Convolutional neural network based on temporal-spatial feature learning for motor imagery electroencephalogram signal decoding.
10.7507/1001-5515.202007006
- Author:
Yaqi CHU
1
,
2
,
3
;
Bo ZHU
1
,
2
,
3
;
Xingang ZHAO
1
,
4
;
Yiwen ZHAO
1
,
4
Author Information
1. State Key Laboratory of Robotics, Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang 110016, P.R.China
2. Institutes for Robotics and Intelligent Manufacturing, Chinese Academy of Sciences, Shenyang 110016, P.R.China
3. University of Chinese Academy of Sciences (UCAS), Beijing 100049, P.R.China.
4. Institutes for Robotics and Intelligent Manufacturing, Chinese Academy of Sciences, Shenyang 110016, P.R.China.
- Publication Type:Journal Article
- Keywords:
brain-computer interface;
convolutional neural network;
motor imagery electroencephalogram;
signal decoding;
temporal-spatial feature
- MeSH:
Algorithms;
Brain-Computer Interfaces;
Electroencephalography;
Humans;
Imagination;
Neural Networks, Computer;
Reproducibility of Results
- From:
Journal of Biomedical Engineering
2021;38(1):1-9
- CountryChina
- Language:Chinese
-
Abstract:
With the advantage of providing more natural and flexible control manner, brain-computer interface systems based on motor imagery electroencephalogram (EEG) have been widely used in the field of human-machine interaction. However, due to the lower signal-noise ratio and poor spatial resolution of EEG signals, the decoding accuracy is relative low. To solve this problem, a novel convolutional neural network based on temporal-spatial feature learning (TSCNN) was proposed for motor imagery EEG decoding. Firstly, for the EEG signals preprocessed by band-pass filtering, a temporal-wise convolution layer and a spatial-wise convolution layer were respectively designed, and temporal-spatial features of motor imagery EEG were constructed. Then, 2-layer two-dimensional convolutional structures were adopted to learn abstract features from the raw temporal-spatial features. Finally, the softmax layer combined with the fully connected layer were used to perform decoding task from the extracted abstract features. The experimental results of the proposed method on the open dataset showed that the average decoding accuracy was 80.09%, which is approximately 13.75% and 10.99% higher than that of the state-of-the-art common spatial pattern (CSP) + support vector machine (SVM) and filter bank CSP (FBCSP) + SVM recognition methods, respectively. This demonstrates that the proposed method can significantly improve the reliability of motor imagery EEG decoding.
