Topology-Aware Graph Augmentation for Predicting Clinical Trajectories in Neurocognitive Disorders
This work addresses a domain-specific problem for researchers and clinicians in neurocognitive disorders, offering an incremental improvement over existing graph contrastive learning methods.
The paper tackles the problem of low model generalizability in predicting clinical trajectories for neurocognitive disorders due to scarce labeled fMRI data by proposing a topology-aware graph augmentation framework, which outperforms state-of-the-art methods on 1,688 fMRI scans.
Brain networks/graphs derived from resting-state functional MRI (fMRI) help study underlying pathophysiology of neurocognitive disorders by measuring neuronal activities in the brain. Some studies utilize learning-based methods for brain network analysis, but typically suffer from low model generalizability caused by scarce labeled fMRI data. As a notable self-supervised strategy, graph contrastive learning helps leverage auxiliary unlabeled data. But existing methods generally arbitrarily perturb graph nodes/edges to generate augmented graphs, without considering essential topology information of brain networks. To this end, we propose a topology-aware graph augmentation (TGA) framework, comprising a pretext model to train a generalizable encoder on large-scale unlabeled fMRI cohorts and a task-specific model to perform downstream tasks on a small target dataset. In the pretext model, we design two novel topology-aware graph augmentation strategies: (1) hub-preserving node dropping that prioritizes preserving brain hub regions according to node importance, and (2) weight-dependent edge removing that focuses on keeping important functional connectivities based on edge weights. Experiments on 1, 688 fMRI scans suggest that TGA outperforms several state-of-the-art methods.