Organizations: Computational and Data Science, Middle Tennessee State University, 1301 E. Main St., Murfreesboro, 37132, TN, USA. · Department of Mathematical Sciences, Middle Tennessee State University, 1301 E. Main St., Murfreesboro, 37132, TN, USA.
Autism Spectrum Disorder (ASD) is a neurodevelopmental condition characterized by atypical brain connectivity. One of the crucial steps in addressing ASD is its early detection. This study introduces a novel computational framework that employs an Attention-Based Graph Convolutional Network, referred to as the GATGraphClassifier, for detecting ASD. We utilize Functional Magnetic Resonance Imaging (fMRI) data from the Autism Brain Imaging Data Exchange (ABIDE) repository to construct functional connectivity matrices using Pearson correlation, which captures interactions between various brain regions. These matrices are then transformed into graph representations, where the nodes and edges represent the brain regions and functional connections, respectively. The GATGraphClassifier employs attention mechanisms to identify critical connectivity patterns, thereby enhancing the model's interpretability and diagnostic accuracy. Our proposed framework demonstrates superior performance across all standard classification metrics compared to existing state-of-the-art methods. Notably, we achieved an average accuracy of 88.79% on the test data over 30 independent runs, surpassing the benchmark model's performance by 12.27%. In addition, we identified the crucial brain regions associated with ASD, consistent with the previous studies, and a few novel regions. This study not only contributes to the advancement of ASD detection but also shows the potential for broader adaptability of GATGraphClassifier in analyzing complex relational data in various fields, where understanding intricate connectivity and interaction patterns is essential.
Figures & tables
Figure 1: Flow of Overall Approach
Figure 2
Layer(s)
Details
Output Shape
Activation
Input
-
(N, 316)
-
GAT Block (1)
GATConv (8 heads), BatchNorm, Dropout (p=0.1)
(N, 2048)
ELU
GAT Blocks (2–7)
GATConv (8 heads), BatchNorm, Dropout (p=0.2)
(N, 2048)
ELU
Pooling
Global Mean Pool
(N, 2048)
-
FC1
Linear (2048 → 1024), Dropout (p=0.2)
(N, 1024)
ReLU
FC2
Linear (1024 → 2)
(N, 2)
LogSoftmax
Table 1: GAT-Based Model Architecture
Figure 4Figure 5Figure 6
Metric
Average
Minimum
Maximum
Standard Deviation
Accuracy
88.79
79.55
96.59
3.72
Precision
86.50
75.00
95.24
4.34
Recall
90.16
75.61
97.56
5.02
AUC
0.96
0.90
0.99
0.02
F1 Score
0.88
0.78
0.96
0.04
Table 2: Performance Metrics of our Model
Method
Accuracy
Precision
Recall
AUC
ASD-DiagNet ( Eslami et al., 2019 )
69.07
65.25
69.58
0.71
ASD-SWNet ( Zhang et al., 2024 )
76.52
76.15
80.65
0.81
Hi-GCN ( Jiang et al., 2020 )
72.63
65.16
70.54
0.79
GCN ( Parisot et al., 2018 )
70.58
68.32
74.29
0.72
Our Approach
88.79
86.50
90.16
0.96
Table 3: Performance of our Model vs. Existing Models (from ( Zhang et al., 2024 ) )
Shapley (with Score)
Attention (with Score)
Vermis 8 ( 3.294×10−4 )
Vermis 6 (3.706)
Cingulum Ant R ( 3.292×10−4 )
Vermis 8 (3.658)
Temporal Inf L ( 2.004×10−4 )
Vermis 9 (3.625)
SupraMarginal R ( 1.947×10−4 )
Temporal Inf R (3.558)
Supp Motor Area L ( 1.800×10−4 )
Cerebelum 4 5 R (3.444)
Table 4: Top 5 Most Important Brain Regions based on Shapley Score and Attention Mechanism Score
Autism spectrum disorder (ASD) is a complex neurodevelopmental condition characterized by atypical functional brain connectivity and subtle structural alterations. rs-fMRI has been widely used to identify disruptions in large-scale brain networks, while structural MRI provides complementary information about morphological organization. Despite their complementary nature, effectively integrating these heterogeneous imaging modalities within a unified framework remains challenging. This study proposes a multimodal graph learning framework that preserves the dominant role of functional connectivity while integrating structural imaging and phenotypic information for ASD classification. The proposed framework is evaluated on ABIDE-I dataset. Each subject is represented as a node within a population graph. Functional and structural features are extracted as modality-specific node attributes, while inter-subject relationships are modeled using a pairwise association encoder (PAE) based on phenotypic information. Two Edge Variational GCNs are trained to learn subject-level embeddings. To enable effective multimodal integration, we introduce a novel asymmetric transformer-based cross-attention mechanism that allows functional embeddings to selectively incorporate complementary structural information while preserving functional dominance. The fused embeddings are then passed to a MLP for ASD classification. Using stratified 10-fold cross-validation, the framework achieved an AUC of 87.3% and an accuracy of 84.4%. Under leave-one-site-out cross-validation (LOSO-CV), the model achieved an average cross-site accuracy of 82.0%, outperforming existing methods by approximately 3% under 10-fold cross-validation and 7% under LOSO-CV. The proposed framework effectively integrates heterogeneous multimodal data from the multi-site ABIDE-I dataset, improving automated ASD classification across imaging sites.
Department of Medicine, Danube Private University (DPU), Krems, Austria · Department of Medical Physics and Biomedical Engineering, Medical University of Vienna, Vienna, Austria · Department of Medical Informatics, Faculty of Medicine, Gonabad University of Medical Sciences, Gonabad, Iran +1
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.
Syeda Hareem Madani, Noureen Bibi, Adam Rafiq Jeraj +3
Salim Habib University · VentureDive · National University of Computer and Emerging Sciences
Attention Deficit Hyperactivity Disorder (ADHD) is a prevalent neurodevelopmental disorder whose neuroimaging-based diagnosis remains challenging due to complex time-varying disruptions in brain connectivity. Functional MRI (fMRI) provides a powerful non-invasive modality for identifying functional alterations. Existing deep learning (DL) studies employ diverse neuroimaging features; however, static functional connectivity remains widely used, whereas dynamic connectivity modeling is comparatively underexplored. Moreover, many DL models lack interpretability. In this work, we propose D-GATNet, an interpretable temporal graph-based framework for automated ADHD classification using dynamic functional connectivity (dFC). Sliding-window Pearson correlation constructs sequences of functional brain graphs with regions of interest as nodes and connectivity strengths as edges. Spatial dependencies are learned via a multi-layer Graph Attention Network, while temporal dynamics are modeled using 1D convolution followed by temporal attention. Interpretability is achieved through graph attention weights revealing dominant ROI interactions, ROI importance scores identifying influential regions, and temporal attention emphasizing informative connectivity segments. Experiments on the Peking University site of the ADHD-200 dataset using stratified 10-fold cross-validation with a 5-seed ensemble achieved 85.18% +_5.64 balanced accuracy and 0.881 AUC, outperforming state-of-the-art methods. Attention analysis reveals cerebellar and default mode network disruptions, indicating potential neuroimaging biomarkers.
Qurat Ul Ain, Alptekin Temizel, Soyiba Jawed
Department of Computer and Software Engineering, College of Electrical and Mechanical Engineering, National University of Sciences and Technology, Islamabad 44000, Pakistan · Graduate School of Informatics, Middle East Technical University, 06800, Ankara, Turkey