Pooling Regularized Graph Neural Network for fMRI Biomarker Analysis

TL;DR

This paper introduces PR-GNN, a novel graph neural network framework with regularized pooling layers designed to identify brain regions associated with neurological disorders from fMRI data, improving classification accuracy and biomarker detection.

Contribution

The paper presents a new interpretable GNN framework with a salient region selection mechanism and regularized pooling layers for biomarker analysis in neuroimaging.

Findings

PR-GNN outperforms baseline methods in classification accuracy.

Salient ROI detection aligns with known ASD biomarkers.

Framework provides flexible ROI selection at individual or group level.

Abstract

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.