Paper Info Reviews Author Feedback Meta-Review Back to top
List of Papers Browse by Subject Areas Author List

Abstract

Tagged magnetic resonance imaging (MRI) has been successfully used to track the motion of internal tissue points within moving organs. Typically, to analyze motion using tagged MRI, cine MRI data in the same coordinate system are acquired, incurring additional time and costs. Consequently, tagged-to-cine MR synthesis holds the potential to reduce the extra acquisition time and costs associated with cine MRI, without disrupting downstream motion analysis tasks. Previous approaches have processed each frame independently, thereby overlooking the fact that complementary information from occluded regions of the tag patterns could be present in neighboring frames exhibiting motion. Furthermore, the inconsistent visual appearance, e.g., tag fading, across frames can reduce synthesis performance. To address this, we propose an efficient framework for tagged-to-cine MR sequence synthesis, leveraging both spatial and temporal information with relatively limited data. Specifically, we follow a split-and-integral protocol to balance spatial-temporal modeling efficiency and consistency. The light spatial-temporal transformer (LiST$^2$) is designed to exploit the local and global attention in motion sequence with relatively lightweight training parameters. The directional product relative position-time bias is adapted to make the model aware of the spatial-temporal correlation, while the shifted window is used for motion alignment. Then, a recurrent sliding fine-tuning (ReST) scheme is applied to further enhance the temporal consistency. Our framework is evaluated on paired tagged and cine MRI sequences, demonstrating superior performance over comparison methods.

Links to Paper and Supplementary Materials

Main Paper (Open Access Version): https://papers.miccai.org/miccai-2024/paper/0448_paper.pdf

SharedIt Link: pending

SpringerLink (DOI): pending

Supplementary Material: N/A

Link to the Code Repository

N/A

Link to the Dataset(s)

N/A

BibTex

@InProceedings{Liu_TaggedtoCine_MICCAI2024,
        author = { Liu, Xiaofeng and Xing, Fangxu and Bian, Zhangxing and Arias-Vergara, Tomas and Pérez-Toro, Paula Andrea and Maier, Andreas and Stone, Maureen and Zhuo, Jiachen and Prince, Jerry L. and Woo, Jonghye},
        title = { { Tagged-to-Cine MRI Sequence Synthesis via Light Spatial-Temporal Transformer } },
        booktitle = {proceedings of Medical Image Computing and Computer Assisted Intervention -- MICCAI 2024},
        year = {2024},
        publisher = {Springer Nature Switzerland},
        volume = {LNCS 15007},
        month = {October},
        page = {pending}
}

Reviews

Review #1

  • Please describe the contribution of the paper

    The study is designed to generate cine MR images from tagged MRI. The study introduces a video to video transformation network that can derive cine MR images from tagged cine sequence. The study present a spatial-temporal transformer (List^2) to exploit the local and global attention in motino sequence. Also present a recurrent sliding fine-tuning scheme to further enhance the temporal consistency

  • 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.Application of transformer based method to video to video synthesis task and better performance compared to convolution-based architecture.

    1. Interesting application for tagged cine to cine, and its clinical value.
  • 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. There are no justification about their model, and why they need it. Given the extensive use of attention-based transformers in the field, the novel contribution of the proposed network is not evident.
    2. Also, the author should include both theoretical and empirical justifications for why the proposed model could be suited to the specific challenges presented in this study.
    3. The result images do not stand out compared to other images. Moreover, the result images looks a bit blurry.
    4. The experimental design, focusing on 20 healthy controls pronouncing “a souk,” seems quite narrow and may not sufficiently demonstrate the model’s applicability to diverse real-world scenarios.
  • Please rate the clarity and organization of this paper

    Satisfactory

  • Please comment on the reproducibility of the paper. Please be aware that providing code and data is a plus, but not a requirement for acceptance.

    The submission does not provide sufficient information for reproducibility.

  • Do you have any additional comments regarding the paper’s reproducibility?

    Following the paper, I can understand the purpose of LiST^2 and how to implement. However, I failed to understand what is ReST and why it is necessary and how to implement it. It would be better if the author have explained it more thoroughly.

  • Please provide detailed and constructive comments for the authors. Please also refer to our Reviewer’s guide on what makes a good review. Pay specific attention to the different assessment criteria for the different paper categories (MIC, CAI, Clinical Translation of Methodology, Health Equity): https://conferences.miccai.org/2024/en/REVIEWER-GUIDELINES.html
    1. The images (Fig.2.) should be improved to begin. By looking at the images I find to see the improvement over other methods.
    2. The architecture does not seem to be novel. The author used local self-attention from Swin transformer which is already a very popular method.
    3. If the author concentrated on the problem itself, and validated under various scenarios (other pronunciations) it would be better rather than focusing on the network architecture itself
  • Rate the paper on a scale of 1-6, 6 being the strongest (6-4: accept; 3-1: reject). Please use the entire range of the distribution. Spreading the score helps create a distribution for decision-making

    Weak Reject — could be rejected, dependent on rebuttal (3)

  • Please justify your recommendation. What were the major factors that led you to your overall score for this paper?
    1. Lack clarity of the work.
    2. Not enough comparison studies.
    3. Unclear images and figures.
    4. Interesting application and it’s challenges
  • Reviewer confidence

    Confident but not absolutely certain (3)

  • [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 authors attempt to synthesize cine from tagged MR images using a light spatial-temporal transformer, which explicitly explores the complementary cross-frame information. The proposed method was demonstrated with speech MRI data, using paired tagged/cine MRI frames for training/testing the model.

  • 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. This work is practically significant.
    2. The method is novel, the manuscript is well written, the results are convincing.
  • 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.

    For specific critiques, please check the “detailed and constructive comments”.

  • Please rate the clarity and organization of this paper

    Very Good

  • Please comment on the reproducibility of the paper. Please be aware that providing code and data is a plus, but not a requirement for acceptance.

    The submission does not provide sufficient information for reproducibility.

  • Do you have any additional comments regarding the paper’s reproducibility?

    No.

  • Please provide detailed and constructive comments for the authors. Please also refer to our Reviewer’s guide on what makes a good review. Pay specific attention to the different assessment criteria for the different paper categories (MIC, CAI, Clinical Translation of Methodology, Health Equity): https://conferences.miccai.org/2024/en/REVIEWER-GUIDELINES.html
    1. Are the tagged frames aligned to the cine frames in preprocessing steps? Why or why not?
    2. The acquisition parameters (TR/TE/matrix size, etc) for tagged/cine MR imaging should be included in “Experiments and Results”.
    3. I would suggest discussing whether the proposed method can be used for synthesizing cardiac cine.
  • Rate the paper on a scale of 1-6, 6 being the strongest (6-4: accept; 3-1: reject). Please use the entire range of the distribution. Spreading the score helps create a distribution for decision-making

    Accept — should be accepted, independent of rebuttal (5)

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

    A good work with novel idea, clear writing, and convincing results.

  • Reviewer confidence

    Very confident (4)

  • [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 #3

  • Please describe the contribution of the paper

    The paper proposed a paired tagged-to-cine MRI translation method based on MRI sequence using video translation techniques. The paper developed an efficient model (called LiSTT) for video translation, incorporating LightViT, SwinTransformer, and temporal position embedding. The paper proposed a recurrent-style fine-tune stage (called ReST) for better temporal consistency. Five-cross validation experiments demonstrated the superiority of the proposed method over existing ones.

  • 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 paper is well-written and organized.
    2. The proposed LiSTT and ReST, which combined several recent techniques, showed promising improvements over existing methods.
    3. The effectiveness of the design components in the proposed model was mostly verified through experiments.
    4. The figures are clear and well-designed.
  • 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. Although the video translation quality was evaluated, the benefits for downstream tasks (e.g., motion measurement) are not verified, which is important to actual clinical applications.
    2. The method was only validated on a single dataset.
  • Please rate the clarity and organization of this paper

    Excellent

  • Please comment on the reproducibility of the paper. Please be aware that providing code and data is a plus, but not a requirement for acceptance.

    The submission does not mention open access to source code or data but provides a clear and detailed description of the algorithm to ensure reproducibility.

  • Do you have any additional comments regarding the paper’s reproducibility?

    The paper provided details on the algorithm. The techniques used in the paper are mostly open-source. Thus, the reproducibility should be fine. The data seems not available.

  • Please provide detailed and constructive comments for the authors. Please also refer to our Reviewer’s guide on what makes a good review. Pay specific attention to the different assessment criteria for the different paper categories (MIC, CAI, Clinical Translation of Methodology, Health Equity): https://conferences.miccai.org/2024/en/REVIEWER-GUIDELINES.html
    1. The paper mentioned DVDnet but cited FastDVDnet. I assume the two methods are different. Which method was implemented in the paper exactly?
    2. “During testing, translating one tagged MRI section to the corresponding cine MR images took about 0.2 seconds.” I would suggest adding a discussion on efficiency comparison, for example, computation/memory cost and model size of the methods.
  • Rate the paper on a scale of 1-6, 6 being the strongest (6-4: accept; 3-1: reject). Please use the entire range of the distribution. Spreading the score helps create a distribution for decision-making

    Accept — should be accepted, independent of rebuttal (5)

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

    The paper is clear. The details of the proposed method are provided. The paper combined several recent technologies to develop a video translation method applied to the tagged-to-cine MRI translation problem. The results are promising. Though there are some small weaknesses to address in future development, overall, this is a good paper.

  • Reviewer confidence

    Confident but not absolutely certain (3)

  • [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



Author Feedback

N/A



Meta-Review

Meta-review not available, early accepted paper.



back to top