Local and global convolutional transformer-based motor imagery EEG classification

被引:1
|
作者
Zhang, Jiayang [1 ]
Li, Kang [1 ]
Yang, Banghua [2 ]
Han, Xiaofei [1 ]
机构
[1] Univ Leeds, Sch Elect Engn, Leeds, England
[2] Shanghai Univ, Sch Mechatron Engn & Automat, Shanghai, Peoples R China
基金
中国国家自然科学基金;
关键词
brain-computer interface; motor imagery; transformer; attention mechanism; Convolutional Neural Network; NEURAL-NETWORKS; RECOGNITION; BCI;
D O I
10.3389/fnins.2023.1219988
中图分类号
Q189 [神经科学];
学科分类号
071006 ;
摘要
Transformer, a deep learning model with the self-attention mechanism, combined with the convolution neural network (CNN) has been successfully applied for decoding electroencephalogram (EEG) signals in Motor Imagery (MI) Brain-Computer Interface (BCI). However, the extremely non-linear, nonstationary characteristics of the EEG signals limits the effectiveness and efficiency of the deep learning methods. In addition, the variety of subjects and the experimental sessions impact the model adaptability. In this study, we propose a local and global convolutional transformer-based approach for MI-EEG classification. The local transformer encoder is combined to dynamically extract temporal features and make up for the shortcomings of the CNN model. The spatial features from all channels and the difference in hemispheres are obtained to improve the robustness of the model. To acquire adequate temporal-spatial feature representations, we combine the global transformer encoder and Densely Connected Network to improve the information flow and reuse. To validate the performance of the proposed model, three scenarios including within-session, cross-session and two-session are designed. In the experiments, the proposed method achieves up to 1.46%, 7.49% and 7.46% accuracy improvement respectively in the three scenarios for the public Korean dataset compared with current state-of-the-art models. For the BCI competition IV 2a dataset, the proposed model also achieves a 2.12% and 2.21% improvement for the cross-session and two-session scenarios respectively. The results confirm that the proposed approach can effectively extract much richer set of MI features from the EEG signals and improve the performance in the BCI applications.
引用
收藏
页数:12
相关论文
共 50 条
  • [1] CTNet: a convolutional transformer network for EEG-based motor imagery classification
    Zhao, Wei
    Jiang, Xiaolu
    Zhang, Baocan
    Xiao, Shixiao
    Weng, Sujun
    [J]. SCIENTIFIC REPORTS, 2024, 14 (01):
  • [2] EEG Classification with Transformer-Based Models
    Sun, Jiayao
    Xie, Jin
    Zhou, Huihui
    [J]. 2021 IEEE 3RD GLOBAL CONFERENCE ON LIFE SCIENCES AND TECHNOLOGIES (IEEE LIFETECH 2021), 2021, : 92 - 93
  • [3] Effects of local and global spatial patterns in EEG motor-imagery classification using convolutional neural network
    Liao, Jacob Jiexun
    Luo, Joy Jiayu
    Yang, Tao
    So, Rosa Qi Yue
    Chua, Matthew Chin Heng
    [J]. BRAIN-COMPUTER INTERFACES, 2020, 7 (3-4) : 47 - 56
  • [4] Three-Branch Temporal-Spatial Convolutional Transformer for Motor Imagery EEG Classification
    Chen, Weiming
    Luo, Yiqing
    Wang, Jie
    [J]. IEEE ACCESS, 2024, 12 : 79754 - 79764
  • [5] EEG-based motor imagery classification using convolutional neural networks with local reparameterization trick
    Huang, Wenqie
    Chang, Wenwen
    Yan, Guanghui
    Yang, Zhifei
    Luo, Hao
    Pei, Huayan
    [J]. EXPERT SYSTEMS WITH APPLICATIONS, 2022, 187
  • [6] MI-CAT: A transformer-based domain adaptation network for motor imagery classification
    Zhang, Dongxue
    Li, Huiying
    Xie, Jingmeng
    [J]. NEURAL NETWORKS, 2023, 165 : 451 - 462
  • [7] Multiscale Convolutional Transformer for EEG Classification of Mental Imagery in Different Modalities
    Ahn, Hyung-Ju
    Lee, Dae-Hyeok
    Jeong, Ji-Hoon
    Lee, Seong-Whan
    [J]. IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING, 2023, 31 : 646 - 656
  • [8] Multiscale Convolutional Transformer for EEG Classification of Mental Imagery in Different Modalities
    Ahn, Hyung-Ju
    Lee, Dae-Hyeok
    Jeong, Ji-Hoon
    Lee, Seong-Whan
    [J]. IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING, 2023, 31 : 646 - 656
  • [9] EEG classification algorithm of motor imagery based on CNN-Transformer fusion network
    Liu, Haofeng
    Liu, Yuefeng
    Wang, Yue
    Liu, Bo
    Bao, Xiang
    [J]. 2022 IEEE INTERNATIONAL CONFERENCE ON TRUST, SECURITY AND PRIVACY IN COMPUTING AND COMMUNICATIONS, TRUSTCOM, 2022, : 1302 - 1309
  • [10] A classification method for EEG motor imagery signals based on parallel convolutional neural network
    Han, Yuexing
    Wang, Bing
    Luo, Jie
    Li, Long
    Li, Xiaolong
    [J]. BIOMEDICAL SIGNAL PROCESSING AND CONTROL, 2022, 71