2026· IEEE Transactions on Instrumentation and Measurement· Vol 75, pp. 4013112-4013112· 0 citations· 46 references
TL;DR
Findings indicate that the learned brain network representations capture distinct and quantifiable abnormalities in ASD-related neural dynamics, establishing the proposed framework as a robust, interpretable, and measurement-aligned tool for objective ASD assessment.
Abstract
Electroencephalography (EEG)-based assessment of autism spectrum disorder (ASD) remains challenging due to the nonstationary, high-dimensional, and heterogeneous nature of neural signals, as well as the difficulty of jointly modeling spatiotemporal dynamics and functional connectivity patterns in a unified framework. Existing approaches often rely on either single-domain features or static graph representations, limiting their ability to capture dynamic neural interactions associated with ASD. To address these limitations, this work proposes a novel EEG-based functional brain network framework with graph learning for automated ASD detection. The framework comprises two key components: 1) a dual-attention-fused Brain-Net graph representation that models EEG-derived functional connectivity and 2) a dynamic spatiotemporal edge progression graph attention (DySTEP-GAT) network for robust classification. Resting-state EEG signals acquired from multiple cortical regions are transformed into multiband functional connectivity matrices, forming refined Brain-Net graphs that encode node-level spatiotemporal characteristics and edge-level interchannel synchronization. A semantic similarity-driven dual-attention mechanism is then employed to fuse rhythm-specific connectivity information into unified functional brain network representations. The proposed DySTEP-GAT incorporates three measurement-centric modules—dynamic edge selection block (DESB), spatiotemporal GAT block (STGATB), and progressive graph embedding block (PGEB)—to effectively model nonstationary neural coupling patterns associated with ASD. Experimental validation on two publicly available EEG datasets demonstrates that the proposed framework achieves high ASD detection accuracy (96 %–97 %), along with improved precision (0.95) and $F1$ -score (0.95) compared to recent state-of-the-art methods. Ablation studies further confirm the effectiveness of attention-based multiband fusion. These findings indicate that the learned brain network representations capture distinct and quantifiable abnormalities in ASD-related neural dynamics, establishing the proposed framework as a robust, interpretable, and measurement-aligned tool for objective ASD assessment.
Autism spectrum disorder (ASD) is a heterogeneous neurodevelopmental condition whose early identification remains challenging because clinical assessment still relies heavily on behavioral observation and expert judgment. Electroencephalography (EEG) offers a noninvasive approach for capturing neural dynamics. Still, EEG-based ASD classification remains difficult because of signal nonstationarity, limited sample sizes, and the risk of subject-identity leakage. This study proposes a comparative temporal deep learning framework for EEG-based ASD classification by evaluating principal component analysis (PCA) and multiscale principal component analysis (MS-PCA) as feature representations combined with a recurrent neural network with bidirectional long short-term memory (RNN-BiLSTM) and a temporal convolutional network with self-attention (TCN-SA). Resting-state eyes-open EEG signals were acquired from only 10 participants, consisting of 5 individuals with ASD and 5 typically developing controls, using a 16-channel acquisition system. The signals were filtered using a fourth-order Butterworth band-pass filter, transformed into PCA or MS-PCA representations, segmented into 4 s windows with 50% overlap, and evaluated using subject-wise 5-fold cross-validation to reduce subject-identity leakage. The results showed that MS-PCA produced higher descriptive performance than PCA in both temporal architectures, with the strongest descriptive result obtained by the MS-PCA + TCN-SA scheme, which achieved a mean accuracy of 97.96 ± 2.37% and balanced precision, recall, F1-score, and specificity. However, the inferential comparison between PCA and MS-PCA did not reach statistical significance at the 0.05 level, and the cohort size was limited to 10 participants. Therefore, these findings should be interpreted as preliminary descriptive evidence within the present cohort rather than evidence of diagnostic readiness or robust clinical applicability. Larger, independent, and demographically diverse EEG datasets with richer clinical characterization are required to confirm the observed trend and evaluate the generalizability of the proposed framework.
M. Muliyadi, Melinda, Yuwaldi Away et al.· Journal of Electronics Elect...· 0 citations
Parkinson’s disease is a progressive neurodegenerative disorder whose reliable diagnosis from non-invasive electrophysiological signals remains challenging because the discriminative information is distributed across spectral bands and across spatially interacting cortical regions. This study proposes an Adaptive Multi-band Dynamic Functional-Connectivity Graph Attention Network (AMD-FCGAT) that represents resting-state electroencephalography as a set of band-resolved brain graphs and learns disease-specific connectivity signatures through attention. For each rhythmic band the framework estimates functional connectivity using the weighted phase-lag index, the phase-locking value and spectral coherence, constructs a sparse electrode graph, and encodes it with a residual multi-head graph attention encoder. An adaptive cross-band gate fuses the band-specific embeddings, and a dynamic temporal self-attention module models the evolution of connectivity across sliding windows before an attention readout produces the classification decision. Experiments on the publicly available University of California San Diego resting-state dataset show that the proposed model attains 97.31 percent accuracy, 97.22 percent F1-score and an area under the curve of 0.998 under a stratified protocol, while a stricter subject-independent evaluation retains 90.97 percent accuracy. Ablation confirms that the cross-band fusion and temporal attention modules contribute the largest gains, and interpretability analysis localises the strongest alterations to central and parietal beta-band synchrony, in agreement with established Parkinsonian neurophysiology. The framework offers an accurate and transparent tool for electroencephalography-based screening.
Sumit Gupta, U. N. Ranjitha, J. Gul Shaira Banu et al.· International journal of com...· 0 citations
The proposed electroencephalography-based ASD detection method was evaluated against previous approaches using the same datasets, showing a substantial improvement in performance, and the need to evaluate the proposed approach on larger datasets to further ensure model generalization.
B. S. Falih, Mohannad K. Sabir, A. Aydın· Brain Topography· 0 citations
Imagined speech (IS) refers to the cognitive process of internally generating words without overt articulation and provides a promising communication pathway for individuals with severe neurological impairments. Electroencephalogram (EEG)-based brain–computer interface (BCI) systems enable automatic decoding of IS; however, reliable measurement and representation of EEG signals remain challenging due to their nonstationary nature and complex interchannel interactions. Most existing methods process EEG channels independently, neglecting functional connectivity patterns critical to speech-related neural dynamics. To address this limitation, our work proposes a sensor-aware graph signal processing (GSP)-based framework that models multichannel EEG measurements as graph-structured data, with electrodes represented as sensing nodes and weighted edges encoding functional relationships. Three graph construction strategies, namely correlation-based, Gaussian kernel-based, and Fundis-based weighting, are investigated to capture intersensor dependencies. Spectral Graph Wavelet Transform (SGWT) features are extracted to jointly characterize spatial connectivity, graph spectral information, and localized EEG signal variations at multiple graph scales, followed by statistical pooling for stability. The proposed framework is evaluated on a publicly available EEG dataset comprising four categories. The Fundis-based SGWT approach achieves classification accuracies of <inline-formula> <tex-math notation="LaTeX">$99.57{\,}\% {\,}\pm {\,}0.11{\,}\%$ </tex-math></inline-formula> for long words, <inline-formula> <tex-math notation="LaTeX">$99.60{\,}\% {\,}\pm {\,}0.15{\,}\%$ </tex-math></inline-formula> for short–long words, <inline-formula> <tex-math notation="LaTeX">$99.45{\,}\% {\,}\pm {\,}0.26{\,}\%$ </tex-math></inline-formula> for short words, and <inline-formula> <tex-math notation="LaTeX">$99.45{\,}\% {\,}\pm {\,}0.20{\,}\%$ </tex-math></inline-formula> for vowels, while maintaining computational efficiency with only 0.074 million parameters, 2.88 ms inference latency, and a compact model size of 4.35 MB. These results demonstrate the effectiveness of graph-based EEG modeling for real-time, low-power wearable BCI systems.
Suman Rekha Dip, H. Meena· IEEE Transactions on Instrum...· 0 citations
This work proposes a cross-attention-guided subject-adaptive graph network (CAS-GNN) model that integrates structural MRI and resting-state functional connectivity data, effectively fusing complementary multimodal information and provides a promising tool for interpretable and robust ASD diagnosis, accelerating biomarker discovery and development.
Yan Tang, Chao Yang, Yihang Xu et al.· Brain Informatics· 0 citations
Electroencephalography (EEG)-based emotion recognition has gained significant attention for applications in mental health monitoring and brain–computer interfaces; however, achieving ro- bust cross-subject generalization remains a major challenge due to high inter-subject variability. This study proposes an adaptive Graph Neural Network (GNN) framework for EEG-based emo- tion recognition, where EEG channels are modeled as graph nodes and functional connectivity is dynamically constructed using correlation-based relationships. Unlike conventional approaches that rely on fixed or fully connected graph structures, the proposed method generates window- wise adaptive graphs, enabling the model to capture time-varying and subject-invariant neural interactions.
EEG signals are segmented into overlapping temporal windows, and frequency-domain features are extracted using power spectral density across standard EEG bands. A Graph Convolutional Network (GCN) is employed to learn spatial dependencies, and the model is evaluated under a strict subject-independent protocol using Leave-One-Subject-Out (LOSO) cross-validation on the DEAP dataset.
Experimental results demonstrate a mean accuracy of 58.95% and a macro-F1 score of 0.438, highlighting the inherent difficulty of cross-subject EEG emotion recognition. Despite moderate accuracy, the proposed framework provides a more realistic evaluation compared to subject- dependent studies and establishes a robust baseline for generalizable emotion recognition. The findings confirm that adaptive graph construction improves functional brain connectivity representation while emphasizing the need for advanced generalization techniques such as domain adaptation and temporal modeling for future research.
Rupali Shishupal, Dr. Jagannath Nalawade· Journal of Intelligent Decis...· 0 citations
