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

Abstract

Accurate automatic segmentation of the White Line of Toldt (WLT) is crucial for guiding colorectal cancer surgeries and improving patient outcomes. However, the complex anatomical structures and low signal-to-noise ratio involved in relevant regions of WLT pose significant challenges to existing segmentation models. Recent studies highlight fractal dimension as a powerful tool for analyzing the complexity of topological structures, offering an effective approach to representing anatomical features in medical images. Building on its success, we present the first well-annotated laparoscopic WLT segmentation (LTS) dataset and propose FSA-Net, a fractal-driven synergistic anatomy-aware network, specially designed for laparoscopic WLT segmentation. Specifically, FSA-Net consists of two core modules: the local texture-aware convolution (LTC) module and the fractal-guided anatomy-consistent attention (FAA) module. The LTC module adaptively adjusts the convolutional kernel offsets based on fractal dimensions to capture intra-anatomical features, while the FAA module employs a fractal-driven key-value pair filtering strategy to enhance the modeling of correlations across inter-anatomical structures. Extensive experimental results validate the effectiveness of our method. The resources will be available at https://github.com/Bigmouth233/FSA-Net.

Links to Paper and Supplementary Materials

Main Paper (Open Access Version): https://papers.miccai.org/miccai-2025/paper/2470_paper.pdf

SharedIt Link: Not yet available

SpringerLink (DOI): Not yet available

Supplementary Material: Not Submitted

Link to the Code Repository

https://github.com/Bigmouth233/FSA-Net

Link to the Dataset(s)

N/A

BibTex

@InProceedings{WuKec_FSANet_MICCAI2025,
        author = { Wu, Kecheng and Xing, Zhaohu and Cai, Zerong and Gao, Feng and Li, Wenxue and Zhu, Lei},
        title = { { FSA-Net: Fractal-driven Synergistic Anatomy-aware Network for Segmenting White Line of Toldt in Laparoscopic Images } },
        booktitle = {proceedings of Medical Image Computing and Computer Assisted Intervention -- MICCAI 2025},
        year = {2025},
        publisher = {Springer Nature Switzerland},
        volume = {LNCS 15968},
        month = {September},
        page = {288 -- 298}
}

Reviews

Review #1

  • Please describe the contribution of the paper

    The author create a dataset for the white line of toldt segmentation and proposed a novel network FSA-Net to achieve better performance on the white line of toldt and polpy segmention.

  • 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. The paper is well-organized, and the experiments are comprehensive.

    2. The authors collected and annotated a White Line of Toldt (WLT) segmentation dataset, filling a gap in this underexplored area.

  • 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 authors should elaborate further on the clinical importance of the White Line of Toldt (WLT), supported by relevant references.

    2. In the WLT segmentation results presented in Table 1, the performance appears suboptimal. This is understandable, as the WLT lacks a clearly defined boundary, as shown in the figures. It would be valuable to clarify whether WLT segmentation provides practical benefits in real-world surgical applications.

  • 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.

    (3) Weak Reject — could be rejected, dependent on rebuttal

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

    My main concern is whether WLT segmentation is truly necessary or clinically justified.

  • 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 #2

  • Please describe the contribution of the paper

    This paper proposes the FSA-Net, which is a fractal-driven synergistic anatomy-aware network used for segmenting white line of toldt in laparoscopic 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.

    Two modules are involved in the proposed network, including the local texture-aware convolution module and the fractal-guided anatomy-consistent attention module. Experimental results show the effectiveness of the proposed method.

  • 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. In the experiments, only one dataset is used for experiments. More datasets and cross-dataset evaluation would be helpful.
    2. There lack of comparisons with laparoscopic specific methods. All the other compared models are for general and colonoscopic image type, which seems unfair.
    3. The method “Boundary refinement network for colorectal polyp segmentation in colonoscopy images” could be mentioned in the paper.
  • 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.

    (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?
    1. In the experiments, only one dataset is used for experiments. More datasets and cross-dataset evaluation would be helpful.
    2. There lack of comparisons with laparoscopic specific methods. All the other compared models are for general and colonoscopic image type, which seems unfair.
    3. The method “Boundary refinement network for colorectal polyp segmentation in colonoscopy images” could be mentioned in the paper.
  • Reviewer confidence

    Confident but not absolutely certain (3)

  • [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.

    The authors have addressed my concerns.



Review #3

  • Please describe the contribution of the paper

    This paper introduces FSA-Net, a fractal-driven network for segmenting the White Line of Toldt (WLT) in laparoscopic surgery images. It uses fractal dimensions to model anatomical complexity, proposes two modules (LTC for local textures and FAA for cross-region context), and releases a new dataset (LTS). Results show it beats SOTA methods on WLT and polyp benchmarks.

  • 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. The paper propose a novel use of fractal analysis to handle tricky anatomical structures.
    2. The authors provide the first well-annotated WLT dataset, filling a gap for laparoscopic research.
    3. The solid experiments show that the proposed method outperforms others on both custom and public datasets, with clear ablation studies.
  • 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. Overly technical method sections (e.g., fractal math) without intuitive explanations—hard for non-math readers to grasp.
    2. Claims about being “first” for the dataset lack comparison to potential similar-but-unmentioned works in laparoscopic imaging.
    3. No analysis of failure cases (e.g., low-contrast WLT regions where fractal methods might struggle).
    4. Small LTS dataset (1.7k images from 145 patients) risks overfitting; no mention of patient diversity (age, BMI, etc.) or bias checks.
  • 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

    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?

    This work tackles a niche but critical problem in surgery with smart fractal integration and a new dataset. The results are convincing, and the modules (LTC/FAA) seem genuinely useful. But the lack of efficiency details and small dataset hold it back slightly. Still, the novelty and practical impact for medical imaging make it a solid accept—assuming the code/dataset drop as promised.

  • 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 sincerely appreciate the reviewers’ comments and address their concerns below.

Reviewer#1 Q1: More datasets evaluation. We appreciate the suggestion. As our LTS dataset is currently the only annotated dataset for White Line of Toldt segmentation, we present more evaluation results on relevant datasets: (ISIC2018: Dice 0.924, IoU 0.872 for our method (vs. 0.916/0.860 for Swin-UMamba; 0.911/0.849 for CASCADE; 0.875/0.787 for PraNet). ColonDB: Dice 0.837, IoU 0.759 for our method (vs. 0.818/0.729 for Swin-UMamba; 0.825/0.745 for CASCADE; 0.712/0.640 for PraNet)). Q2: Lack of laparoscopic specific methods. We compared our method to 3 laparoscopic specific methods on our LTS dataset: (LSKANet [1] Dice 0.531, IoU 0.399; TAFE [2] Dice 0.522, IoU 0.391; LDCNet [3] Dice 0.496, IoU 0.374). Our FSA-Net achieves better scores (Dice 0.557, IoU 0.428). Q3: Citation of the related work. We will include a proper citation of the recommended method in our manuscript.

Reviewer#2 Q1: Overly technical method without intuitive explanations. Thanks for your suggestion. Fig. 2 offers an illustration of the fractal analysis, and we will further refine the manuscript to aid understanding. Q2: Lack of comparison to similar works. To the best of our knowledge, we are the first to leverage fractal math in the context of laparoscopic White Line of Toldt segmentation. For more comparisons with related works in laparoscopic imaging, please refer to our response to Q2 of Reviewer #1. Q3: No analysis of failure cases. We acknowledge the omission of failure case analysis due to page limitations and will revise the manuscript to discuss failure cases. Q4: Small LTS dataset risks overfitting; No mention of patient diversity. We will add patient diversity information to our manuscript. Our dataset comprises patients with varied demographic characteristics (age, BMI, geographic background, etc.). We present 5-fold cross-validation results on LTS dataset: (Dice: 0.549, 0.562, 0.553, 0.545, 0.559→Mean±Std=0.554 ± 0.006; IoU: 0.420, 0.429, 0.426, 0.416, 0.427→Mean±Std=0.424 ± 0.005). The results suggest no evidence of overfitting.

Reviewer#3 Q1: Elaboration on the significance of White Line of Toldt (WLT). We appreciate the suggestion and will refine the manuscript with support from relevant references [4-5]. The WLT marks the junction of rectum and peritoneum, providing an avascular plane for safe mobilization of the mesorectum. Dissection along this plane is essential in procedures such as total mesorectum excision, enabling resection with minimal blood loss while preserving critical structures like the ureter and pelvic nerve. Q2: Clarify benefits in real-world surgical applications. We fully recognize the concern. We would like to clarify that WLT segmentation can provide effective visual guidance for accessing the avascular plane during surgery. We present a user study involving 3 experienced surgeons (5-10 years of surgical experience) and 3 trainees. Each participant rated the difficulty of identifying the WLT (scale 1-5) in 50 cases, with and without our predicted masks. Results show a reduction in difficulty scale with masks: Experienced surgeons: (without mask) 2.6 ± 0.81, 2.4 ± 0.85, 2.8 ± 0.65; (with mask) 2.3 ± 0.57, 2.0 ± 0.51, 2.1 ± 0.64. Trainees: (without mask) 3.8 ± 0.74, 4.1 ± 0.72, 3.7 ± 0.88; (with mask) 2.7 ± 0.62, 3.1 ± 0.49, 3.0 ± 0.56. The results suggest that our segmentation facilitates WLT identification, particularly for junior surgeons, thereby supporting access to the correct plane.

References [1]LSKANet: Long strip kernel attention network for robotic surgical scene segmentation. (2023) [2]Surgical Scene Segmentation by Transformer with Asymmetric Feature Enhancement. (2025) [3]LDCNet: Lightweight dynamic convolution network for laparoscopic procedures image segmentation. (2024) [4]An optimal surgical plane for laparoscopic functional total mesorectal excision in rectal cancer. (2021) [5]The ‘Holy Plane’ of rectal surgery. (1988)



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”.

    N/A

  • 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.

    Accept

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

    The initial points of weakness raised by R3 are about clinical and real world relevance. I think the response from authors is reasonable and this is the kind of thing that can be easily added. So despite lack of post-rebuttal interaction from reviewers, I recommend Accept.



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