DuSCN-FusionNet: An Interpretable Dual-Channel Structural Covariance Fusion Framework for ADHD Classification Using Structural MRI

Qurat Ul Ain; Alptekin Temizel; Soyiba Jawed

cs.CVMar 27, 2026

DuSCN-FusionNet: An Interpretable Dual-Channel Structural Covariance Fusion Framework for ADHD Classification Using Structural MRI

Authors: Qurat Ul Ain, Alptekin Temizel, Soyiba Jawed

Organizations: 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

Abstract

Attention Deficit Hyperactivity Disorder (ADHD) is a highly prevalent neurodevelopmental condition; however, its neurobiological diagnosis remains challenging due to the lack of reliable imaging-based biomarkers, particularly anatomical markers. Structural MRI (sMRI) provides a non-invasive modality for investigating brain alterations associated with ADHD; nevertheless, most deep learning approaches function as black-box systems, limiting clinical trust and interpretability. In this work, we propose DuSCN-FusionNet, an interpretable sMRI-based framework for ADHD classification that leverages dual-channel Structural Covariance Networks (SCNs) to capture inter-regional morphological relationships. ROI-wise mean intensity and intra-regional variability descriptors are used to construct intensity-based and heterogeneity-based SCNs, which are processed through an SCN-CNN encoder. In parallel, auxiliary ROI-wise variability features and global statistical descriptors are integrated via late-stage fusion to enhance performance. The model is evaluated using stratified 10-fold cross-validation with a 5-seed ensemble strategy, achieving a mean balanced accuracy of 80.59% and an AUC of 0.778 on the Peking University site of the ADHD-200 dataset. DuSCN-FusionNet further achieves precision, recall, and F1-scores of 81.66%, 80.59%, and 80.27%, respectively. Moreover, Grad-CAM is adapted to the SCN domain to derive ROI-level importance scores, enabling the identification of structurally relevant brain regions as potential biomarkers.

Figures & tables

Fig. 1: Overview of the proposed DuSCN-FusionNet framework for ADHD classification.

Fig. 2: ROC curve for DuSCN-FusionNet in ADHD vs. HC classification (AUC = 0.778), outperforming random chance.

Fig. 3: Confusion matrix of DuSCN-FusionNet predictions for ADHD and HCs, summarizing classification performance across subjects.

Fig. 4: ROI-level interpretability analysis using Grad-CAM on the SCN-CNN branch.

Fig. 5: Top 12 most salient brain regions identified by Grad-CAM. ROIs above the 90th percentile threshold are retained as candidate structural biomarkers for ADHD classification.

Method	Sample (ADHD/HC)	Features	Classifier(s)	ACC (%)	Bio-markers
Wang et al. [12]	146 / 441	Raw sMRI	3D MVA-CNN	78.8	No
Kuttala et al. [13]	77 / 94	Multilayer Perceptron	Dense Attentive GAN	85.38	No
Zou et al. [23]	197 / 362	GM, WM, CSF	3D CNN	65.86	No
DuSCN-FusionNet-(ours)	78 / 116	SCNs + IQR + Global Features	CNN-MLP	80.59	Yes

TABLE I: Comparison of ADHD classification performance with previous studies and the proposed DuSCN-FusionNet pipeline.

Pipeline Variant	ACC (%)	PRE (%)	REC (%)	F1 (%)
Without Auxiliary features	62.01 $\pm$ 13.75	63.57 $\pm$ 14.59	62.01 $\pm$ 13.75	60.03 $\pm$ 13.30
Without Seed Ensemble	72.14 $\pm$ 6.17	74.04 $\pm$ 6.17	72.14 $\pm$ 6.17	71.50 $\pm$ 6.52
DuSCN-FusionNet	80.59 $\pm$ 9.39	81.66 $\pm$ 8.87	80.59 $\pm$ 9.39	80.27 $\pm$ 9.28

TABLE II: Ablation study of the DuSCN-FusionNet pipeline showing classification performance.

Explore similar work

D-GATNet: Interpretable Temporal Graph Attention Learning for ADHD Identification Using Dynamic Functional Connectivity
Mar 27, 2026Qurat Ul Ain, Alptekin Temizel, Soyiba JawedClassification Neuroimaging
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
HyperAMS-Net: Adaptive Multi-Scale Spatial Hypergraph Network for Brain Disorder Classification
Sep 17, 2026Proloy Kumar Mondal, Md Kamran Hussin Chowdhury, Hoi Leong LeeMedical Diagnosis Multi-Scale Feature Fusion
Department of Computer Science and Engineering, University of North Texas, Denton, Texas, USA · Inje University, Gimhae-si, Republic of Korea · Faculty of Electronic Engineering and Technology, Universiti Malaysia Perlis, Arau, Malaysia
HybridDualPathNet: Cross-Fusion Attention of EfficientNetV2 and Swin Transformer for Brain MRI Classification of Alzheimer's and Parkinson's Disease
Oct 12, 2026Jarpula Nirjala, Rahul PatelMedical Image Classification Alzheimer's Disease Classification
Department of Electronics and Communication Engineering, Indian Institute of Information Technology, Surat, India

Mar 27, 2026J/K move · Enter open · S save

D-GATNet: Interpretable Temporal Graph Attention Learning for ADHD Identification Using Dynamic Functional Connectivity

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

Classification Neuroimaging Graph Attention Networks

Open paper AskPDF arXiv

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.

Abstract

Explore similar work

D-GATNet: Interpretable Temporal Graph Attention Learning for ADHD Identification Using Dynamic Functional Connectivity

HyperAMS-Net: Adaptive Multi-Scale Spatial Hypergraph Network for Brain Disorder Classification

HybridDualPathNet: Cross-Fusion Attention of EfficientNetV2 and Swin Transformer for Brain MRI Classification of Alzheimer's and Parkinson's Disease