Comprehensive common spatial patterns with temporal structure information of EEG data : minimizing nontask related EEG component
In the context of electroencephalogram (EEG)-based brain-computer interfaces (BCI), common spatial patterns (CSP) is widely used for spatially filtering multichannel EEG signals. CSP is a supervised learning technique depending on only labeled trials. Its generalization performance deteriorates due...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Journal Article |
| Language: | English |
| Published: |
2013
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/95913 http://hdl.handle.net/10220/11253 |
| _version_ | 1826114300856500224 |
|---|---|
| author | Wang, Haixian Xu, Dong |
| author2 | School of Computer Engineering |
| author_facet | School of Computer Engineering Wang, Haixian Xu, Dong |
| author_sort | Wang, Haixian |
| collection | NTU |
| description | In the context of electroencephalogram (EEG)-based brain-computer interfaces (BCI), common spatial patterns (CSP) is widely used for spatially filtering multichannel EEG signals. CSP is a supervised learning technique depending on only labeled trials. Its generalization performance deteriorates due to overfitting occurred when the number of training trials is small. On the other hand, a large number of unlabeled trials are relatively easy to obtain. In this paper, we contribute a comprehensive learning scheme of CSP (cCSP) that learns on both labeled and unlabeled trials. cCSP regularizes the objective function of CSP by preserving the temporal relationship among samples of unlabeled trials in terms of linear representation. The intrinsically temporal structure is characterized by an $ell_1$ graph. As a result, the temporal correlation information of unlabeled trials is incorporated into CSP, yielding enhanced generalization capacity. Interestingly, the regularizer of cCSP can be interpreted as minimizing a nontask related EEG component, which helps cCSP alleviate nonstationarities. Experiment results of single-trial EEG classification on publicly available EEG datasets confirm the effectiveness of the proposed method. |
| first_indexed | 2024-10-01T03:36:55Z |
| format | Journal Article |
| id | ntu-10356/95913 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T03:36:55Z |
| publishDate | 2013 |
| record_format | dspace |
| spelling | ntu-10356/959132020-05-28T07:17:23Z Comprehensive common spatial patterns with temporal structure information of EEG data : minimizing nontask related EEG component Wang, Haixian Xu, Dong School of Computer Engineering DRNTU::Engineering::Computer science and engineering In the context of electroencephalogram (EEG)-based brain-computer interfaces (BCI), common spatial patterns (CSP) is widely used for spatially filtering multichannel EEG signals. CSP is a supervised learning technique depending on only labeled trials. Its generalization performance deteriorates due to overfitting occurred when the number of training trials is small. On the other hand, a large number of unlabeled trials are relatively easy to obtain. In this paper, we contribute a comprehensive learning scheme of CSP (cCSP) that learns on both labeled and unlabeled trials. cCSP regularizes the objective function of CSP by preserving the temporal relationship among samples of unlabeled trials in terms of linear representation. The intrinsically temporal structure is characterized by an $ell_1$ graph. As a result, the temporal correlation information of unlabeled trials is incorporated into CSP, yielding enhanced generalization capacity. Interestingly, the regularizer of cCSP can be interpreted as minimizing a nontask related EEG component, which helps cCSP alleviate nonstationarities. Experiment results of single-trial EEG classification on publicly available EEG datasets confirm the effectiveness of the proposed method. 2013-07-12T01:44:17Z 2019-12-06T19:23:17Z 2013-07-12T01:44:17Z 2019-12-06T19:23:17Z 2012 2012 Journal Article Wang, H., & Xu, D. (2012). Comprehensive common spatial patterns with temporal structure information of EEG data: Minimizing nontask related EEG component. IEEE Transactions on Biomedical Engineering, 59(9), 2496-2505. https://hdl.handle.net/10356/95913 http://hdl.handle.net/10220/11253 10.1109/TBME.2012.2205383 en IEEE transactions on biomedical engineering © 2012 IEEE. |
| spellingShingle | DRNTU::Engineering::Computer science and engineering Wang, Haixian Xu, Dong Comprehensive common spatial patterns with temporal structure information of EEG data : minimizing nontask related EEG component |
| title | Comprehensive common spatial patterns with temporal structure information of EEG data : minimizing nontask related EEG component |
| title_full | Comprehensive common spatial patterns with temporal structure information of EEG data : minimizing nontask related EEG component |
| title_fullStr | Comprehensive common spatial patterns with temporal structure information of EEG data : minimizing nontask related EEG component |
| title_full_unstemmed | Comprehensive common spatial patterns with temporal structure information of EEG data : minimizing nontask related EEG component |
| title_short | Comprehensive common spatial patterns with temporal structure information of EEG data : minimizing nontask related EEG component |
| title_sort | comprehensive common spatial patterns with temporal structure information of eeg data minimizing nontask related eeg component |
| topic | DRNTU::Engineering::Computer science and engineering |
| url | https://hdl.handle.net/10356/95913 http://hdl.handle.net/10220/11253 |
| work_keys_str_mv | AT wanghaixian comprehensivecommonspatialpatternswithtemporalstructureinformationofeegdataminimizingnontaskrelatedeegcomponent AT xudong comprehensivecommonspatialpatternswithtemporalstructureinformationofeegdataminimizingnontaskrelatedeegcomponent |