List of Papers Browse by Subject Areas Author List
Abstract
Ultrasound-guided fine needle aspiration (FNA) biopsy is a common minimally invasive diagnostic procedure. However, an aspiration needle tracker addressing rapid reciprocating motion is still missing. MrTrack, an aspiration needle tracker with a mamba-based register mechanism, is proposed. MrTrack leverages a Mamba-based register extractor to sequentially distill global context from each historical search map, storing these temporal cues in a register bank. The Mamba-based register retriever then retrieves temporal prompts from the register bank to provide external cues when current vision features are temporarily unusable due to rapid reciprocating motion and imaging degradation. A self-supervised register diversify loss is proposed to encourage feature diversity and dimension independence within the learned register, mitigating feature collapse. Comprehensive experiments conducted on both robotic and manual aspiration biopsy datasets demonstrate that MrTrack not only outperforms state-of-the-art trackers in accuracy and robustness but also achieves superior inference efficiency. Project page: https://github.com/PieceZhang/MrTrack
Links to Paper and Supplementary Materials
Main Paper (Open Access Version): https://papers.miccai.org/miccai-2025/paper/0251_paper.pdf
SharedIt Link: Not yet available
SpringerLink (DOI): Not yet available
Supplementary Material: https://papers.miccai.org/miccai-2025/supp/0251_supp.zip
Link to the Code Repository
https://github.com/PieceZhang/MrTrack
Link to the Dataset(s)
N/A
BibTex
@InProceedings{ZhaYue_MrTrack_MICCAI2025,
author = { Zhang, Yuelin and Ding, Qingpeng and Lei, Long and Feng, Yongxuan and Tang, Raymond Shing-Yan and Cheng, Shing Shin},
title = { { MrTrack: Register Mamba for Needle Tracking with Rapid Reciprocating Motion during Ultrasound-Guided Aspiration Biopsy } },
booktitle = {proceedings of Medical Image Computing and Computer Assisted Intervention -- MICCAI 2025},
year = {2025},
publisher = {Springer Nature Switzerland},
volume = {LNCS 15960},
month = {September},
page = {410 -- 420}
}
Reviews
Review #1
- Please describe the contribution of the paper
The author presented a Mamba-based need tracking framework for ultrasound-guided fine needle aspiration. Additionally, a self-supervised register diversification loss has been proposed in this manuscript.
- 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 Mamba-based register extraction-retrieval framework is innovative
- The register diversify loss is an unique approach
- Comparison results with state-of-the-art is promising
- 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.
- Free parameters used for the loss function and the training parameters are not tuned (no selection criteria mentioned)
- The author did not mention anything about the ultrasound imaging parameters such as gain, dynamic range, frequency, etc., effect of these parameters may improve the generalizability of the proposed method.
- Statistical significance in the comparison should include to proof the effect of the proposed method over the state-of-the-art.
- 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 mention open access to source code or data but provides a clear and detailed description of the algorithm to ensure 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?
The needle tracking framework and its loss function are unique and innovative. In future the method can be utilized for clinical applications. But a detailed experiment is needed, such as tuning the free parameters, including training parameters. Statistical significance estimation across all the comparison methods can prove the performance difference.
- 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 authors propose a method based on the Mamba framework for tracking and detection of fine needles during aspiration biopsy. This task is typically hard even for state of the art detection methods, because of the fast and non-linear motion of fine needles, and the characteristics of ultrasound imaging. In this paper the Mamba framework (which already showed promising results in other computer vision tasks) is augmented with registers to enhance its temporal tracking. The method is extensively validated on datasets generated with automated motion of the needle and probe, as well with ex-vivo trials by human operators. The paper serves as an introduction to this quite novel framework for computer vision tasks, which could be useful to a good number of members of the CAI community.
- 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 document is of very high quality. Every section is clear and complete. The evaluation is thorough and the presentation of the results is cristalline. The method represents the first introduction of a very promising tool to research in CAI, and it may inspire many to overcome the current state of the art in this class of problems.
- 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.
None to report.
- 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 mention open access to source code or data but provides a clear and detailed description of the algorithm to ensure 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
Thank you for your work, it was a pleasure to read the paper.
Only a minor remark, since the manual insertion dataset was only used for testing and it showed good performance. Were all automated insertion videos used both for training and validation, or were they split to avoid overfitting?
- 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 deserves to be known as much as possible among the community. It may represent an important new tool to tackle challenging problems. The quality of the document and the thoroughness of the validation are exemplary.
- 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 presents a Mamba-based needle tracking approach, termed MrTrack, designed to address rapid reciprocating needle motion under ultrasound guidance. The method introduces an innovative Mamba-based registration mechanism, which outperforms state-of-the-art approaches in experiments conducted on an agar phantom embedded with pork tissue. Overall, the work demonstrates strong clinical relevance and aligns well with the interests of the MICCAI community. While further clinical studies are necessary to validate its effectiveness, the proposed approach shows clear potential and contributes meaningfully to advancements in the field of needle tracking.
- 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.
- This paper is well written and structured, making it easy to follow.
- The proposed approach shows its novelty, and the rationale behind the proposed components is well justified.
- The experimental design is representative and clinically relevant, effectively reflecting real-world scenarios to a certain extent.
- This work shows strong clinical value, and shows promising potential to enhance various needle tracking tasks in US guidance.
- 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.
-
A major concern is how the proposed approach accounts for tissue motion caused by patient breathing or discomfort during clinical application. In the current experimental setup, the phantom is relatively stationary, making the needle the primary source of motion. However, in real-world scenarios, organ motion—particularly in the liver or kidney—can be significant, and corresponding changes in ultrasound image appearance, including potential artifacts, may occur. It is unclear whether the Mamba-based mechanism can effectively handle such variability. The authors should discuss the applicability and robustness of their approach under realistic conditions involving tissue motion and ultrasound artifacts.
- For transparency, the in-plane-static (IPS) and in-plane-moving (IPM) techniques need to be clearly explained. The reviewer cannot find the definitions from the mentioned reference [14].
- In line 12 Page 8, the author mentioned a challenging case involving poor needle visibility. For some out-of-plane cases (in a short time), the needle may disappear completely or partially. The reviewer suggests that the authors provide a more detailed discussion on how the proposed method handles such cases.
- For reproducibility and transparency, the authors should make the code publicly available.
-
- 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 mention open access to source code or data but provides a clear and detailed description of the algorithm to ensure 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
Minors:
- Line 29, Page 2, “…using temporal context. creating a dynamic template with long-term …” The period between “context” and “creating” should be a comma.
- 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 proposed approach shows clear potential and contributes meaningfully to advancements in the field of needle tracking
- 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
Author Feedback
We would like to thank all reviewers for their thoughtful and constructive feedback. We are encouraged by the recognition of our work’s novelty, clinical potential, and strong experimental design.
Regarding the concerns about hyper-parameter selection and ultrasound imaging parameters raised by R1, we agree that tuning these parameters may improve model performance and generalizability. However, we would like to highlight that the model can achieve satisfactory performance without extensive hyper-parameter search. Therefore, we used default settings or simple values without tuning (e.g., alpha = beta = 0.01). We will add discussion on this topic in the camera-ready version. The missing information about the ultrasound imaging system will also be included.
For R2’s question about data splitting, we clarify that there was no overlap between the training and testing sets, as described in the data splitting scheme in Section 4.1, where the automated insertion data was split into training, validation, and testing sets in a ratio of 7:1:2.
Regarding the concern about handling tissue/organ motion raised by R3, we appreciate this important point. We are actively working with clinicians on needle tracking under realistic or simulated organ motion, which will be our next stage of research. We will add discussion on these challenges and our future plans in the revision. For the clarification of IPS/IPM, we will explicitly define these terms in the revised manuscript and update the reference. Regarding the question of how the model handles poor needle visibility or out-of-plane cases, currently the model cannot explicitly address situations where the needle completely disappears. In our experiments, MrTrack leverages historical temporal context via the register bank, which helps maintain tracking robustness during short-term occlusions or visibility loss. However, we recognize that when the needle is completely out-of-plane or invisible for an extended period, the tracker’s performance may degrade, as there is currently no explicit template updating or recovery strategy. In the revised manuscript, we will clarify this limitation and further discuss potential solutions, such as incorporating confidence-based template updates or integrating out-of-plane detection modules, to enhance the robustness of MrTrack under these challenging conditions.
We appreciate the suggestions from all reviewers and believe these will help guide future improvements to our framework.
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