An improved model using convolutional sliding window-attention network for motor imagery EEG classification

Frontiers in Neuroscience

Neuroscience

Jianxu Zheng¹  Hua Feng¹  Shiqi Cao²  Zijian Wang³  Xuhang Li³  Yuxuan Huang³  Binxing Xu³  Yu Liu³ 

[1] Department of Neurosurgery and State Key Laboratory of Trauma, Burn and Combined Injury, Southwest Hospital, Third Military Medical University (Army Medical University), Chongqing, China;Department of Orthopaedics of TCM Clinical Unit, The Sixth Medical Center, Chinese PLA General Hospital, Beijing, China;School of Computer Science and Technology, Donghua University, Shanghai, China;

关键词: EEG;    motor imagery;    brain computer interface;    deep learning;    CNN;    attention;   

DOI  :  10.3389/fnins.2023.1204385

 received in 2023-04-12, accepted in 2023-07-26,  发布年份 2023

来源: Frontiers

【 摘 要 】

IntroductionThe classification model of motor imagery-based electroencephalogram (MI-EEG) is a new human-computer interface pattern and a new neural rehabilitation assessment method for diseases such as Parkinson's and stroke. However, existing MI-EEG models often suffer from insufficient richness of spatiotemporal feature extraction, learning ability, and dynamic selection ability.MethodsTo solve these problems, this work proposed a convolutional sliding window-attention network (CSANet) model composed of novel spatiotemporal convolution, sliding window, and two-stage attention blocks.ResultsThe model outperformed existing state-of-the-art (SOTA) models in within- and between-individual classification tasks on commonly used MI-EEG datasets BCI-2a and Physionet MI-EEG, with classification accuracies improved by 4.22 and 2.02%, respectively.DiscussionThe experimental results also demonstrated that the proposed type token, sliding window, and local and global multi-head self-attention mechanisms can significantly improve the model's ability to construct, learn, and adaptively select multi-scale spatiotemporal features in MI-EEG signals, and accurately identify electroencephalogram signals in the unilateral motor area. This work provided a novel and accurate classification model for MI-EEG brain-computer interface tasks and proposed a feasible neural rehabilitation assessment scheme based on the model, which could promote the further development and application of MI-EEG methods in neural rehabilitation.

【 授权许可】

Unknown   
Copyright © 2023 Huang, Zheng, Xu, Li, Liu, Wang, Feng and Cao.