MSANet: a multi-scale spatiotemporal adaptive network for EEG-based driver fatigue recognition
9 affiliations
- 1School of Software Engineering, Zhengzhou University of Light Industry
- 2Department of Pediatrics, Zhongshan Hospital, Xiamen University
- 3Department of Neurology, The Third Hospital of Xiamen
- 4Department of Neurology and Geriatrics, Fujian Institute of Geriatrics, Fujian Medical University Union Hospital
- 5Xiamen Shuimu Ruizhi Intelligent Technology Co., Ltd.
- 6Department of Neurology and Department of Neuroscience, The First Affiliated Hospital, School of Medicine, Xiamen University
- 7Xiamen Medical Quality Control Center for Neurology
- 8Department of Geriatrics, The Third Hospital of Xiamen
- 9Department of Geriatrics, The First Affiliated Hospital of Xiamen University
Abstract
Electroencephalography (EEG)-based driver fatigue recognition remains challenging because fatigue-related patterns are distributed across multiple temporal scales, inter-channel relationships vary across samples, and inter-individual differences complicate cross-subject transfer. To address these issues, a multi-scale spatiotemporal adaptive network (MSANet) is proposed for EEG-based driver fatigue recognition. The framework introduces a multi-granularity patch temporal gate (MGPTG) to model locally continuous temporal contexts, stable graph node enhancement (SGNE) to enhance channel-node features before dynamic graph construction, and an adaptive classifier gate (ACG) to combine cosine-based and affine decision geometries. MSANet was evaluated on two public simulated-driving EEG datasets using a cross-subject transfer protocol, in which the target-subject samples were stratified with a partition ratio of 0.5 into a labeled adaptation subset and a held-out test subset. MSANet achieved an accuracy (ACC) of 83.64% and an F1-score (F1) of 83.74% on Dataset I, and 74.64% ACC and 72.92% F1 on Dataset II. The results indicate strong numerical performance on Dataset I and competitive performance on Dataset II. Subject-level statistical analysis was additionally conducted to distinguish mean performance differences from differences that are consistently supported across subjects. Ablation results showed lower mean performance after removing MGPTG, SGNE, or ACG under the evaluated settings. These findings support further evaluation of the proposed stage-specific temporal, graph-node, and classification mechanisms for cross-subject transfer in EEG-based driver fatigue recognition.
Related
- Long-term Coexistence of Epidemics and Risk Awareness: Impacts of Adaptive Human Response and Fatigue
- Neurodynamical heterogeneous spiking neural network for spatiotemporal cognitive signal processing
- Three-dimensional animation compression optimization based on structured processing and adaptive spatiotemporal segmentation algorithm
- VAD-EmotionNet: a spatial-spectral attention network for EEG emotion recognition in the 3D affective space
- Adaptive spatiotemporal filtering of social electrosensory signals by cerebellar feedback
BibTeX
@article{gu2026msanet,
title = {{MSANet: a multi-scale spatiotemporal adaptive network for EEG-based driver fatigue recognition}},
author = {Peipei Gu and Ruibo Wang and Ming Zhang and Naian Xiao and Fangqin Liu and Bin Jiang and Jiayang Guo and Xiaosong Peng},
journal = {Frontiers in Neuroscience},
year = {2026},
month = oct,
doi = {10.3389/fnins.2026.1939908},
url = {https://www.frontiersin.org/articles/10.3389/fnins.2026.1939908/full},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - MSANet: a multi-scale spatiotemporal adaptive network for EEG-based driver fatigue recognition
AU - Peipei Gu
AU - Ruibo Wang
AU - Ming Zhang
AU - Naian Xiao
AU - Fangqin Liu
AU - Bin Jiang
AU - Jiayang Guo
AU - Xiaosong Peng
PY - 2026
DA - 2026/10/06
JO - Frontiers in Neuroscience
DO - 10.3389/fnins.2026.1939908
UR - https://www.frontiersin.org/articles/10.3389/fnins.2026.1939908/full
AB - Electroencephalography (EEG)-based driver fatigue recognition remains challenging because fatigue-related patterns are distributed across multiple temporal scales, inter-channel relationships vary across samples, and inter-individual differences complicate cross-subject transfer. To address these issues, a multi-scale spatiotemporal adaptive network (MSANet) is proposed for EEG-based driver fatigue recognition. The framework introduces a multi-granularity patch temporal gate (MGPTG) to model locally continuous temporal contexts, stable graph node enhancement (SGNE) to enhance channel-node features before dynamic graph construction, and an adaptive classifier gate (ACG) to combine cosine-based and affine decision geometries. MSANet was evaluated on two public simulated-driving EEG datasets using a cross-subject transfer protocol, in which the target-subject samples were stratified with a partition ratio of 0.5 into a labeled adaptation subset and a held-out test subset. MSANet achieved an accuracy (ACC) of 83.64% and an F1-score (F1) of 83.74% on Dataset I, and 74.64% ACC and 72.92% F1 on Dataset II. The results indicate strong numerical performance on Dataset I and competitive performance on Dataset II. Subject-level statistical analysis was additionally conducted to distinguish mean performance differences from differences that are consistently supported across subjects. Ablation results showed lower mean performance after removing MGPTG, SGNE, or ACG under the evaluated settings. These findings support further evaluation of the proposed stage-specific temporal, graph-node, and classification mechanisms for cross-subject transfer in EEG-based driver fatigue recognition.
SN - 1662-453X
SN - 1662-4548
ER -