TY - JOUR
T1 - Joint Filter-Band-Combination and Multi-View CNN for Electroencephalogram Decoding
AU - Fan, Zhuyao
AU - Xi, Xugang
AU - Gao, Yunyuan
AU - Wang, Ting
AU - Fang, Feng
AU - Houston, Michael
AU - Zhang, Yingchun
AU - Li, Lihua
AU - Lu, Zhong
N1 - Publisher Copyright:
© 2001-2011 IEEE.
PY - 2023
Y1 - 2023
N2 - Motor imagery (MI) electroencephalogram (EEG) signals have an important role in brain-computer interface (BCI) research. However, effectively decoding these signals remains a problem to be solved. Traditional EEG signal decoding algorithms rely on parameter design to extract features, whereas deep learning algorithms represented by convolution neural network (CNN) can automatically extract features, which is more suitable for BCI applications. However, when EEG data is taken as input in raw time series, traditional 1D-CNNs are unable to acquire both frequency domain and channel association information. To solve this problem, this study proposes a novel algorithm by inserting two modules into CNN. One is the Filter Band Combination (FBC) Module, which preserves as many frequency domain features as possible while maintaining the time domain characteristics of EEG. Another module is Multi-View structure that can extract features from the output of FBC module. To prevent over fitting, we used a cosine annealing algorithm with restart strategy to update the learning rate. The proposed algorithm was validated on the BCI competition dataset and the experiment dataset, using accuracy, standard deviation, and kappa coefficient. Compared with traditional decoding algorithms, our proposed algorithm achieved an improvement of the maximum average correct rate of 6.6% on the motion imagery 4-classes recognition mission and 11.3% on the 2-classes classification task.
AB - Motor imagery (MI) electroencephalogram (EEG) signals have an important role in brain-computer interface (BCI) research. However, effectively decoding these signals remains a problem to be solved. Traditional EEG signal decoding algorithms rely on parameter design to extract features, whereas deep learning algorithms represented by convolution neural network (CNN) can automatically extract features, which is more suitable for BCI applications. However, when EEG data is taken as input in raw time series, traditional 1D-CNNs are unable to acquire both frequency domain and channel association information. To solve this problem, this study proposes a novel algorithm by inserting two modules into CNN. One is the Filter Band Combination (FBC) Module, which preserves as many frequency domain features as possible while maintaining the time domain characteristics of EEG. Another module is Multi-View structure that can extract features from the output of FBC module. To prevent over fitting, we used a cosine annealing algorithm with restart strategy to update the learning rate. The proposed algorithm was validated on the BCI competition dataset and the experiment dataset, using accuracy, standard deviation, and kappa coefficient. Compared with traditional decoding algorithms, our proposed algorithm achieved an improvement of the maximum average correct rate of 6.6% on the motion imagery 4-classes recognition mission and 11.3% on the 2-classes classification task.
KW - Electroencephalogram
KW - brain-computer interface
KW - convolutional neural networks
KW - motor imagery
UR - http://www.scopus.com/inward/record.url?scp=85153794108&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85153794108&partnerID=8YFLogxK
U2 - 10.1109/TNSRE.2023.3269055
DO - 10.1109/TNSRE.2023.3269055
M3 - Article
C2 - 37083516
AN - SCOPUS:85153794108
SN - 1534-4320
VL - 31
SP - 2101
EP - 2110
JO - IEEE Transactions on Neural Systems and Rehabilitation Engineering
JF - IEEE Transactions on Neural Systems and Rehabilitation Engineering
ER -