Learning Structural-Functional Brain Representations through Multi-Scale Adaptive Graph Attention for Cognitive Insight
This work addresses the problem of multimodal brain data integration for cognitive insight, representing an incremental advancement in domain-specific methods.
The paper tackled the challenge of jointly modeling brain structure and function to explain intelligence by introducing MAGNet, a Transformer-style graph neural network that adaptively learns structure-function interactions, and it outperformed baselines on the ABCD dataset.
Understanding how brain structure and function interact is key to explaining intelligence yet modeling them jointly is challenging as the structural and functional connectome capture complementary aspects of organization. We introduced Multi-scale Adaptive Graph Network (MAGNet), a Transformer-style graph neural network framework that adaptively learns structure-function interactions. MAGNet leverages source-based morphometry from structural MRI to extract inter-regional morphological features and fuses them with functional network connectivity from resting-state fMRI. A hybrid graph integrates direct and indirect pathways, while local-global attention refines connectivity importance and a joint loss simultaneously enforces cross-modal coherence and optimizes the prediction objective end-to-end. On the ABCD dataset, MAGNet outperformed relevant baselines, demonstrating effective multimodal integration for advancing our understanding of cognitive function.