Multiscale Community-Based Fingerprinting of Signed Functional Networks
Authors: Sema Athamnah, Selin Aviyente
Organizations: Department of Biomedical Engineering, Michigan State University, USA. · Department of Electrical and Computer Engineering, Michigan State University, USA.
Objective: Recent studies demonstrate that functional connectomes contain subject-specific signatures, or fingerprints, that can identify individuals across repeated sessions and tasks. Existing methods mostly rely on edge-level features that are sensitive to noise, difficult to interpret, and limited in their ability to generalize across tasks and datasets. Methods: We propose a multiscale community-based functional connectome fingerprinting framework that characterizes each individual by the mesoscale structure of their functional networks. We introduce a signed multilayer community detection framework that incorporates both correlated and anti-correlated brain activity to identify subject-specific community structures across tasks and sessions. Graph-theoretic metrics are then computed from the resulting joint community structures to derive low-dimensional community-level fingerprint representations. Results: The proposed framework is evaluated on 810 healthy control subjects from the Human Connectome Project (HCP). The results show that community-based fingerprints provide a reliable and interpretable substrate for individualized brain characterization across sessions and tasks. Conclusion: Mesoscale community structure provides meaningful and discriminative subject-specific fingerprints. Significance: The proposed framework offers a promising foundation for precision neuroimaging and personalized neuroscience applications.
Figures & tables
Notations
Explanations
G±
Signed graph.
N
Number of nodes.
V
Set of nodes.
E
Set of edges.
A±
Positive (negative) adjacency matrix.
P±
Positive (negative) null-model adjacency matrix.
Table 1: Notation definitions.
Figure 1: Overview of the proposed community-based fingerprinting framework. (a) Multiscale Community Detection: A signed multilayer modularity (SMM) optimization framework is applied to generate a set of multiscale partitions by varying the resolution ( γ ) and interlayer coupling ( ω ) parameters (Section 3.1 ). (b) Fingerprint Extraction: For each partition, community-level graph-theoretic features are extracted from each subject’s functional connectivity (FC) network to construct community-based fingerprint (CBF) vectors (Section 3.2 ). (c) Partition Selection: The resulting partitions are ranked according to the mean pairwise Euclidean distance (MPED) of their corresponding CBF vectors, and the top- L partitions are selected (Section 3.3 ). (d) Subject Identification: The selected top- L partitions are used to construct the training and testing CBFs from the corresponding training and testing FC networks. For each selected partition and feature type, each testing CBF vector is matched to its corresponding training CBF vector using a distance-based nearest-neighbor classifier for subject identification (Section 3.4 ).
Figure 2: Within-task fingerprinting results for a sample size of S=100 , averaged across 10 randomly selected subject groups.
Figure 3: Within-task fingerprinting results across different methods for sample sizes (a) S=200 , (b) 300 , and (c) 400 .
Figure 4: Between-task fingerprinting results across different methods for a sample size of S=400 . Results for S=200 , and 300 are provided in the Supplementary Materials.
Figure 5: Discriminative connectivity maps illustrating the spatial distribution of the top-1% discriminative community-based fingerprints (first and third rows), with the corresponding contributions of the seven Yeo functional brain systems (second and fourth rows) for each HCP task ( S=400 ).
Effect
DF
F-value
p -value
p -value (GG)
fMRI task
7,63
1181.60
2.79×10−64∗
1.08×10−31∗
Method
7,63
5834.68
4.89×10−86∗
6.48×10−28∗
fMRI task × Method
49,441
180.61
1.47×10−260∗
1.74×10−30∗
Table 2: Results of the two-way repeated-measures ANOVA for sample size S =100.
Table 3: Bonferroni-corrected post-hoc comparisons for the method factor, for sample size S =100.
Effect
DF
F-value
p -value
fMRI task
7,98
2531.54
1.03×10−107
Sample Size
2,98
373.31
1.46×10−46
Method
7,98
2269.96
2.09×10−105
Method × Sample Size
14,98
13.77
1.90×10−17
Method × fMRI task
49,98
77.45
1.68×10−60
Sample Size × fMRI task
14,98
4.13
1.27×10−5
Table 4: Results of the three-way ANOVA for sample size S =200, 300, and 400.
Table 5: Bonferroni-corrected post-hoc comparisons for the method factor for sample size S =200, 300, and 400.
Integrating structural and functional connectomes remains challenging because their relationship is non-linear and organized over nested modular hierarchies. We propose a hierarchical multiscale structure-function coupling framework for connectome integration that jointly learns individualized modular organization and hierarchical coupling across structural connectivity (SC) and functional connectivity (FC). The framework includes: (i) Prototype-based Modular Pooling (PMPool), which learns modality-specific multiscale communities by selecting prototypical ROIs and optimizing a differentiable modularity-inspired objective; (ii) an Attention-based Hierarchical Coupling Module (AHCM) that models both within-hierarchy and cross-hierarchy SC-FC interactions to produce enriched hierarchical coupling representations; and (iii) a Coupling-guided Clustering loss (CgC-Loss) that regularizes SC and FC community assignments with coupling signals, allowing cross-modal interactions to shape community alignment across hierarchies. We evaluate the model's performance across four cohorts for predicting brain age, cognitive score, and disease classification. Our model consistently outperforms baselines and other state-of-the-art approaches across three tasks. Ablation and sensitivity analyses verify the contributions of key components. Finally, the visualizations of learned coupling reveal interpretable differences, suggesting that the framework captures biologically meaningful structure-function relationships.
Jianwei Chen, Zhengyang Miao, Wenjie Cai +10
School of Medicine, University of Dundee, UK · College of Medicine and Biological Information Engineering, Northeastern University, China · School of Airspace Science and Engineering, Shandong University, China +5
Brain networks exhibit a modular community structure that varies across individuals and neurological conditions. However, existing self-supervised learning (SSL) methods often overlook this heterogeneity, relying on generic masking strategies that fail to capture subject-specific functional organization. We propose BrainPICM, a self-supervised framework for brain network analysis via progressive individualized community aware masking. BrainPICM formulates ROI-to-community mapping as a progressive unbalanced optimal transport process, yielding soft assignments and per-ROI confidence scores. Guided by these confidence estimates, a curriculum-style masking strategy gradually incorporates low-confidence, potentially pathological regions into training, enabling the model to learn both stable modular structures and individual variations. Additionally, a deviation-aware aggregation module quantifies functional reorganization by measuring mass redistribution relative to a population template, enhancing interpretability and downstream prediction. Experiments on three fMRI datasets (ABIDE-I, ADHD-200, ADNI) show that BrainPICM consistently outperforms state-of-the-art supervised and SSL methods in diagnostic accuracy, indicating that explicitly injecting modular community structure into masked modeling yields more functionally consistent and generalizable representations. The source code for this approach will be released at https://github.com/Hrychen7/BrainPICM.
Hairui Chen, Yanwu Yang, Jianfeng Cao +3
Harbin Institute of Technology at Shenzhen, Shenzhen, China · Peng Cheng Laboratory, Shenzhen, China · University of Tübingen, Tübingen, Germany +1
Independently trained neural networks have no shared neuron-index reference frame, so comparing them requires accounting for coordinate freedom. Neural Collapse sharpens this problem: networks converge toward a shared, low-dimensional geometry, raising the question of whether trajectory-specific functional variation remains distinguishable after convergence. We distinguish three claims - detectability, transplantability, and causal persistence - and address the first. Using five independently trained networks reconstructing Neural Collapse on MNIST, we apply a verified affine-correct alignment mapping donor heads into recipient coordinates. Donor-specific functional fingerprints remain distinguishable after recipient-level baseline correction: all 20 ordered donor-recipient pairs are correctly identified, with an exact permutation p=0.0083, robust to a leakage audit. These findings establish detectability under the test used here, but not transplantability or causal persistence. The study shows how alignment, ambiguity diagnostics, and leakage control combine to test cross-network variation in a controlled setting; whether this generalizes beyond it is open.
Truong Xuan Khanh, Phan Thanh Duc
H&K Research Studio, Clevix LLC, Hanoi, Vietnam · Banking Academy of Vietnam, Hanoi, Vietnam