List of Papers Browse by Subject Areas Author List
Abstract
Whole brain segmentation, which divides the entire brain volume into anatomically labeled regions of interest (ROIs), is a crucial step in brain image analysis. Traditional methods often rely on intricate pipelines that, while accurate, are time-consuming and require expertise due to their complexity. Alternatively, end-to-end deep learning methods offer rapid whole brain segmentation but often sacrifice accuracy due to neglect of geometric features. In this paper, we propose a novel framework that integrates the key curvature feature, previously utilized by complex surface-based pipelines but overlooked by volume-based methods, into deep neural networks, thereby achieving both high accuracy and efficiency. Specifically, we first train a coarse anatomical segmentation model focusing on high-contrast tissue types, i.e., white matter (WM), gray matter (GM), and subcortical regions. Next, we reconstruct the cortical surfaces using the WM/GM interface and compute curvature features for each vertex on the surfaces. These curvature features are then mapped back to the image space, where they are combined with intensity features to train a finer cortical parcellation model. We also simplify the process of cortical surface reconstruction and curvature computation, thereby enhancing the overall efficiency of the framework. Additionally, our framework is flexible and can incorporate any neural network as its backbone. It can serve as a plug-and-play component to enhance the whole brain segmentation results of any segmentation network. Experimental results on the public Mindboggle-101 dataset demonstrate improved segmentation performance with comparable speed compared to various deep learning methods.
Links to Paper and Supplementary Materials
Main Paper (Open Access Version): https://papers.miccai.org/miccai-2024/paper/2514_paper.pdf
SharedIt Link: https://rdcu.be/dV5J6
SpringerLink (DOI): https://doi.org/10.1007/978-3-031-72114-4_2
Supplementary Material: N/A
Link to the Code Repository
N/A
Link to the Dataset(s)
N/A
BibTex
@InProceedings{Zha_ACurvatureGuided_MICCAI2024,
author = { Zhao, Fenqiang and Tang, Yuxing and Lu, Le and Zhang, Ling},
title = { { A Curvature-Guided Coarse-to-Fine Framework for Enhanced Whole Brain Segmentation } },
booktitle = {proceedings of Medical Image Computing and Computer Assisted Intervention -- MICCAI 2024},
year = {2024},
publisher = {Springer Nature Switzerland},
volume = {LNCS 15009},
month = {October},
page = {13 -- 22}
}
Reviews
Review #1
- Please describe the contribution of the paper
This paper proposed a curvature-guided coarse-to-fine framework to improve the whole brain segmentation performance.
- 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.
- This paper is well written and easy to follow.
- The proposed method is direct.
- After the guidance of the proposed curvature calculated from a simple marching cube method, the segmentation accuracy is largely improved based on popular segmentation 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.
- Brain segmentation methods are not concluded well in the section Introduction, some studies such as https://doi.org/10.1007/s00138-023-01415-0 will enrich this part.
- In section 2.2, why did the authors consider the topology correction, spherical mapping, and surface registration as non-essential steps? The authors just mentioned these steps can be replaced by using deep learning-based models. Did the authors get the experimental results based on these steps to prove this point?
- From Table 2, experimental results show that the proposed method is not comparable to the FreeSurfer. It may mean that the segmentations of Mindboggle101 datasets are fine-tuned based on FreeSurfer results, and after introducing the curvature features the deep learning model tries to learn the segmentation procedure like FreeSurfer. More convincing experiments using some other human-labeled dataset should be conducted to demonstrate the effectiveness of the proposed method.
- 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
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
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 proposed method is easy to reimplemeneted and with better performance, but some description about the surface reconstruction are not suitable.
- 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
This work presents an automatic brain region parcellation framework guided by curvature features in a coarse-to-fine manner. The dataset consists of 101 brain MRI data with manually labeled brain regions from the Mindboggle-101 dataset. The results demonstrate that the proposed coarse-to-fine segmentation framework and curvature-guided tissue segmentation can enhance segmentation performance.
- 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) Employ a coarse-to-fine strategy for brain region parcellation, prioritizing segmentation of higher-contrast targets before detailing more intricate brain regions. 2) Utilize curvature information to guide cortical region parcellation, a crucial consideration for manual annotation, significantly enhancing segmentation performance.
- 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) Cortical curvature calculation heavily relies on well-reconstructed surface meshes, with topology correction being crucial for accurate surface reconstruction. Dismissing the importance of the topology correction step would be unwarranted, as incorrect curvature features can compromise segmentation performance. 2) Despite the advancements in deep-learning methods expediting processes, the topology-correction step outlined in [20] may still require several minutes. However, if authors integrate the topology correction using methods reported in [20], the entire inference time could be reduced to less than 10 seconds. 3) The concept of learning-based brain region parcellation has been previously proposed and investigated. It’s imperative for authors to acknowledge prior works and conduct comparative experiments, as seen in [1, 2]. [1] Billot, Benjamin, et al. “Robust machine learning segmentation for large-scale analysis of heterogeneous clinical brain MRI datasets.” Proceedings of the National Academy of Sciences 120.9 (2023): e2216399120. [2] Tustison, Nicholas J., et al. “The ANTsX ecosystem for quantitative biological and medical imaging.” Scientific reports 11.1 (2021): 9068.
- 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.
- Do you have any additional comments regarding the paper’s reproducibility?
The reproducibility of this paper is good except for the link of the code that is not provided.
- 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
1) Lack of techonical details: The procedure for calculating curvature from the surface is not explicitly outlined, particularly the method for converting curvature from 2D to 3D space is not specified. 2) Lack of comparison experiments: It will be more helpful if the authors could compare with more strong learning-based segmentation baselines such as SyntheSeg++, antpynet. 3) The reconstructed curvature should be displayed in the final version, and a comparison using accurate curvature should be included.
- 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?
I appreciate the complete ablation studies and the significant improvement from several baseline in this work. However, the comparison with strong segmentation baseline is insufficient. I wish authors could include more comparison methods. I would improve the rate if the author can reply my concern in rebuttal session.
- 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 presents a framework for improving traditional brain segmentation using curvature features from brain atlas. They provide their method as a plug and play to improve segmentation using any trained neural network as background.
- 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.
The paper is well-written. The motivation behind the experiment setup is clear. The results and discussion help the readers understand the effect of curvature features on improvement of segmentation results.
- 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.
One weekness would be the amount of improvment in the Dice Score reported in this paper. On average there is around 1 or 2% improvement in the segmentation. Would that be considered significant enough to affect any diagnosis or prediction based on segmneted images.
- 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 submission does not provide sufficient information for 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
Overall, paper is well-written and easy to follow. The authors should discuss the significance of the improvement in their segmentation results. But, overall this is a clean method to improve segmentation results as location-aware and curvature features are included in the segmentation pipeline.
- 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
Accept — should be accepted, independent of rebuttal (5)
- Please justify your recommendation. What were the major factors that led you to your overall score for this paper?
The exhastive set up of experiments with latest UNet and nnUnet help the readers to establish the conclusion of how curvature features would help in segmentation. The design of the experiment is clean and the paper is well written.
- 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
Author Feedback
N/A
Meta-Review
Meta-review not available, early accepted paper.