Abstract

Automated tooth arrangement is a crucial stage in digital orthodontic planning. Existing learning-based methods are based on large-scale expert-designed treatment plans, but high-quality arrangement results are difficult to obtain. Semi-supervised learning is commonly applied in scenarios with limited labeled data. However, due to the challenge of evaluating the confidence of pseudo-labels, previous works have not effectively explored semi-supervised tooth arrangement as a regression problem. To address this, we propose a semi-supervised tooth arrangement framework guided by dental arch priors and iterative confidence evaluation. We establish a teacher-student-based semi-supervised framework and introduce a weak-to-strong consistency regularization tailored for 3D point clouds. Inspired by optimization problems, we iteratively analyze errors to assess the confidence of pseudo-labels generated by the teacher network, mitigating the challenge of filtering low-quality pseudo-labels in regression. In addition, we predict the dental arch width to reduce the complexity of learning intricate transformations and leverage it as orthodontic prior information to improve arrangement accuracy. Our framework fills a critical gap in the field, and its core ideas can be generalized to other regression tasks. On a high-quality dataset, our method achieves competitive results with minimal labeled data. Code and typical data will be released soon.

Links to Paper and Supplementary Materials

Main Paper (Open Access Version): https://papers.miccai.org/miccai-2025/paper/4812_paper.pdf

SharedIt Link: Not yet available

SpringerLink (DOI): Not yet available

Supplementary Material: Not Submitted

Link to the Code Repository

https://github.com/oblivionis-tgw/ITMatch

Link to the Dataset(s)

N/A

BibTex

@InProceedings{WanChe_ITMatch_MICCAI2025,
        author = { Wang, Chengyuan and He, Zhihui and Chen, Li and Yang, Shidong and Sun, Guiyu and Duan, Fan and Wang, Shuo and Zhou, Yanheng},
        title = { { ITMatch: Arch-Guided Semi-Supervised Tooth Arrangement via Iterative Confidence Evaluation } },
        booktitle = {proceedings of Medical Image Computing and Computer Assisted Intervention -- MICCAI 2025},
        year = {2025},
        publisher = {Springer Nature Switzerland},
        volume = {LNCS 15967},
        month = {September},

}


Reviews

Review #1

  • Please describe the contribution of the paper

    This paper highlights the high labor intensity of manual annotation in the Automated Tooth Arrangement task and emphasizes the necessity of a semi-supervised learning approach to overcome this challenge. It is the first work to propose a semi-supervised learning framework for this task, integrating commonly used techniques such as consistency regularization and a teacher-student architecture to enable effective and natural semi-supervised training. Notably, it addresses a key issue in semi-supervised learning—the reliability of pseudo-labels—by introducing an Iterative Confidence Evaluation strategy, which enhances training stability. Furthermore, the model incorporates a commonly used dental arch prior—specifically, the dental arch width—into the encoding-decoding process, thereby guiding the model toward more accurate predictions.

  • 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 effectiveness of the proposed training framework is thoroughly validated through quantitative comparisons across multiple learning scenarios, including fully supervised and various semi-supervised settings. Additionally, the ablation studies are well-structured, using tables and figures to clearly demonstrate the contributions of individual components, thereby strengthening the authors’ claims.

    This work is the first to introduce a semi-supervised learning framework to the tooth arrangement task. The design effectively combines consistency regularization with a teacher-student model and introduces an intuitive confidence metric for pseudo-labels, resulting in a stable and coherent training pipeline.

  • 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.

    Achieving clinical-level performance using consistency regularization requires robust data augmentation that accurately reflects the complexity of real-world clinical data. However, the paper lacks detailed descriptions and justifications regarding the practicality of the augmentation strategies employed, making it difficult to assess the feasibility of deploying the method in clinical settings.

    The datasets used for both training and inference appear to be composed entirely of full-teeth cases. As such, the framework may struggle to provide reliable arch guidance in more clinically common scenarios, such as asymmetric dentition or missing teeth. This limitation reduces the applicability of the method to real-world clinical cases and may hinder its ability to deliver accurate predictions in those settings.

  • 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

    This study is the first to introduce a semi-supervised learning framework to the tooth arrangement task, combining consistency regularization and a teacher-student architecture to enable a natural and effective training process. The proposed approach is highly suitable for clinical settings where labeled data are limited. However, to clarify the scientific contribution of this work, the following limitation needs to be addressed. When leveraging consistency regularization, achieving clinical-level performance requires strong data augmentation strategies that adequately reflect the complexity of real clinical data. Therefore, a more detailed explanation of the augmentation techniques used, as well as a discussion of their practicality, is necessary.

  • 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?

    This paper meets the standards of a MICCAI-level contribution by being the first to propose a natural and effective semi-supervised learning pipeline for the tooth arrangement task. Furthermore, the experiments and ablation studies convincingly support the validity of the core techniques proposed in the paper.

  • 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

    This paper proposes the first semi-supervised regression framework for automated tooth arrangement.

  • 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. This work introduces the first semi-supervised regression framework specifically designed for automated tooth arrangement.
    2. The proposed Arch Perception and Guidance method is well adapted to the unique requirements of the tooth arrangement task.
    3. Thorough ablation studies are provided.
  • 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. The discussion of related work is too brief. A more detailed analysis of previous methods (references [18, 3, 15, 8]) would help better position the proposed approach and clarify its improvements.
    2. The claim in the abstract that the “core ideas can be generalized to other regression tasks” is not supported or discussed in the main text.
  • 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.

  • 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

    Reference [19] should be replaced.

  • 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?

    This paper makes a contribution to the research field of tooth arrangement. I hope the authors can address the issues I mentioned above in their revision.

  • Reviewer confidence

    Somewhat confident (2)

  • [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

    The main contribution of this paper is the introduction of ITMatch, a novel semi-supervised regression framework tailored for automated 3D tooth arrangement using point cloud data. Unlike prior works that rely heavily on fully labeled datasets, ITMatch effectively leverages unlabeled data by combining a teacher-student paradigm with a new iterative confidence evaluation (ICE) mechanism that filters unreliable pseudo-labels based on prediction stability across iterations. The framework also integrates domain-specific knowledge through an auxiliary task that predicts dental arch width, guiding the model to learn anatomically plausible tooth arrangements and simplifying complex transformations. Additionally, the authors design a weak-to-strong consistency regularization strategy suited for 3D inputs to improve the model’s robustness under limited supervision.

  • 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. Semi-Supervised Regression Framework for Tooth Arrangement 2.Iterative Confidence Evaluation (ICE)
    2. Task-Specific Consistency Regularization
  • 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.Limited Comparison to Other SSL Regression Methods 2.No Runtime or Efficiency Analysis 3.ICE Limited to Centroid-Based Confidence

  • 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.

  • 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 paper has has well-motivated formulation of semi-supervised regression for 3D tooth arrangement, the introduction of a principled and effective iterative confidence evaluation mechanism, and the integration of clinically meaningful priors through arch-guided learning.

  • Reviewer confidence

    Somewhat confident (2)

  • [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

N/A




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”.

    After careful consideration of this manuscript along with the expert reviews from the double-blind review process, I find the paper’s contributions to be potentially valuable to our field. While the reviewers have raised some valid points that need attention, I believe these can be adequately addressed. I recommend acceptance of this submission with the authors implementing the necessary revisions based on the reviewers’ feedback to strengthen the final version of the paper.



back to top