Authors

Yihao Xia, Yonggang Shi

Abstract

The inter-site variability of diffusion magnetic resonance imaging (dMRI) hinders the aggregation of the dMRI data from multiple centers. This necessitates dMRI harmonization for removing the non-biological site-effects. Recently, the emergence of high-resolution dMRI data across various connectome imaging studies allows the large-scale analysis of cortical micro-structure. Existing harmonization methods, however, perform poorly in the harmonization of dMRI data in cortical areas because they rely on image registration methods to factor out anatomical variations, which have known difficulty in aligning cortical folding patterns. To overcome this fundamental challenge in dMRI harmonization, we propose a framework of personalized dMRI harmonization on the cortical surface to improve the dMRI harmonization of gray matter by adaptively estimating the inter-site harmonization mappings. In our experiments, we demonstrate the effectiveness of the proposed method by applying it to harmonize dMRI across the Human Connectome Project (HCP) and the Lifespan Human Connectome Projects in Development (HCPD) studies and achieved much better performance in comparison with conventional methods based on image registration.

Link to paper

DOI: https://link.springer.com/chapter/10.1007/978-3-031-16446-0_68

SharedIt: https://rdcu.be/cVRUc

Link to the code repository

N/A

Link to the dataset(s)

N/A

Reviews

Review #1

Please describe the contribution of the paper

Harmonizing site-dependent effects on diffusion MRI is critical in multi-site clinical studies. Most methods take a volume-space-based approach to harmonize the imaging measures in a reference space. But these methods are not optimal to study cortical gray matters because of the heterogeneous structures of cortical surfaces. This work introduces a method to harmonize diffusion MRI measures at the cortical surface for individual subjects. The method is based on a distance measure to find corresponding vertices on the surface. The performance of the method is evaluated based on the HCP and HCPD data sets.
Please list the main strengths of the paper; you should write about a novel formulation, an original way to use data, demonstration of clinical feasibility, a novel application, a particularly strong evaluation, or anything else that is a strong aspect of this work. Please provide details, for instance, if a method is novel, explain what aspect is novel and why this is interesting.

1) The problem of the harmonization of surface-based imaging data is very relevant in clinical studies. To the best of my knowledge, this is the first work to consider this problem. 2) A method is developed to analyze inter-subject local correspondence which includes the geometric properties of the surface and the cortical thickness information.
Please list the main weaknesses of the paper. Please provide details, for instance, if you think a method is not novel, explain why and provide a reference to prior work.

1) The proposed local correspondence matching method only considers the geometric information and cortical thickness which does not require diffusion MRI data. It is not clear why matched geometry indicates matched diffusion MRI measures. 2) The “grayordinate” format as included in the cifiti data in HCP has become standard for surface-based analysis. It is not clear why that is not an option for surface-based harmonization.
Please rate the clarity and organization of this paper

Good
Please comment on the reproducibility of the paper. Note, that authors have filled out a reproducibility checklist upon submission. Please be aware that authors are not required to meet all criteria on the checklist - for instance, providing code and data is a plus, but not a requirement for acceptance

I don’t see any limitations on the reproducibility.
Please provide detailed and constructive comments for the authors. Please also refer to our Reviewer’s guide on what makes a good review: https://conferences.miccai.org/2022/en/REVIEWER-GUIDELINES.html

1) It will be helpful to compare with the standard cifity surface-based analysis. 2) The diffusion MRI data from HCP has a very high spatial resolution. For standard clinical data, the resolution will be much lower. Then the partial-volume effect will be a significant problem for surface based analysis. Adding more comments or a solution to this problem will certainly improve this work.
Rate the paper on a scale of 1-8, 8 being the strongest (8-5: accept; 4-1: reject). Spreading the score helps create a distribution for decision-making

5
Please justify your recommendation. What were the major factors that led you to your overall score for this paper?

This work points out an interesting question on dMRI harmonization. But the proposed method is not convincing since dMRI data was not naturally used to define local correspondence and there is no comparison with the standard cifity format.
Number of papers in your stack

3
What is the ranking of this paper in your review stack?

2
Reviewer confidence

Very confident
[Post rebuttal] After reading the author’s rebuttal, state your overall opinion of the paper if it has been changed

N/A
[Post rebuttal] Please justify your decision

N/A

Review #2

Please describe the contribution of the paper

The paper introduces a surface-based harmonization method to reduce inter-site variation in diffusion-weighted images.
Please list the main strengths of the paper; you should write about a novel formulation, an original way to use data, demonstration of clinical feasibility, a novel application, a particularly strong evaluation, or anything else that is a strong aspect of this work. Please provide details, for instance, if a method is novel, explain what aspect is novel and why this is interesting.

The idea of reducing cortical mismatch for better harmonization is novel. The paper is well-written. The authors stated the problem and motivation well and clear.
Please list the main weaknesses of the paper. Please provide details, for instance, if you think a method is not novel, explain why and provide a reference to prior work.

No statistical test was done to prove the significance of the numerical results. There is little to no discussion about the limitation of the work.
Please rate the clarity and organization of this paper

Very Good
Please comment on the reproducibility of the paper. Note, that authors have filled out a reproducibility checklist upon submission. Please be aware that authors are not required to meet all criteria on the checklist - for instance, providing code and data is a plus, but not a requirement for acceptance

Study can be reproduced
Please provide detailed and constructive comments for the authors. Please also refer to our Reviewer’s guide on what makes a good review: https://conferences.miccai.org/2022/en/REVIEWER-GUIDELINES.html
1. The authors stated “Alternatively, surface-based registration can alleviate some of this anatomy misalignment problem for dMRI harmonization, but it is still insufficient to resolve this challenge.”. Please explain why surface-based registration is not sufficient to solve cortical mismatch or at least provide reference for the statement.
2. Are the improvements in Table 1 significant? Please provide the p-value of the test.
3. Is there any explanation that HCPD harmonization task is always better than HCP harmonization task?
4. The authors only harmonize b=3000 shell, then how FA and MD was calculated? Only from b=3000 shell or from the unharmonize b=1000 shell?
Rate the paper on a scale of 1-8, 8 being the strongest (8-5: accept; 4-1: reject). Spreading the score helps create a distribution for decision-making

6
Please justify your recommendation. What were the major factors that led you to your overall score for this paper?

Novel surface-based DWIs harmonization. Minor weakness due to lack of statistical test for numerical results.
Number of papers in your stack

4
What is the ranking of this paper in your review stack?

2
Reviewer confidence

Confident but not absolutely certain
[Post rebuttal] After reading the author’s rebuttal, state your overall opinion of the paper if it has been changed

N/A
[Post rebuttal] Please justify your decision

N/A

Review #4

Please describe the contribution of the paper

dMRI harmonization method that personalizes inter-site mappings
Please list the main strengths of the paper; you should write about a novel formulation, an original way to use data, demonstration of clinical feasibility, a novel application, a particularly strong evaluation, or anything else that is a strong aspect of this work. Please provide details, for instance, if a method is novel, explain what aspect is novel and why this is interesting.
- Well-motivated
- Decently written
- Well structured
- Strong study design
- Strong technical implementation
- Neat figures
Please list the main weaknesses of the paper. Please provide details, for instance, if you think a method is not novel, explain why and provide a reference to prior work.
- Certain colloquial language throughout the submission can be substituted
Please rate the clarity and organization of this paper

Very Good
Please comment on the reproducibility of the paper. Note, that authors have filled out a reproducibility checklist upon submission. Please be aware that authors are not required to meet all criteria on the checklist - for instance, providing code and data is a plus, but not a requirement for acceptance

Not reproducible - although publicly available datasets are used, implementation code has not been made available
Please provide detailed and constructive comments for the authors. Please also refer to our Reviewer’s guide on what makes a good review: https://conferences.miccai.org/2022/en/REVIEWER-GUIDELINES.html
- Recommend authors to avoid colloquial language
- Evaluate on more datasets
Rate the paper on a scale of 1-8, 8 being the strongest (8-5: accept; 4-1: reject). Spreading the score helps create a distribution for decision-making

6
Please justify your recommendation. What were the major factors that led you to your overall score for this paper?

Strong study design and implementation
Number of papers in your stack

2
What is the ranking of this paper in your review stack?

1
Reviewer confidence

Not Confident
[Post rebuttal] After reading the author’s rebuttal, state your overall opinion of the paper if it has been changed

N/A
[Post rebuttal] Please justify your decision

N/A

Primary Meta-Review

Please provide your assessment of this work, taking into account all reviews. Summarize the key strengths and weaknesses of the paper and justify your recommendation. In case you deviate from the reviewers’ recommendations, explain in detail the reasons why. In case of an invitation for rebuttal, clarify which points are important to address in the rebuttal.

The paper discusses a new framework for diffusion MRI harmonization for individuals in a surface-based framework. This is a critical task with the appearance of more and more multi-site data collection projects and the potential weakness of volumetric registration-based methods to accurately align cortical areas. The proposed framework is evaluated on the publicly available HCP and HCPD data sets and presents good preliminary results. The paper is clearly written and well-motivated.

Reproducibility: data is shared, but code is not available.

Discussion regarding how this approach would apply and perform in the case of clinical standard diffusion data as well as details about method limitations are missing.

In the future, a comparison to surface-based analysis with grayordinates, as provided in the HCP cohort, would be recommended. Also, comparison with solutions using diffusion-derived information and comprehensive statistical testing to demonstrate significance would be needed to make this a stronger work.
What is the ranking of this paper in your stack? Use a number between 1 (best paper in your stack) and n (worst paper in your stack of n papers). If this paper is among the bottom 30% of your stack, feel free to use NR (not ranked).

2

Author Feedback

N/A

back to top

Personalized dMRI Harmonization on Cortical Surface