Pooling Regularized Graph Neural Network for fMRI Biomarker Analysis
This work addresses the need for interpretable biomarker analysis in neuroimaging for disorders like Autism Spectrum Disorder, though it is incremental as it builds on existing GNN methods.
The authors tackled the problem of identifying brain biomarkers for neurological disorders by proposing an interpretable Graph Neural Network framework with regularized pooling layers, which outperformed baseline methods in classification accuracy on an Autism Spectrum Disorder fMRI dataset and detected salient regions consistent with known biomarkers.
Understanding how certain brain regions relate to a specific neurological disorder has been an important area of neuroimaging research. A promising approach to identify the salient regions is using Graph Neural Networks (GNNs), which can be used to analyze graph structured data, e.g. brain networks constructed by functional magnetic resonance imaging (fMRI). We propose an interpretable GNN framework with a novel salient region selection mechanism to determine neurological brain biomarkers associated with disorders. Specifically, we design novel regularized pooling layers that highlight salient regions of interests (ROIs) so that we can infer which ROIs are important to identify a certain disease based on the node pooling scores calculated by the pooling layers. Our proposed framework, Pooling Regularized-GNN (PR-GNN), encourages reasonable ROI-selection and provides flexibility to preserve either individual- or group-level patterns. We apply the PR-GNN framework on a Biopoint Autism Spectral Disorder (ASD) fMRI dataset. We investigate different choices of the hyperparameters and show that PR-GNN outperforms baseline methods in terms of classification accuracy. The salient ROI detection results show high correspondence with the previous neuroimaging-derived biomarkers for ASD.