List of Papers Browse by Subject Areas Author List
Abstract
Identification of microvascular obstruction (MVO) in acute myocardial infarction patients is critical for prognosis and has a direct link to mortality risk. Current approaches using late gadolinium enhancement (LGE) for contrast-enhanced cardiac magnetic resonance (CMR) pose risks to the kidney and may not be applicable to many patients. This highlights the need to explore alternative non-contrast imaging methods, such as cine CMR, for MVO identification. However, the scarcity of datasets and the challenges in annotation make the MVO identification in cine CMR challenging and remain largely under-explored. For this purpose, we propose a non-contrast MVO identification framework in cine CMR with a novel coarse-grained mask regularization strategy to better utilize information from LGE annotations in training. We train and test our model on a dataset comprising 680 cases. Our model demonstrates superior performance over competing methods in cine CMR-based MVO identification, proving its feasibility and presenting a novel and patient-friendly approach to the field. The code is available at https://github.com/code-koukai/MVO-identification.
Links to Paper and Supplementary Materials
Main Paper (Open Access Version): https://papers.miccai.org/miccai-2024/paper/0670_paper.pdf
SharedIt Link: https://rdcu.be/dVZeo
SpringerLink (DOI): https://doi.org/10.1007/978-3-031-72378-0_22
Supplementary Material: https://papers.miccai.org/miccai-2024/supp/0670_supp.zip
Link to the Code Repository
https://github.com/code-koukai/MVO-identification
Link to the Dataset(s)
N/A
BibTex
@InProceedings{Yan_CoarseGrained_MICCAI2024,
author = { Yan, Yige and Cheng, Jun and Yang, Xulei and Gu, Zaiwang and Leng, Shuang and Tan, Ru San and Zhong, Liang and Rajapakse, Jagath C.},
title = { { Coarse-Grained Mask Regularization for Microvascular Obstruction Identification from non-contrast Cardiac Magnetic Resonance } },
booktitle = {proceedings of Medical Image Computing and Computer Assisted Intervention -- MICCAI 2024},
year = {2024},
publisher = {Springer Nature Switzerland},
volume = {LNCS 15001},
month = {October},
page = {231 -- 241}
}
Reviews
Review #1
- Please describe the contribution of the paper
-
This paper proposes a novel approach for MVO identification by extracting spatiotemporal features from non-contrast cine CMR, which has not been well-explored previously.
-
This paper introduces a coarse-grained mask regularization strategy to leverage information from LGE data.
-
- 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 structure of the paper is good. The experiment results are good. The ablation studies also demonstrate the effectiveness of the method. The authors also conduct the parameter analysis.
- 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.
The logic in paragraph one can be improved. Many important papers in related works might be missed. The definition of microvascular obstruction is not clear. The clinical significance of microvascular is not clear.
- Please rate the clarity and organization of this paper
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 authors claimed to release the source code and/or dataset upon acceptance of the submission.
- Do you have any additional comments regarding the paper’s reproducibility?
good
- 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
Please improve the paper writting of Paragraph 1. The authors should add more citations about the studies on Non-Contrast detection. The clinical significance of Microvascular Obstruction Identification should be more clear.
- 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 Accept — could be accepted, dependent on rebuttal (4)
- Please justify your recommendation. What were the major factors that led you to your overall score for this paper?
The paper presentation and writting are good. The experiments are extensive.
- 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
This paper proposes a novel method for identifying microvascular obstruction (MVO) from non-contrast cine cardiac MRI, addressing the risks associated with contrast agents like gadolinium. The authors introduce a coarse-grained mask regularization strategy leveraging information from late gadolinium enhancement (LGE) annotations to enhance non-contrast imaging capabilities. This method is tested on a dataset of 680 cases and demonstrates superior performance over existing methods.
- 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 topic is highly relevant as it addresses a significant clinical need to reduce gadolinium usage due to its potential risks. The proposed method is innovative, particularly the coarse-grained mask regularization that uses existing LGE data to improve MVO detection in non-contrast cine CMR.
-
The methodology is technically robust, integrating advanced concepts like residual 2D+1D network (R(2+1)D) and BERT for processing spatiotemporal features and temporal refinement, respectively. The multi-task learning approach and the innovative use of a coarse-grained mask for regularization are well thought out and effectively implemented.
- The authors provide comprehensive experimental results, including comparisons with several other methods and ablation studies, which validate the effectiveness of their approach. The use of standard metrics like AUC, Specificity, Recall, and F1-score, along with a detailed dataset description, enhances the credibility of the results.
- 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.
-
The study is conducted on a relatively small dataset, which may limit the generalizability of the findings. More extensive validation on diverse datasets would be beneficial to establish the robustness of the method.
-
Some technical details are missing or unclear, such as the exact implementation nuances of the BERT and R(2+1)D components within the framework. More detailed explanations or pseudocode could enhance reproducibility and understanding.
-
Future work could explore the scalability of the method to other forms of cardiac imaging or its adaptation for real-time application. Additionally, investigating the impact of further increasing the dataset size or diversity could be valuable.
-
- 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 authors claimed to release the source code and/or dataset upon acceptance of the submission.
- Do you have any additional comments regarding the paper’s reproducibility?
N/A
- 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
- The paper presents an innovative approach to mask regularization but lacks detailed explanation about the transition from coarse-grained masks to the training process. Clarifying how these masks are utilized during the training to influence the learning process, and any strategies used to handle misalignments could provide better insights into the effectiveness of your approach. 2.Considering the clinical importance of MVO identification, discussing the potential of this method for real-time application in clinical settings could be very impactful. Insights into computational requirements, inference time, and potential integration with clinical imaging systems would be valuable additions.
- 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 Accept — could be accepted, dependent on rebuttal (4)
- Please justify your recommendation. What were the major factors that led you to your overall score for this paper?
Clinical Impact
- 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 #3
- Please describe the contribution of the paper
This paper proposes a framework for identifying MVO using non-contrast cine CMR.
Contributions:
- It introduces a novel approach to identify MVO by extracting spatiotemporal features from non-contrast cine CMR, which has not been explored extensively before.
- To enhance model training, a coarse-grained masked regularization strategy is introduced to leverage information from LGE data.
- Experimental results demonstrate the feasibility of identifying MVO from non-contrast cine CMR.
- 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.
-
This paper explores a new task that has the potential to fill the gap left by existing methods reliant on harmful contrast agents, which holds promising clinical prospects.
-
The introduction of a coarse-grained mask regularization strategy in this paper addresses a practical challenge by converting pixel-wise yet misaligned location information into a block-wise regional context, aiming to mitigate the impact of misalignment between CMR and LGE data. This idea is intriguing.
-
The writing is well-structured, with clear explanations of the roles of each module, making it very easy to understand.
-
- 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.
-
Although the task is novel, it seems that this paper may not be the first to tackle it. The Introduction section could provide more background on related work addressing the same task.
-
Regarding the Coarse-Grained Mask Regularization idea, its effectiveness could be further validated through experiments. For example, intentionally using data with misalignment effects and observing whether the regularization strategy has any impact would provide valuable insights.
-
- 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 authors claimed to release the source code and/or dataset upon acceptance of the submission.
- Do you have any additional comments regarding the paper’s reproducibility?
N/A
- 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
Refer to the weaknesses section (6).
Curiosity strikes: Can this method currently be used to identify the presence of MVO (classification), and is it possible to further segment the MVO area on CMR later? What challenges does this face?
- 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?
This paper explores the feasibility of using CMR to identify MVO, which holds significant clinical application value. The framework proposed in this paper is well-suited for this task and introduces an intriguing coarse-grained mask regularization strategy. The writing is clear, and the experimental results validate the roles of each module effectively.
- 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
1) Acknowledgement: We sincerely appreciate the time and effort that all the reviewers have invested in reviewing our manuscript. We are grateful for the positive feedback on the good idea (R3, R4), clinical impacts (R3, R4), comprehensive experiment and the good results (R1, R3), and good writing structure (R1, R4).
2) Concerns: On the clarity of technical details (Reviewer #3), the logic in paragraph one, and the definition and clinical significance of MVO (Reviewer #1): We thank the reviewers for the comments. Influenced by your valuable suggestions, we are committed to refining our paper for the final version. On the effectiveness and generalizability of the method, and future scalability (Reviewer #3 & Reviewer #4): The reviewers’ advice is highly insightful, highlighting specific areas for improvement. We will incorporate these suggestions and aim for broader experiments in future research.
Meta-Review
Meta-review not available, early accepted paper.