Abstract

This study, we introduce a novel Topological Cycle Graph Attention Network (CycGAT), designed to delineate a functional backbone within brain functional graphs—key pathways essential for signal transmission—from non-essential, redundant connections that form cycles around this core structure. We first introduce a cycle incidence matrix that establishes an independent cycle basis within a graph, mapping its relationship with edges. We propose a cycle graph convolution that leverages a cycle adjacency matrix, derived from the cycle incidence matrix, to specifically filter edge signals in a domain of cycles. Additionally, we strengthen the representation power of the cycle graph convolution by adding an attention mechanism, which is further augmented by the introduction of edge positional encodings in cycles, to enhance the topological awareness of CycGAT. We demonstrate CycGAT’s localization through simulation and its efficacy on an ABCD study’s fMRI data (n=8765), comparing it with baseline models. CycGAT outperforms these models, identifying a functional backbone with significantly fewer cycles, crucial for understanding neural circuits related to general intelligence. Our code will be released once accepted.

Links to Paper and Supplementary Materials

Main Paper (Open Access Version): https://papers.miccai.org/miccai-2024/paper/1737_paper.pdf

SharedIt Link: https://rdcu.be/dV59l

SpringerLink (DOI): https://doi.org/10.1007/978-3-031-72120-5_67

Supplementary Material: https://papers.miccai.org/miccai-2024/supp/1737_supp.pdf

Link to the Code Repository

https://github.com/JH-415/CycGAT

Link to the Dataset(s)

https://abcdstudy.org/

BibTex

@InProceedings{Hua_TopologicalCycle_MICCAI2024,
        author = { Huang, Jinghan and Chen, Nanguang and Qiu, Anqi},
        title = { { Topological Cycle Graph Attention Network for Brain Functional Connectivity } },
        booktitle = {proceedings of Medical Image Computing and Computer Assisted Intervention -- MICCAI 2024},
        year = {2024},
        publisher = {Springer Nature Switzerland},
        volume = {LNCS 15011},
        month = {October},
        page = {723 -- 732}
}

Reviews

Review #1

Please describe the contribution of the paper

The paper introduces a novel network architecture (CycGAT) designed to differentiate essential signal transmission pathways in brain functional graphs from redundant connections. It utilizes a cycle incidence matrix and a cycle adjacency matrix to enhance the representation and processing of cycle-related information in the graph. The model incorporates edge positional encodings to further strengthen its topological awareness, demonstrating excellent performance on a large fMRI dataset from the ABCD study
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. CycGAT introduces an innovative use of cycle-based graph analysis which is novel in the field of brain functional connectivity. It enhances the ability to identify and distinguish between essential and redundant connections within brain functional graphs；
2. The paper reports that CycGAT outperforms existing baseline models on the ABCD dataset, indicating its efficacy in handling large-scale, real-world data；
3. The authors provide a comprehensive description of their methods, including the mathematical foundations of the cycle incidence and adjacency matrices, which are crucial for understanding and replicating the study.
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 complexity of the methodology, particularly the mathematical descriptions, might be challenging for readers not familiar with advanced graph theory and neural network architectures.
2. minor revision: The paper primarily focuses on one dataset. Additional testing across various datasets would strengthen the generalizability of the findings.
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.
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

1) It is recommended that the authors extend the CycGAT on other fMRI datasets or even other types of graph-structured data to demonstrate the model’s versatility and robustness. 2) Please ensure that the released code is accessible and well-documented, which would aid in fostering reproducibility and further innovation in the field.
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 method of the paper lacks novelty and can help make advancement in the analysis of brain functional connectivity.
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 study introduced a novel Topological Cycle Graph Attention Network (CycGAT), designed to delineate a functional back- bone within brain functional graphs from non-essential, redundant connections that form cycles around this core structure.
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 main idea and the challenge that is aimed to be investigated are interesting. 2- The method is properly employed in the real-world data.
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.

In the results section, Table 1 presents the general intelligence classification accuracy. The p-value is obtained from two-sample t-tests examining the performance of each method in reference to the proposed CycGAT. While it shows that the results for CycGAT appear to be better, the improvement is not significant.
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?

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

The paper is well written and the basic ideas and its execution is correct. I only have one concern, if the improvement with the proposed is the one provided in table 1, it is not a significant improvement. The authors should present other aspects of their provided method.
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 is well written and the basic ideas and execution are correct. But the improvement on the real world data reported in the manuscript is not significant.
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 proposes a cycle graph convolution that leverages a cycle adjacency matrix to specifically filter edge signals in a domain of cycles. Additionally it strengthen the representation power of the cycle graph convolution by adding an attention mechanism, which sounds novel.
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.

They inspired from graph convolution network to classify general intelligence groups by learning from the edges’ features and their neighborhood connections in a domain of cycles and from attention mechanism to smooth edge signals, which sounds novel.
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. Why the proposed method works better than the compared ones are not analyzed in detail.
2. More experimental results are expected, Ablation study is also encouraged to be conducted for the work.
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?

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 see comments in the weakness section.
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?

Novelty is good enough but results and more experiments can be further improved.
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

Dear Reviewers, Thank you for your valuable comments. We will ensure that the code is clearly organized and accessible on GitHub, with the link provided in the camera-ready version. Additionally, we will consider your advice when we extend this conference paper to a journal submission. This includes adding more experiments on different imaging modalities and other brain datasets. Thank you again for your insights.

Meta-Review

Meta-review not available, early accepted paper.

back to top

Topological Cycle Graph Attention Network for Brain Functional Connectivity

Author(s):