistake in the submission of this review, as the suggested literature is focused on ultrasound for oral soft tissue applications, whereas our paper is about 3D cardiac ultrasound. As such, we do not feel the suggested references would be appropriate to include in our paper.