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

Abstract

Cerebral microbleeds (CMBs) are small hemorrhagic lesions that pose significant challenges for accurate segmentation due to the high rate of false positives and false negatives. CMBs have two subtypes: lobar and deep microbleeds (MBs). Motivated by the strong association between deep MBs and hypertension, we propose a blood pressure-driven nnU-Net (BP-nnUNet) that integrates blood pressure (BP) prompt into the state-of-the-art nnU-Net framework through three key strategies. First, we estimate BP using the pre-trained Meta-matching model, that requires only MRI images. This allows our method to be successfully applied to public datasets with missing clinical demographics. Second, we categorize CMBs into lobar and deep MB, enriching input text prompts with multiple classes while constraining the BP effect to deep MBs. Lastly, we introduce a novel anatomically-aware joint prompt fusion module that combines lobar and deep MB prompts. Experiments on both in-house and public datasets demonstrate that our BP-nnUNet outperforms existing CMB segmentation models and universal models incorporating medical prompts. Ablation studies validate the effectiveness of integrating subtype-level and case-level prompts, as well as our fusion module. Our method paves the way for the incorporation of clinically relevant information into a segmentation framework. Our code is available at https://github.com/junmokwon/BP-nnUNet.

Links to Paper and Supplementary Materials

Main Paper (Open Access Version): https://papers.miccai.org/miccai-2025/paper/0187_paper.pdf

SharedIt Link: Not yet available

SpringerLink (DOI): Not yet available

Supplementary Material: Not Submitted

Link to the Code Repository

https://github.com/junmokwon/BP-nnUNet

Link to the Dataset(s)

https://valdo.grand-challenge.org/

BibTex

@InProceedings{KwoJun_Blood_MICCAI2025,
        author = { Kwon, Junmo and Kim, Jonghun and Kim, Taehyeon and Seo, Sang Won and Cho, Hwan-ho and Park, Hyunjin},
        title = { { Blood Pressure Assisted Cerebral Microbleed Segmentation via Meta-matching } },
        booktitle = {proceedings of Medical Image Computing and Computer Assisted Intervention -- MICCAI 2025},
        year = {2025},
        publisher = {Springer Nature Switzerland},
        volume = {LNCS 15960},
        month = {September},
        page = {78 -- 88}
}

Reviews

Review #1

  • Please describe the contribution of the paper

    The authors proposed a joint prompt of medical text and BP data to fully leverage the clinical characteristics of CMB subtypes.

  • Please list the major strengths of the paper: you should highlight 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.

    Good writing and story telling.

  • Please list the major weaknesses of the paper. Please provide details: for instance, if you state that a formulation, way of using data, demonstration of clinical feasibility, or application is not novel, then you must provide specific references to prior work.
    1. Figure 4 shows very good segmentation performance but in the table 2 the DSC score is not very high, I want to see some failure cases.
    2. In Figure 4, I also want to point it out the proposed method did well for cases In house #2 and VALDO #2. If the ablation results could show some visualizations would be better since the scores did not show too much.
    3. In table 2, why text prompt make the dsc lower in the in-house data?
    4. Since the microbleeds have two types, I would like to know which type has better segmentation performance since they use different strategy.
    5. MICCAI 2021 VALDO challenge task2 only provides binary segmentation mask (0: background, 1: microbleed). How do you define the lobar and deep microbleed?
  • 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 submission does not provide sufficient information for reproducibility.

  • Optional: If you have any additional comments to share with the authors, please provide them here. Please also refer to our Reviewer’s guide on what makes a good review and pay specific attention to the different assessment criteria for the different paper categories: https://conferences.miccai.org/2025/en/REVIEWER-GUIDELINES.html

    N/A

  • 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.

    (4) Weak Accept — could be accepted, dependent on rebuttal

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

    Method and writing

  • Reviewer confidence

    Very confident (4)

  • [Post rebuttal] After reading the authors’ rebuttal, please state your final opinion of the paper.

    N/A

  • [Post rebuttal] Please justify your final decision from above.

    N/A



Review #2

  • Please describe the contribution of the paper

    The paper combines several recent innovations to create a microbleed segmentation method that performs very well. The method, BP-nnUNet, takes a T1-weighted MRI and predicts BP features using a pre-trained meta-matching model, and then uses BiomedCLIP to create features that are used to highlight lobar and deep microbleeds as separate classes in the final segmentation. Results indicate that all aspects of the model are contributing towards its good performance in two datasets (an in-house one and the VALDO2021 challenge).

  • Please list the major strengths of the paper: you should highlight 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 major strengths include:

    • Strong quantitative results (especially DSC and F1), with clear improvement across multiple metrics in two datasets compared to recent SOTA methods.

    • Good ablation studies showing the benefits of the different aspects of the method (e.g. meta-matching BP features, text prompts, fusion architecture).

    • Novelty of combining the BP features with BiomedCLIP to create separate lobar and deep microbleed predictions.

  • Please list the major weaknesses of the paper. Please provide details: for instance, if you state that a formulation, way of using data, demonstration of clinical feasibility, or application is not novel, then you must provide specific references to prior work.

    • One main weakness is that the lobar and deep microbleed segmentations are never evaluated separately, as it would be good to know the relative performance in these two categories, as that is clinically relevant.

    • Overall the paper is well written and sufficient detail is provided, although it would have been good to know a little more about the size of the internal features that were being fused.

  • 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 submission has provided an anonymized link to the source code, dataset, or any other dependencies.

  • Optional: If you have any additional comments to share with the authors, please provide them here. Please also refer to our Reviewer’s guide on what makes a good review and pay specific attention to the different assessment criteria for the different paper categories: https://conferences.miccai.org/2025/en/REVIEWER-GUIDELINES.html

    N/A

  • 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.

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

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

    The strength of the results coupled with the novelties, which are both technically and biologically interesting, make this an excellent paper.

  • Reviewer confidence

    Confident but not absolutely certain (3)

  • [Post rebuttal] After reading the authors’ rebuttal, please state your final opinion of the paper.

    N/A

  • [Post rebuttal] Please justify your final decision from above.

    N/A



Review #3

  • Please describe the contribution of the paper

    This paper uses blood pressure as a guide and support to improve the performance of cerebral microbleed segmentation in MRI images.

  • Please list the major strengths of the paper: you should highlight 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.Fusion of imaging and non-imaging data, in this case MRI and blood pressure (a physiological data). 2.Introduced a joint prompt fusion module that achieves better performance than cross attention.

  • Please list the major weaknesses of the paper. Please provide details: for instance, if you state that a formulation, way of using data, demonstration of clinical feasibility, or application is not novel, then you must provide specific references to prior work.

    I am concerned about the blood pressure estimates. Does the error in blood pressure estimation affect the performance of the final segmentation model? If possible, I would like the authors to add the effect of estimating blood pressure and measuring blood pressure on model performance.

    The methods don’t seem to be categorized specifically for Lobar Microbleeds and Deep Microbleeds? Can I assume that in general both subtypes of microbleeds are widespread, so the authors used joint prompt fusion.

  • 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 submission has provided an anonymized link to the source code, dataset, or any other dependencies.

  • Optional: If you have any additional comments to share with the authors, please provide them here. Please also refer to our Reviewer’s guide on what makes a good review and pay specific attention to the different assessment criteria for the different paper categories: https://conferences.miccai.org/2025/en/REVIEWER-GUIDELINES.html

    N/A

  • 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.

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

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

    The method and experiments.

  • Reviewer confidence

    Confident but not absolutely certain (3)

  • [Post rebuttal] After reading the authors’ rebuttal, please state your final opinion of the paper.

    N/A

  • [Post rebuttal] Please justify your final decision from above.

    N/A



Author Feedback

We deeply appreciate the reviewers for their positive assessment and constructive feedback. We will incorporate minor changes suggested by the reviewers into the final manuscript.

  1. The primary concern raised by all reviewers is that lobar and deep microbleeds (MBs) were not evaluated separately but rather in a binary cerebral microbleed (CMB) classification fashion. We acknowledge the need for evaluation in lobar and deep MBs and plan to address this in future work.
  2. We thank R1 for pointing out the need to add implementation details on internal feature fusion. We will add this information in the final manuscript.
  3. We thank R2 and R3 for their constructive feedback. R2 expressed concern regarding the absence of failure cases. While our proposed method does encounter some failure cases, they were not included in the manuscript due to the 8-page limit. We will provide a thorough discussion of failure cases in future work. R3 asked about the effect of error in blood pressure (BP) estimation. We agree that errors in BP estimation could impact our results. Unfortunately, neither an in-house dataset nor the public dataset used in our experiments provides actual BP measurements. We are actively seeking an alternative dataset with real measurements to enable a comparison with our estimated BPs.
  4. R2 asked how the binary CMB labels can be converted to lobar and deep MBs. We defined lobar and deep MBs using a proxy label as proposed in the previous work [1]. MBs located within the lobar brain regions are classified as lobar MBs. MBs surrounded by infratentorial and deep supratentorial regions are categorized as deep MBs. [1] Kwon et al. “Anatomically-Guided Segmentation of Cerebral Microbleeds in T1-weighted and T2*-weighted MRI”, MICCAI 2024.



Meta-Review

Meta-review #1

  • Your recommendation

    Provisional Accept

  • If your recommendation is “Provisional Reject”, then summarize the factors that went into this decision. In case you deviate from the reviewers’ recommendations, explain in detail the reasons why. You do not need to provide a justification for a recommendation of “Provisional Accept” or “Invite for Rebuttal”.

    N/A



back to top