Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

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

Syeda Hareem Madani, Noureen Bibi, Adam Rafiq Jeraj, Sumra Khan, Anas Zafar, Rizwan Qureshi

TL;DR

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.

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.

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

TL;DR

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.

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.
Paper Structure (13 sections, 7 figures, 3 tables)

This paper contains 13 sections, 7 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Methodology
  3. Data Acquisition and Participants
  4. Structural Preprocessing
  5. Functional Preprocessing
  6. Graph Construction and Atlas Selection
  7. Data Augmentation Strategy - Gausian Noise Injection
  8. Deep Learning Architectures
  9. Explainable AI (XAI) Analysis
  10. Results and Discussion
  11. Results
  12. Discussion
  13. Conclusion

Figures (7)

  • Figure 1: The complete fMRI preprocessing pipeline. The workflow progresses from raw data acquisition through standard FSL preprocessing steps to feature extraction and final model training.
  • Figure 2: Visual results of the FSL structural preprocessing pipeline
  • Figure 3: Functional connectivity adjacency matrix. Warmer colors indicate stronger temporal correlation between brain regions
  • Figure 4: Top 10 discriminative brain regions. The prominence of DMN regions confirms biological validity.
  • Figure 5: Results for the AAL-based model classification
  • ...and 2 more figures