Beyond Anatomy: Explainable ASD Classification from rs-fMRI via Functional Parcellation and Graph Attention Networks

2026-03-02

AI Summary

This study addresses the limitations of conventional anatomical atlas–based approaches for classifying autism spectrum disorder (ASD) from resting-state fMRI, which struggle to capture individualized functional connectivity patterns due to fixed regional boundaries. The authors propose a graph neural network (GNN)–based classification framework that systematically demonstrates, for the first time, the superior efficacy of a functional parcellation atlas (MSDL) over anatomical atlases in ASD classification. The method integrates multi-site data harmonization, Gaussian noise augmentation, and a graph attention mechanism to enhance robustness. Neuroscientific interpretability is achieved through GNNExplainer and gradient-based saliency analysis, effectively mitigating reliance on acquisition artifacts. Evaluated on the ABIDE I dataset, the model achieves 95.0% accuracy and an AUC of 0.98—outperforming anatomical atlas–based approaches by 10.7 percentage points and surpassing existing GNN methods.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Natural Language Processing: Language Grounding & Multi-modal NLP

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization

Abstract

Anatomical brain parcellations dominate rs-fMRI-based Autism Spectrum Disorder (ASD) classification, yet their rigid boundaries may fail to capture the idiosyncratic connectivity patterns that characterise ASD. We present a graph-based deep learning framework comparing anatomical (AAL, 116 ROIs) and functionally-derived (MSDL, 39 ROIs) parcellation strategies on the ABIDE I dataset. Our FSL preprocessing pipeline handles multi-site heterogeneity across 400 balanced subjects, with site-stratified 70/15/15 splits to prevent data leakage. Gaussian noise augmentation within training folds expands samples from 280 to 1,680. A three phase pipeline progresses from a baseline GCN with AAL (73.3% accuracy, AUC=0.74), to an optimised GCN with MSDL (84.0%, AUC=0.84), to a Graph Attention Network ensemble achieving 95.0% accuracy (AUC=0.98), outperforming all recent GNN-based benchmarks on ABIDE I. The 10.7-point gain from atlas substitution alone demonstrates that functional parcellation is the most impactful modelling decision. Gradient-based saliency and GNNExplainer analyses converge on the Posterior Cingulate Cortex and Precuneus as core Default Mode Network hubs, validating that model decisions reflect ASD neuropathology rather than acquisition artefacts. All code and datasets will be publicly released upon acceptance.

Problem

• Autism Spectrum Disorder
• rs-fMRI
• functional parcellation
• anatomical parcellation
• connectivity patterns

Innovation

• functional parcellation
• graph attention network
• explainable AI
• rs-fMRI
• autism spectrum disorder