SurfGNN: A robust surface-based prediction model with interpretability for coactivation maps of spatial and cortical features

TL;DR

SurfGNN is an interpretable surface-based graph neural network that effectively predicts neonatal brain age from cortical features, outperforming existing methods and providing regional activation insights.

Contribution

This work introduces SurfGNN, a novel GNN model with topology-sampling and region-specific learning for cortical surface prediction tasks.

Findings

Achieves at least 9% improvement over state-of-the-art methods.

Attains a mean absolute error of 0.827 weeks in neonatal brain age prediction.

Generates interpretable regional activation maps.

Abstract

Current brain surface-based prediction models often overlook the variability of regional attributes at the cortical feature level. While graph neural networks (GNNs) excel at capturing regional differences, they encounter challenges when dealing with complex, high-density graph structures. In this work, we consider the cortical surface mesh as a sparse graph and propose an interpretable prediction model-Surface Graph Neural Network (SurfGNN). SurfGNN employs topology-sampling learning (TSL) and region-specific learning (RSL) structures to manage individual cortical features at both lower and higher scales of the surface mesh, effectively tackling the challenges posed by the overly abundant mesh nodes and addressing the issue of heterogeneity in cortical regions. Building on this, a novel score-weighted fusion (SWF) method is implemented to merge nodal representations associated with each cortical feature for prediction. We apply our model to a neonatal brain age prediction task using a dataset of harmonized MR images from 481 subjects (503 scans). SurfGNN outperforms all existing state-of-the-art methods, demonstrating an improvement of at least 9.0% and achieving a mean absolute error (MAE) of 0.827+0.056 in postmenstrual weeks. Furthermore, it generates feature-level activation maps, indicating its capability to identify robust regional variations in different morphometric contributions for prediction.