Related Experiment Video
Updated: Aug 5, 2026
05:41
Synchronous Triplanar Reconstruction Integrated with Color Doppler Mapping for Precise and Rapid Localization of Thyroid Lesions
Published on: February 9, 2024
HFG-Net: High-frequency guided multi-view graph convolution and dynamic spatio-temporal fusion for ASD diagnosis
Jiannan Kang1, Zhiyuan Fan2, Zongbing Xiao3
1Hebei University, No. 2666 Qiyi East Road, Baoding 071000, Hebei Province, P.R. China, Baoding, Hebei, 071002, China.
Journal of Neural Engineering
|July 28, 2026
Summary
A new deep learning model, HFG-Net, accurately diagnoses Autism Spectrum Disorder (ASD) using electroencephalogram (EEG) data by analyzing high-frequency brainwave patterns and complex brain connectivity. This approach offers a robust and interpretable method for ASD diagnosis.
Area of Science:
- Neuroscience
- Artificial Intelligence
- Biomedical Engineering
Background:
- Autism Spectrum Disorder (ASD) presents significant heterogeneity in electroencephalogram (EEG) features, challenging current deep learning models.
- Existing methods like CNNs and GCNs struggle with spatio-temporal dependencies and non-Euclidean graph abnormalities in EEG data.
Purpose of the Study:
- To develop a novel deep learning framework, HFG-Net, for improved EEG-based ASD diagnosis.
- To address limitations in capturing spatio-temporal dynamics and handling data heterogeneity in ASD electrophysiological signals.
Main Methods:
- Proposed HFG-Net, incorporating a Channel-Temporal Multi-scale Attention (CTMA) mechanism for spatio-temporal coupling.
- Introduced High-Frequency Guided Multi-View Graph Construction using Beta/Gamma bands for robust topology.
- Implemented a Dynamic Synergistic Fusion module for adaptive feature integration.
Main Results:
- HFG-Net achieved 95.79% classification accuracy and 92.89% F1-score on a clinical dataset (120 subjects).
- The model demonstrated superior performance compared to state-of-the-art methods.
- Achieved 98.43% recognition rate for mild ASD cases, with interpretable analysis aligning with neuropathological findings.
Conclusions:
- HFG-Net effectively models synergistic spatio-temporal dynamics and adapts to sample heterogeneity in EEG data.
- The framework provides an efficient, robust, and interpretable paradigm for EEG-based ASD diagnosis.
- High-frequency guided analysis offers a promising direction for understanding ASD neurophysiology.