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

Abstract

Recent medical image segmentation methods have started to apply implicit neural representation (INR) to segmentation networks to learn continuous data representations. Though effective, they suffer from inferior performance. In this paper, we delve into the inferiority and discover that the underlying reason behind it is the indiscriminate treatment for context fusion that fails to properly exploit misaligned contexts. Therefore, we propose a novel Implicit-parameterized INR Network (I2Net), which dynamically generates the model parameters of INRs to adapt to different misaligned contexts. We further propose novel gate shaping and learner orthogonalization to induce I2Net to handle misaligned contexts in orthogonal ways. We conduct extensive experiments on two medical datasets, i.e. Glas and Synapse, and a generic dataset, i.e. Cityscapes, to show the superiority of our I2Net. Code: https://github.com/ChineseYjh/I2Net.

Links to Paper and Supplementary Materials

Main Paper (Open Access Version): https://papers.miccai.org/miccai-2024/paper/1184_paper.pdf

SharedIt Link: https://rdcu.be/dZxdI

SpringerLink (DOI): https://doi.org/10.1007/978-3-031-72111-3_31

Supplementary Material: https://papers.miccai.org/miccai-2024/supp/1184_supp.pdf

Link to the Code Repository

https://github.com/ChineseYjh/I2Net

Link to the Dataset(s)

N/A

BibTex

@InProceedings{Yu_I2Net_MICCAI2024,
        author = { Yu, Jiahao and Duan, Fan and Chen, Li},
        title = { { I2Net: Exploiting Misaligned Contexts Orthogonally with Implicit-Parameterized Implicit Functions for Medical Image Segmentation } },
        booktitle = {proceedings of Medical Image Computing and Computer Assisted Intervention -- MICCAI 2024},
        year = {2024},
        publisher = {Springer Nature Switzerland},
        volume = {LNCS 15008},
        month = {October},
        page = {328 -- 338}
}

Reviews

Review #1

  • Please describe the contribution of the paper

    This paper introduces a novel implicit-parameterized INR network to capture various patterns behind contexts for medical image segmentation. Further, novel gate shaping and learner orthogonalization are designed to learn orthogonal context patterns. Experiments on two medical image datasets and one natural image dataset demonstrate the effectiveness of the method.

  • 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.
    1. The proposed method, I^2Net, is designed to exploit misaligned contexts orthogonally for medical image segmentation.
    2. Experiments on two medical image datasets and one natural image dataset demonstrate the effectiveness of the method.
    3. An ablation study was conducted to verify the effectiveness of the key components.
  • 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.
    1. The author should explain why they do not report IOSNet’s results on the Glas dataset.
    2. The paper failed to compare with important baselines, such as nnFormer, MISSFormer, and Swin UNETR, on the Synapse dataset.
  • 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?

    No

  • 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 Section 6.

  • 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?

    Refer to Section 6.

  • Reviewer confidence

    Very confident (4)

  • [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

    Authors delve into the inferiority and discover that the underlying reason behind it is the indiscriminate treatment for context fusion that fails to properly exploit misaligned contexts. Authors propose a novel Implicit-parameterized INR Network (I2Net), which dynamically generates the model parameters of INRs to adapt to different misaligned contexts. Besides, they propose novel gate shaping and learner orthogonalization to induce I2Net to handle misaligned contexts in an orthogonal way.

  • 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.
    1. This paper aims to address the challenge of existing INRs for medical image segmentation by exploiting various discrimination patterns of contexts.
    2. The ideas of gating shaping and learner orthogonalization are very interesting, which might exert positive influences to the community.
    3. Experiments are conducted thoroughly to prove the efficacy of I2Net.
  • 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.
    1. For the task of medical image segmentation, please compare the proposed model with nnUNet, which is a very strong backbone. Also, there are a lack of experimental results for the recent INRs, including SwIPE, MORSE.
    2. Also, the pipeline can be easily expanded to the 3D architecture. Maybe a detailed evaluation on volumetric datasets will be more convincing.
  • 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?

    Authors have given enough experimental details for readers for an implementation.

  • 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 also see the main strengths and weaknesses part.

  • 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 aims to address the challenge of misaligned contexts faced by existing INRs. And the proposed contributions are very impressive. However, the experimental evaluations are not thorough enough.

  • 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

    The paper introduces the Implicit-parameterized Implicit Neural Representation Network (I^2Net) for medical image segmentation, improving performance by handling misaligned context latent codes. It uses a dynamic implicit function to adapt INR-based decoders to various discrimination patterns through pattern learners (PLs) and implicit gates, which selectively apply these patterns. The study also introduces techniques like gate shaping to prevent network degeneration and learner orthogonalization to ensure PLs learn distinct patterns. Key contributions include a novel approach to address indiscriminate context fusion in INR decoders, general applicability in INR-related fields, and validation of I2Net’s effectiveness across multiple datasets, including medical and semantic segmentation scenarios.

  • 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.
    1. The proposed method shows better performance over previous aligning methods.
  • 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.
    1. Lack of comparison with nnUnet, SwinUnet, and UNETR, which have better performance on the Synapse dataset.
    2. It would be better to report the inference time of implicit function based methods and traditional segmentation methods like Unet/TransUnet/nnUnet.
    3. It’s important to compare your methods with other network ensemble based methods. This comparison will provide a comprehensive evaluation and validate the effectiveness of your proposed approach.
  • 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.

  • 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

    Please refer the weakness.

  • 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?

    Please refer the weakness.

  • 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

N/A



Meta-Review

Meta-review not available, early accepted paper.



back to top