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

Abstract

Intracardiac echocardiography (ICE) has the potential to play a crucial role in structural heart disease (SHD) interventions by providing high-quality imaging in real time, without many of the key drawbacks of established imaging modalities. However, ICE’s limited field-of-view (FoV) requires continuous readjustments of the catheter position to fully visualize the dynamic cardiac environment, which impairs spatial navigation and increases procedure time and complexity. Dynamic panoramic reconstruction can mitigate this limitation. However, state-of-the-art methods depend on precise catheter tracking, the accuracy of which is affected by the presence of noise and anatomical motion. While registration can correct these errors, existing approaches are computationally prohibitive for large imaging volumes due to repeated iterations over image data, further amplified by the added time dimension. To address these challenges, we present a novel method for truly dynamic panoramic reconstruction by leveraging the repetitive nature of cardiac motion under a cyclic environment assumption. To our knowledge, our method is the first to employ dynamic pose graph optimization (PGO) specifically designed for 4D ICE tracking. Our results demonstrate enhanced tracking accuracy and improved panoramic reconstruction quality, potentially providing real-time, dynamic anatomical guidance for clinicians. The improved alignment of overlapping ICE volumes and increased temporal tracking resolution represent a substantial advancement in 4D ICE imaging, enhancing navigation and decision-making during complex cardiac interventions.

Links to Paper and Supplementary Materials

Main Paper (Open Access Version): https://papers.miccai.org/miccai-2025/paper/4263_paper.pdf

SharedIt Link: Not yet available

SpringerLink (DOI): Not yet available

Supplementary Material: https://papers.miccai.org/miccai-2025/supp/4263_supp.zip

Link to the Code Repository

N/A

Link to the Dataset(s)

N/A

BibTex

@InProceedings{HerSeb_ICEPoGO_MICCAI2025,
        author = { Herz, Sebastian AND Wysocki, Magdalena AND Tristram, Felix AND Hickler, Julia AND Neary-Zajiczek, Lydia AND Hennersperger, Christoph AND Navab, Nassir AND Wörz, Stefan},
        title = { { ICE-PoGO: Improving Dynamic Panoramic Reconstruction of 4D ICE Imaging through Pose Graph Optimization } },
        booktitle = {proceedings of Medical Image Computing and Computer Assisted Intervention -- MICCAI 2025},
        year = {2025},
        publisher = {Springer Nature Switzerland},
        volume = {LNCS 15968},
        month = {September},
        page = {320 -- 330}
}

Reviews

Review #1

  • Please describe the contribution of the paper

    The main contribution of this paper is the development of a novel method for dynamic panoramic reconstruction in 4D intracardiac echocardiography (ICE), addressing the limitations of current techniques in structural heart disease interventions. Unlike existing approaches that rely on computationally intensive registration and are sensitive to tracking errors, this method introduces the dynamic pose graph optimization (PGO) for 4D ICE tracking, which uses the repetitive nature of cardiac motion.

  • 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 novelty of this paper lies primarily in the use of dynamic pose graph optimization to construct a motion model that facilitates the fusion of multiple echocardiographic image volumes. The proposed framework is validated using a beating heart phantom.

  • 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.

    However, my primary concern is the insufficient detail in the methodology section. Specifically, there is no explanation of the 3D registration method, which is essential for fusing multiple 3D echocardiographic volumes into a single panoramic image. Additionally, Figure 1 is difficult to interpret. Why are only five cardiac phases shown? How are the intra-phase graphs connected? There is no description of how the EM sensor was mounted on the ICE catheter.

    Furthermore, the experimental validation is limited, providing insufficient evidence to demonstrate the effectiveness of the proposed framework. First of all, electromagnetic (EM) tracking is susceptible to interference from large metal objects. Therefore, the experiments should include the placement of large metal objects in various locations to better simulate the real operating environment of a catheterization lab.

    The paper would benefit from a discussion on potential solutions for translating the proposed framework into a real clinical environment. Addressing practical challenges, such as integrating the system into existing clinical workflows, ensuring compatibility with standard equipment, and mitigating electromagnetic interference, would significantly strengthen the work. Overall, I believe the paper has some potential and would be better suited for an expanded version submitted to a relevant journal.

  • Please rate the clarity and organization of this paper

    Poor

  • 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.

    (2) Reject — should be rejected, independent of rebuttal

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

    I do not believe the current version should be accepted, as it lacks essential details on 3D image registration, provides an insufficient description of the methodology, and includes limited experimental validation.

  • 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 proposes a method for dynamic panoramic reconstruction in ICE that formulates the problem as a pose graph optimization problem. It assumes the motion of the heart is cyclic and builds one graph per cycle phase. These individual graphs are connected by inter-graph edges yielding an hyper pose graph that allows for the joint optimization of the poses across the different phases of the cardiac motion. Experiments performed on a phantom show improved accuracy when compared to the baseline approach.

  • 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 idea of formulating the dynamic panoramic reconstruction problem using pose graph optimization is novel.
    • The low runtime of the proposed method is a plus.
    • Decomposing the affine transformation to separate catheter motion from cardiac deformation is interesting.
    • The ablation study shows that the proposed method yields the best results and reveals that the proposed intra-phase graphs combined with sparse EM tracking provides the most significant accuracy improvement.
  • 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.
    • The fact that the proposed method only works for cyclic and well-behaved cardiac motions is a downside, limiting the applicability of the method.
    • Section 2.2: How is the assignment of US volumes to intra-phase graphs performed? Does it assume the heart rate of the patient is known and is constant? Please clarify this aspect.
    • Experiments on real patients, even if only qualitatively analysed, would add value to this submission
  • 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?

    Although the topic of this submission falls outside my area of expertise, I believe I understood the contribution of this paper and consider it novel and practically valuable. I have a few concerns described in the Weaknesses section that I would like the authors to address.

  • Reviewer confidence

    Somewhat confident (2)

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

    Accept

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

    I consider the authors’ rebuttal satisfactory and thus I am keeping my initial recommendation.



Review #3

  • Please describe the contribution of the paper

    The main contribution of this paper is the introduction of a novel method for dynamic panoramic reconstruction in 4D intracardiac echocardiography (ICE), which significantly enhances catheter tracking and spatial alignment across multiple cardiac phases. This is accomplished through the integration of pose graph optimization (PGO) into the 4D ICE imaging pipeline, enabling efficient and scalable alignment of a large number of 3D ultrasound volumes under realistic, dynamic physiological conditions.

    The paper is well written, with a clearly articulated clinical motivation and a technically sound, innovative solution. Most notably, the strong alignment between the clinical challenge and the proposed computational approach is commendable—a level of coherence that is increasingly rare.

  • 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. Novel Dynamic Panoramic Reconstruction Framework for 4D ICE This is the first method specifically designed for dynamic panoramic reconstruction in intracardiac echocardiography (ICE), addressing a critical need for real-time anatomical visualization during interventional procedures.

    2. Clever Use of Cardiac Physiology through the Cyclic Environment Assumption The cyclic environment assumption—that cardiac anatomy returns to a similar spatial configuration at the same point in each cardiac cycle—is leveraged to enforce temporal consistency across 4D ultrasound volumes. This allows the model to approximate the dynamic anatomy as “quasi-static” within a given phase, enabling meaningful global optimization across time. This is a clinically grounded insight that helps the authors build a computationally efficient, fully dynamic model.

    3. Scalable and Efficient Implementation with Real-Time Potential The method is demonstrated to be computationally feasible, with typical registration and optimization steps completing in seconds on a standard CPU setup. This opens the door to real-time or near real-time integration in the clinical workflow, which is a key requirement for interventional imaging technologies.

    4. Strong Quantitative and Qualitative Evaluation The authors construct a quantifiable phantom-based testbed, complete with 3D-printed cone landmarks and a cardiac-mimicking pump-driven silicone heart model. The ablation study clearly shows the impact of each system component—EM tracking, inter- and intra-phase edges, affine registration, etc.—demonstrating a rigorous understanding of the method’s inner workings.

  • 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.

    While the paper introduces a novel and well-executed method, several limitations should be acknowledged. The method relies on the assumption of cyclic cardiac motion, which may not hold in clinical scenarios involving irregular heart rhythms, and its robustness to such deviations is not assessed. Additionally, the evaluation is limited to comparisons with EM tracking and sequential registration, without benchmarking against recent learning-based methods, making it difficult to fully position the work within the current state of the art, e.g,:

    Mao, Z., Zhao, L., Huang, S., Fan, Y., Lee, A.P.W.: DSR: Direct Simultaneous Registration for Multiple 3D Images. In: Medical Image Computing and Computer Assisted Intervention – MICCAI 2022, pp. 98–107. Springer, Cham (2022). Zhang, L., et. Al. Simultaneous Registration of Location and Orientation in Intravascular Ultrasound Pullback Pairs via 3D Graph-Based Optimization. IEEE Transactions on Medical Imaging, 34(12), 2550–2561 (2015).

    The approach also depends on sparse electromagnetic (EM) tracking, which may not be available or reliable in all clinical environments. Finally, the absence of publicly available code and limited implementation details may hinder reproducibility and adoption by the community.

  • 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.

    (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?

    I recommend acceptance of this paper due to its strong technical innovation, clear clinical motivation, and well-executed methodology.

  • 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 would like to thank the reviewers for their constructive feedback on our paper. We are glad you acknowledged the “novelty” of our dynamic pose graph optimization for dynamic panoramic reconstruction of 4D ICE (R1, R2, R3) “addressing a critical need for real-time anatomical visualization during interventional procedures” (R3), “ the interesting decomposition of affine transformations to separate catheter motion from cardiac deformation” (R2), “the clever use of cardiac physiology” (R3), and the “scalability, efficiency, and real-time potential” of our implementation (R2, R3), along with the strong quantitative and qualitative result (R2, R3). Existing panoramic reconstruction methods depend on accurate EM tracking—often unreliable in clinical settings due to field disturbances—our method optimizes poses even with noisy EM data, improving dynamic panoramic visualization. We provide the following clarifications to improve the clarity of our contributions.

Regarding the experiment with metal objects (R1), we would like to clarify that we use a phantom with an electrically driven water pump, which introduces substantial noise with the EM Field generated by its electric motor. Conceptually, this is very similar to introducing a metal object into the environment. Table 1 shows the pump’s impact on pose error. By comparing the static and dynamic cases, smoothed EM tracking has a 43% higher Feature Error and 22.4% higher Global Error when the pump is switched on. The variance also increased by 16.3% and 33.5%, respectively. Our approach is significantly improving tracking accuracy in both cases. This is demonstrated in the results in Table 1 and in Figure 3, where the visualised reconstruction is more anatomically plausible.

Regarding the clinical application (R2, R3), cyclic motion assumption (R2, R3) and intra-phase graphs assignment (R2): In clinical use, ECG gating ensures robustness to heart rate variability and has been a common assumption in the literature [8, 11]. Irregular heart rhythms, common in structural heart procedures, can be managed by detecting anomalies in the ECG signal and excluding the corresponding images from the optimization. To avoid volume gaps or jumps, the operator is expected to reduce the catheter insertion speed accordingly.

Regarding insufficient technical details (R1), in the final version we will clarify: 3D Registration We registered ultrasound volumes pairwise, using a mask for non-image areas, random pixel sampling, and Conjugate Gradient Line Search optimization. Various similarity metrics were tested, but no significant differences were observed, leading to the selection of mean square similarity for its efficiency. As registration details had minimal impact on results, we focused this submission on graph optimization. EM Sensor mounting The EM sensor is integrated into the ICE catheter in the proximity of the US transducer with known relative placement from manufacturing.
Number of Phases Our approach allows any number of cardiac phases. We chose 5 to capture the full motion range of our phantom while limiting manual landmark annotation. More phases ease affine registration but increase computational cost. Intra-Phase Edges In summary, adjacent frames are connected directly, while loop closures are identified using nearest neighbors. We will update the figure to include a legend and color coding for these connections.

We thank all reviewers for their thoughtful and constructive feedback, which has improved the clarity and quality of our submission. We are especially encouraged by the positive reception, highlighting the clinical need for robust dynamic visualisation that our method is addressing. We believe it is important to share these insights at MICCAI25 to enable discussion and guide future research.



Meta-Review

Meta-review #1

  • Your recommendation

    Invite for Rebuttal

  • 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”.

    This manuscript received 3 polarizing reviews. Issues identified include the lack of technical details and limited validation, which will be difficult to justify due to the ‘no promised/new results’ policy in MICCAI. Another issue is that the proposed method will only work for cyclic/well-behaved cardiac motion, which limits its clinical utility. These issues need to be addressed in the rebuttal.

  • After you have reviewed the rebuttal and updated reviews, please provide your recommendation based on all reviews and the authors’ rebuttal.

    Accept

  • Please justify your recommendation. You may optionally write justifications for ‘accepts’, but are expected to write a justification for ‘rejects’

    N/A



Meta-review #2

  • After you have reviewed the rebuttal and updated reviews, please provide your recommendation based on all reviews and the authors’ rebuttal.

    Reject

  • Please justify your recommendation. You may optionally write justifications for ‘accepts’, but are expected to write a justification for ‘rejects’

    N/A



Meta-review #3

  • After you have reviewed the rebuttal and updated reviews, please provide your recommendation based on all reviews and the authors’ rebuttal.

    Accept

  • Please justify your recommendation. You may optionally write justifications for ‘accepts’, but are expected to write a justification for ‘rejects’

    N/A



back to top