Abstract:Mounting evidence has demonstrated that complex brain function processes are realized by the interaction of holistic functional brain networks which are spatially distributed across specific brain regions in a temporally dynamic fashion. Therefore, modeling spatio-temporal patterns of holistic functional brain networks plays an important role in understanding brain function. Compared to traditional modeling methods such as principal component analysis, independent component analysis, and sparse coding, superior performance has been achieved by recent deep learning methodologies. However, there are still two limitations of existing deep learning approaches for functional brain network modeling. They either (1) merely modeled a single targeted network and ignored holistic ones at one time, or (2) underutilized both spatial and temporal features of fMRI during network modeling, and the spatial/temporal accuracy was thus not warranted. To address these limitations, we proposed a novel Multi-Head Guided Attention Graph Neural Network (Multi-Head GAGNN) to simultaneously model both spatial and temporal patterns of holistic functional brain networks. Specifically, a spatial Multi-Head Attention Graph U-Net was first adopted to model the spatial patterns of multiple brain networks, and a temporal Multi-Head Guided Attention Network was then introduced to model the corresponding temporal patterns under the guidance of modeled spatial patterns. Based on seven task fMRI datasets from the public Human Connectome Project and resting state fMRI datasets from the public Autism Brain Imaging Data Exchange I of 1448 subjects, the proposed Multi-Head GAGNN showed superior ability and generalizability in modeling both spatial and temporal patterns of holistic functional brain networks in individual brains compared to other state-of-the-art (SOTA) models. Furthermore, the modeled spatio-temporal patterns of functional brain networks via the proposed Multi-Head GAGNN can better predict the individual cognitive behavioral measures compared to the other SOTA models. This study provided a novel and powerful tool for brain function modeling as well as for understanding the brain-cognitive behavior associations.

A Novel Spatio-temporal Hub Identification in Brain Networks by Learning Dynamic Graph Embedding on Grassmannian Manifolds

A NOVEL SPATIO-TEMPORAL HUB IDENTIFICATION METHOD FOR DYNAMIC FUNCTIONAL NETWORKS

Spatio-temporal Modeling of Connectome-Scale Brain Network Interactions Via Time-Evolving Graphs

Joint hub identification for brain networks by multivariate graph inference

Joint Identification of Network Hub Nodes by Multivariate Graph Inference

Spatio-Temporal Graph Hubness Propagation Model for Dynamic Brain Network Classification

Learning Graph Filters for Structure-Function Coupling based Hub Node Identification

Learning Brain Dynamics of Evolving Manifold Functional MRI Data Using Geometric-Attention Neural Network.

A Deep Probabilistic Spatiotemporal Framework for Dynamic Graph Representation Learning with Application to Brain Disorder Identification

Deep Representations for Time-varying Brain Datasets

Manifold Learning in Detecting the Transitions of Dynamic Functional Connectivities Boosts Brain State-Specific Recognition.

Multi-view Graph Embedding with Hub Detection for Brain Network Analysis

Discrete-time Temporal Network Embedding via Implicit Hierarchical Learning in Hyperbolic Space

Capturing Spatial Dynamics Using Time-Resolved Referenced-Informed Network Estimation Techniques.

Modelling variability in dynamic functional brain networks using embeddings

DBGDGM: Dynamic Brain Graph Deep Generative Model

Modeling spatio-temporal patterns of holistic functional brain networks via multi-head guided attention graph neural networks (Multi-Head GAGNNs)

DGST-Former: Dynamic Graph with Spatio-Temporal Transpose-Transformer for Brain fMRI Data

Dynamic Graph Transformer for Brain Disorder Diagnosis

A Novel Method to Identify Mild Cognitive Impairment Using Dynamic Spatio-Temporal Graph Neural Network

A Deep Graph Neural Network Architecture for Modelling Spatio-temporal Dynamics in resting-state functional MRI Data