Learning Graph Filters for Structure-Function Coupling based Hub Node Identification
Not on the map. Why?
Abstract (arXiv)
Over the past two decades, tools from network science have been leveraged to characterize the organization of both structural and functional networks of the brain. One such measure of network organization is hub node identification. Hubs are specialized nodes within a network that link distinct brain units corresponding to specialized functional processes. Conventional methods for identifying hub nodes utilize different types of centrality measures and participation coefficient to profile various aspects of nodal importance. These methods solely rely on the functional connectivity networks constructed from functional magnetic resonance imaging (fMRI), ignoring the structure-function coupling in the brain. In this paper, we introduce a graph signal processing (GSP) based hub detection framework that utilizes both the structural connectivity and the functional activation to identify hub nodes. The proposed framework models functional activity as graph signals on the structural connectivity. Hub nodes are then detected based on the premise that hub nodes are sparse, have higher level of activity compared to their neighbors, and the non-hub nodes' activity can be modeled as the output of a graph-based filter. Based on these assumptions, an optimization framework, GraFHub, is formulated to learn the coefficients of the optimal polynomial graph filter and detect the hub nodes. The proposed framework is evaluated on both simulated data and resting state fMRI (rs-fMRI) data from Human Connectome Project (HCP).
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 23 of its 102 references that are in the map. First posted on arXiv 22 Oct 2024.
Builds on
23 of 102 references.
References · 23
- 2022Rethinking Graph Neural Networks for Anomaly Detectioncited by 187
- 2021Interpreting and Unifying Graph Neural Networks with An Optimization Frameworkcited by 131
- 2015Global Convergence of ADMM in Nonconvex Nonsmooth Optimizationcited by 98
- 2020Graph signal processing for machine learning: A review and new perspectivescited by 73
- 2016Network Topology Inference from Spectral Templatescited by 58
- 2022Graph Filters for Signal Processing and Machine Learning on Graphscited by 53
- 2020A User Guide to Low-Pass Graph Signal Processing and its Applicationscited by 25
- 2015Sampling of graph signals with successive local aggregationscited by 19
- 2018Graph Learning from Filtered Signals: Graph System and Diffusion Kernel Identificationcited by 18
- 2016Small-World Brain Networks Revisitedcited by 16
- 2015MADMM: a generic algorithm for non-smooth optimization on manifoldscited by 12
- 2018Detection of false data injection attacks in smart grids based on graph signal processingcited by 12
11 more
- 2019Decoupling of brain function from structure reveals regional behavioral specialization in humanscited by 10
- 2021Bayesian Estimation of Graph Signalscited by 9
- 2017Filter Design for Autoregressive Moving Average Graph Filterscited by 8
- 2017Evaluating Graph Signal Processing for Neuroimaging Through Classification and Dimensionality Reductioncited by 7
- 2020Blind Inference of Eigenvector Centrality Rankingscited by 7
- 2021Graph-signal Reconstruction and Blind Deconvolution for Structured Inputscited by 5
- 2022Community-based anomaly detection using spectral graph filteringcited by 4
- 2016Locating Temporal Functional Dynamics of Visual Short-Term Memory Binding using Graph Modular Dirichlet Energycited by 2
- 2019Spectral Characterization of functional MRI data on voxel-resolution cortical graphscited by 2
- 2019Estimating Centrality Blindly from Low-pass Filtered Graph Signalscited by 2
- 2016Functional Alignment with Anatomical Networks is Associated with Cognitive Flexibilitycited by 1
Built on by
Citing papers per year: 2025: 1.
1 paper.