TY - JOUR
T1 - Deep EEG feature learning via stacking common spatial pattern and support matrix machine
AU - Liang, Shuang
AU - Hang, Wenlong
AU - Yin, Mingbo
AU - Shen, Hang
AU - Wang, Qiong
AU - Qing, Jin
AU - Choi, Kup Sze
AU - Zhang, Yu
N1 - Funding Information:
This work was supported in part by the Key R&D Program of Guangdong Province, China (2018B030339001), by the National Natural Science Foundation of China (61902197, 61802177), by the Open Project of State Key Laboratory for Novel Software Technology at Nanjing University (KFKT2020B11), by NUPTSF (NY219034), by the Natural Science Foundation of Jiangsu Province (BK20201357), by the Six Talent Peaks Project in Jiangsu Province (RJFW-020), and by the Hong Kong Research Grants Council under Grant (PolyU 152006/19E).
Publisher Copyright:
© 2022 Elsevier Ltd
PY - 2022/4
Y1 - 2022/4
N2 - Deep stacked networks (DSNs) have shown promising performance in electroencephalogram (EEG) pattern decoding by recursively enhancing the separability of input features with the supervised information in the stacks. However, most DSN-based models take pre-extracted EEG features as input, which adversely affects the learning of high-level EEG feature representation when the informative neural patterns are not fully captured by the input features. To overcome this issue, we propose a novel deep stacked architecture called Deep Stacked Feature Representation (DSFR) that allows a network to be fed with raw EEG data for automatic learning of the high-level representation and abstraction. The proposed deep stacked architecture utilizes a series of feature decoding modules (FDMs) as the base building blocks, which incorporate random projections as its stacking elements. In each FDM, a feature extractor common spatial pattern (CSP) and a matrix classifier support matrix machine (SMM) are included and stacked in a chain structure. The random projections of the predictions of SMM from all the previous FDMs are integrated into the raw EEG data, which are then fed into the CSP in the subsequent FDMs to generate the EEG feature representation recursively. The proposed DSFR is carried out in an efficient feed-forward way and does not need parameter fine-tuning with backpropagation, resulting in a simplified optimization process. Extensive experiments are conducted on three publicly available motor imagery (MI)-based EEG datasets to evaluate the performance of the proposed DSFR method. The results show that DSFR outperforms the state-of-the-art methods.
AB - Deep stacked networks (DSNs) have shown promising performance in electroencephalogram (EEG) pattern decoding by recursively enhancing the separability of input features with the supervised information in the stacks. However, most DSN-based models take pre-extracted EEG features as input, which adversely affects the learning of high-level EEG feature representation when the informative neural patterns are not fully captured by the input features. To overcome this issue, we propose a novel deep stacked architecture called Deep Stacked Feature Representation (DSFR) that allows a network to be fed with raw EEG data for automatic learning of the high-level representation and abstraction. The proposed deep stacked architecture utilizes a series of feature decoding modules (FDMs) as the base building blocks, which incorporate random projections as its stacking elements. In each FDM, a feature extractor common spatial pattern (CSP) and a matrix classifier support matrix machine (SMM) are included and stacked in a chain structure. The random projections of the predictions of SMM from all the previous FDMs are integrated into the raw EEG data, which are then fed into the CSP in the subsequent FDMs to generate the EEG feature representation recursively. The proposed DSFR is carried out in an efficient feed-forward way and does not need parameter fine-tuning with backpropagation, resulting in a simplified optimization process. Extensive experiments are conducted on three publicly available motor imagery (MI)-based EEG datasets to evaluate the performance of the proposed DSFR method. The results show that DSFR outperforms the state-of-the-art methods.
KW - Common spatial pattern
KW - Deep stacked network
KW - Electroencephalography
KW - Feature representation
KW - Support matrix machine
UR - http://www.scopus.com/inward/record.url?scp=85123850726&partnerID=8YFLogxK
U2 - 10.1016/j.bspc.2022.103531
DO - 10.1016/j.bspc.2022.103531
M3 - Journal article
AN - SCOPUS:85123850726
SN - 1746-8094
VL - 74
JO - Biomedical Signal Processing and Control
JF - Biomedical Signal Processing and Control
M1 - 103531
ER -