Decoding motor imagery EEG signals by a deep learning network integrating multiscale filtering and temporal-channel attention mechanisms
- 1Department of Information Engineering, Gongqing College of Nanchang University, Jiujiang, China.
- 2School of Information Engineering, Jiujiang Polytechnic University of Science and Technology, Jiujiang, China.
- 3School of Information Engineering, Nanchang University, Nanchang, China.
Abstract
Due to their low signal-to-noise ratio and non-stationarity, accurately classifying EEG signals is still one of the main hindrances that limit the real-world applications of a brain-computer interface (BCI). Deep learning (DL) is a potential method for addressing the problem. However, how to create a suitable DL network for motor imagery (MI)-based BCIs poses severe challenges in the case of insufficient training data. In this paper, a DL network based on multiscale convolutional neural network (CNN) and temporal and channel attention mechanisms is proposed for decoding MI tasks. Temporal and spatial features are extracted by four parallel CNN branches, each of which is followed by a Transformer block for learning the global correlations of local time features. The outputs from the four Transformer blocks are concatenated along channel dimension and then input into a channel attention module for channel fusion. The subsequent feature map is classified by a CNN, which replaces the commonly used fully connected layer for reducing computational complexity. The proposed network was evaluated on three publicly available BCI datasets and compared with five state-of-the-art DL networks. The training data sets are augmented using the approach segmentation and reconstruction. The experimental results demonstrate that the proposed network achieved an average accuracy of 80.21% on Dataset 1 (4-class tasks), 90.19% on Dataset 2 (2-class tasks), and 97.32% on Dataset 3 (4-class tasks), surpassing the strongest baselines FBCNet on Dataset 1, MSFBCNN on Dataset 2 and MSFBCNN on Dataset 3 by 2.59%, 1.27%, and 1.99% respectively. The network effectively decoded the MI-based EEG signals, providing a potential method for BCI applications.
Related
- Multiscale predictive cellular modeling: integrating hypothesis grammars, digital twins, and multi-omics for In silico oncology and precision theranostics
- Decoding spatial attention in the cocktail party problem using whole-head high-density fNIRS
- OpenDvM: Decoding the Visual Mind — A Unified Toolbox for EEG Attention Research
- Limitations of Genomic Foundation Models for Decoding Regulatory Mechanisms in ALS
- Decoding Cell-type Specific Gene Regulation with CellReg3D by Integrating DNA Large Language Models and 3D Genome Structure
BibTeX
@article{zhu2026decoding,
title = {{Decoding motor imagery EEG signals by a deep learning network integrating multiscale filtering and temporal-channel attention mechanisms}},
author = {Yong Zhu and Yongge Shi and Jianhua Wu and Qingguo Wei},
journal = {PLOS One},
year = {2026},
month = oct,
doi = {10.1371/journal.pone.0361026},
url = {https://doi.org/10.1371/journal.pone.0361026},
note = {CC BY},
}
RIS
TY - JOUR
TI - Decoding motor imagery EEG signals by a deep learning network integrating multiscale filtering and temporal-channel attention mechanisms
AU - Yong Zhu
AU - Yongge Shi
AU - Jianhua Wu
AU - Qingguo Wei
PY - 2026
DA - 2026/10/23
JO - PLOS One
DO - 10.1371/journal.pone.0361026
UR - https://doi.org/10.1371/journal.pone.0361026
AB - Due to their low signal-to-noise ratio and non-stationarity, accurately classifying EEG signals is still one of the main hindrances that limit the real-world applications of a brain-computer interface (BCI). Deep learning (DL) is a potential method for addressing the problem. However, how to create a suitable DL network for motor imagery (MI)-based BCIs poses severe challenges in the case of insufficient training data. In this paper, a DL network based on multiscale convolutional neural network (CNN) and temporal and channel attention mechanisms is proposed for decoding MI tasks. Temporal and spatial features are extracted by four parallel CNN branches, each of which is followed by a Transformer block for learning the global correlations of local time features. The outputs from the four Transformer blocks are concatenated along channel dimension and then input into a channel attention module for channel fusion. The subsequent feature map is classified by a CNN, which replaces the commonly used fully connected layer for reducing computational complexity. The proposed network was evaluated on three publicly available BCI datasets and compared with five state-of-the-art DL networks. The training data sets are augmented using the approach segmentation and reconstruction. The experimental results demonstrate that the proposed network achieved an average accuracy of 80.21% on Dataset 1 (4-class tasks), 90.19% on Dataset 2 (2-class tasks), and 97.32% on Dataset 3 (4-class tasks), surpassing the strongest baselines FBCNet on Dataset 1, MSFBCNN on Dataset 2 and MSFBCNN on Dataset 3 by 2.59%, 1.27%, and 1.99% respectively. The network effectively decoded the MI-based EEG signals, providing a potential method for BCI applications.
SN - 1932-6203
ER -