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

Abstract

Brain functional network (FN) extraction is fundamental to advancing our understanding of brain function, providing critical insights into the neural mechanisms underlying cognition and behavior. Data-driven FN analysis methods have been developed to analyze functional magnetic resonance imaging (fMRI) data. However, to ensure cross-subject correspondence, group-level analyses of these methods sacrifice subject-specific variation. This trade-off between group-level alignment and subject-specific discrepancies hinders the accurate characterization of individual brain FNs. In this study, we propose a multi-subject orthogonal sparse matrix decomposition method without the need for group-level analysis, which simultaneously extracts both group-level FNs and individual FNs with cross-subject correspondence. We introduce a novel quasi-orthogonality constraint that enhances the linear independence of FNs, ensuring effective extraction of FNs, while enabling precise control over FN spatial scale. Additionally, by further incorporating a sparsity constraint, our method effectively minimizes spatial overlap between FNs, resulting in sparse representations. For simulated datasets, our method outperforms comparison methods, supporting its low parameter sensitivity and superior ability to extract FNs and time courses. Application to multi-site fMRI datasets, comprising 233 healthy controls (HCs) and 205 schizophrenia patients (SZs), validates the reproducibility of FNs extracted by our method. The results underscore the method’s ability to preserve both cross-subject correspondence and individual variability. Overall, our method advances fMRI analytic capabilities by reconciling population-level consistency with individualized neural signatures, offering enhanced discriminative power for investigating neuropsychiatric disorder mechanisms and brain function.

Links to Paper and Supplementary Materials

Main Paper (Open Access Version): https://papers.miccai.org/miccai-2025/paper/4264_paper.pdf

SharedIt Link: Not yet available

SpringerLink (DOI): Not yet available

Supplementary Material: Not Submitted

Link to the Code Repository

N/A

Link to the Dataset(s)

N/A

BibTex

@InProceedings{HeXin_Multisubject_MICCAI2025,
        author = { He, Xingyu and Calhoun, Vince D. and van Erp, Theo G.M. and Du, Yuhui},
        title = { { Multi-subject Orthogonal Sparse Matrix Decomposition Method for Extracting Individual Brain Functional Networks } },
        booktitle = {proceedings of Medical Image Computing and Computer Assisted Intervention -- MICCAI 2025},
        year = {2025},
        publisher = {Springer Nature Switzerland},
        volume = {LNCS 15966},
        month = {September},

}

Reviews

Review #1

  • Please describe the contribution of the paper

    This paper proposes a multi-subject orthogonal sparse matrix decomposition method that introduces a quasi-orthogonality constraint and a sparsity constraint to simultaneously extract group-level functional networks (FNs) with cross-subject consistency and individualized FNs without requiring group-level analysis. The method’s effectiveness is validated on both simulated and multi-site real fMRI data, revealing potential biomarkers associated with the cerebellum and superior temporal gyrus in schizophrenia patients. This work provides a novel tool for balancing population-level consistency and individual variability in functional network analysis.

  • 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. Without requiring group-level analysis or complex initialization procedures, our proposed method enables concurrent identification of robust group-level FNs and individual FNs with cross-subject comparability.
    2. We propose a novel quasi-orthogonality constraint that ensures the effective extraction of FNs by enhancing their linear independence while allowing for precise modulation of the FN scales.
    3. By further incorporating a sparsity constraint, the proposed method effectively reduces the spatial overlap between FNs, yielding sparse representations.
  • 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) Why did the authors choose to compare their method only with the third paradigm, which directly enforces cross-subject correspondence during individual FN analysis (e.g., Independent Vector Analysis)? It would be more convincing to also include comparisons with the first and second paradigms (2) Lack of code or data preprocessing details hinders reproducibility. 3）The manuscript should follow a standardized formatting template, as the current structure appears somewhat disorganized. 4）In Figures 1 and 2, the spacing between subfigures is too narrow, making the layout appear crowded. Please revise these figures to improve visual clarity.

  • 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.

    (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?

    The core value lies in resolving the individual-group trade-off in FN extraction via quasi-orthogonality and sparsity constraints, supported by rigorous experiments and clear clinical relevance (e.g., schizophrenia biomarkers). Deductions stem from limited comparisons and reproducibility gaps.

  • 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

    Summary: The paper designs a multi-subject orthogonal sparse matrix decomposition functional brain network analysis, which jointly extracts group-level and individual brain networks. The key novelty is in the proposal of a quasi-orthogonality constraint that encourages linear independence among networks, and controls the spatial scale. Spatial overlap is also influenced by the introduction of a sparsity regulariser. Experiments are performed on simulated and real world fMRI data from healthy and schizophrenic patients.

  • 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 methodology is motivated well and explained clearly. The suitability to this application comes across pretty well. The algorithm design is clearly presented along with a theoretical exposition on the optimization strategy

    2. Multiple simulations have been run to test the assumptions in the data and two separate real-world datasets have been used for experimental validation.

    3. Experimental results support the main claims being made by the authors.

  • 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. Comparisons are restricted to matrix decomposition frameworks designed for time-series data e.g. NMF, IVA. It would be interesting to see comparisons against brain network extraction from covariance decomposition models such as Common Principal Components.

    2. Open sourcing of the code to reproduce experiments is glossed over.

  • 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

    1. The authors should improve the formatting and presentation of equations, references, and figures in the paper and use inline references to equations/sections when appropriate. These changes would help improve the accessibility of the papers to those without the strong theoretical background in fMRI and representation learning.

    2. The authors should consider including a small explanation on how to interpret the p values computed by the t test in Table 2.

  • 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?

    Overall the contributions made by the work, both in terms of methodology and experimental evaluation seem sufficient for acceptance into the conference.

  • 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 #3

  • Please describe the contribution of the paper

    The paper proposes orthogonal sparse matrix decomposition to obtain individual and group functional connectivity networks simultaneously. The method is first tested on simulated data and finally applied to two existing datasets corresponding to Schizophrenia studies. The method is shown to extract individual and group networks. Two specific group-level networks differed between controls and patients, which the authors interpret as potential biomarker networks.

  • 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 method is thoroughly described
    • The method performed better than existing factorization alternatives (NMF, IVA) on simulated data
    • The method could potentially be more specific in the estimation of individual networks without losing sensitivity to group networks than classic alternatives such as PCA.
  • 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.
    • The conclusions about potential biomarkers are highly speculative and should be toned down. The gap from the results in this paper and an actual biomarker definition is wide.
    • The text has some minor oversights that create friction
    • Figure 1, panel A should use a meshes rather than bars, otherwise 3D plots are challenging to interpret.
    • Figure 2 condenses too much information and it’s not immediate, e.g., what is the role of panel D (a “FN template derived from COBRE”) because it hasn’t been introduced in the text. I could not figure out what E wants to show and why FN 22 is the only one not showing any correlation and is not highlighted or described anywhere.
  • 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 provide sufficient information for 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?

    The paper would require proofreading and improvements with a better description of the results, but it’s overall a sound contribution.

  • 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



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”.

    N/A



back to top