Functional Connectivity Graph Neural Networks
By: Yang Li, Luopeiwen Yi, Tananun Songdechakraiwut
Potential Business Impact:
Helps computers understand complex patterns in networks.
Real-world networks often benefit from capturing both local and global interactions. Inspired by multi-modal analysis in brain imaging, where structural and functional connectivity offer complementary views of network organization, we propose a graph neural network framework that generalizes this approach to other domains. Our method introduces a functional connectivity block based on persistent graph homology to capture global topological features. Combined with structural information, this forms a multi-modal architecture called Functional Connectivity Graph Neural Networks. Experiments show consistent performance gains over existing methods, demonstrating the value of brain-inspired representations for graph-level classification across diverse networks.
Similar Papers
Unified Generative Latent Representation for Functional Brain Graphs
Neurons and Cognition
Lets computers understand brain activity patterns.
Data-Efficient Neural Training with Dynamic Connectomes
Neurons and Cognition
Shows how computer brains learn by watching their activity.
Edge-boosted graph learning for functional brain connectivity analysis
Machine Learning (CS)
Finds brain diseases earlier by looking at brain connections.