Correlation-based common spatial pattern (CCSP): A novel extension of CSP for classification of motor imagery signal
Common spatial pattern (CSP) is shown to be an effective pre-processing algorithm in order to discriminate different classes of motor-based EEG signals by obtaining suitable spatial filters. The performance of these filters can be improved by regularized CSP, in which available prior information is...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English English |
| Published: |
Public Library Science
2021
|
| Subjects: | |
| Online Access: | https://eprints.ums.edu.my/id/eprint/32837/1/Correlation-based%20common%20spatial%20pattern%20.pdf https://eprints.ums.edu.my/id/eprint/32837/2/Correlation-based%20common%20spatial%20pattern%201.pdf |
| _version_ | 1825714767379038208 |
|---|---|
| author | Khatereh Darvish ghanbar Tohid Yousefi Rezaii Ali Farzamnia Ismail Saad |
| author_facet | Khatereh Darvish ghanbar Tohid Yousefi Rezaii Ali Farzamnia Ismail Saad |
| author_sort | Khatereh Darvish ghanbar |
| collection | UMS |
| description | Common spatial pattern (CSP) is shown to be an effective pre-processing algorithm in order to discriminate different classes of motor-based EEG signals by obtaining suitable spatial filters. The performance of these filters can be improved by regularized CSP, in which available prior information is added in terms of regularization terms into the objective function of conventional CSP. Variety of prior information can be used in this way. In this paper, we used time correlation between different classes of EEG signal as the prior information, which is clarified similarity between different classes of signal for regularizing CSP. Furthermore, the proposed objective function can be easily extended to more than two-class problems. We used three different standard datasets to evaluate the performance of the proposed method. Correlation-based CSP (CCSP) outperformed original CSP as well as the existing regularized CSP, Principle Component Analysis (PCA) and Fisher Discriminate Analysis (FDA) in both two-class and multi-class scenarios. The simulation results showed that the proposed method outperformed conventional CSP by 6.9% in 2-class and 2.23% in multi-class problem in term of mean classification accuracy. |
| first_indexed | 2024-03-06T03:16:42Z |
| format | Article |
| id | ums.eprints-32837 |
| institution | Universiti Malaysia Sabah |
| language | English English |
| last_indexed | 2024-03-06T03:16:42Z |
| publishDate | 2021 |
| publisher | Public Library Science |
| record_format | dspace |
| spelling | ums.eprints-328372022-06-16T08:22:32Z https://eprints.ums.edu.my/id/eprint/32837/ Correlation-based common spatial pattern (CCSP): A novel extension of CSP for classification of motor imagery signal Khatereh Darvish ghanbar Tohid Yousefi Rezaii Ali Farzamnia Ismail Saad QA75.5-76.95 Electronic computers. Computer science Common spatial pattern (CSP) is shown to be an effective pre-processing algorithm in order to discriminate different classes of motor-based EEG signals by obtaining suitable spatial filters. The performance of these filters can be improved by regularized CSP, in which available prior information is added in terms of regularization terms into the objective function of conventional CSP. Variety of prior information can be used in this way. In this paper, we used time correlation between different classes of EEG signal as the prior information, which is clarified similarity between different classes of signal for regularizing CSP. Furthermore, the proposed objective function can be easily extended to more than two-class problems. We used three different standard datasets to evaluate the performance of the proposed method. Correlation-based CSP (CCSP) outperformed original CSP as well as the existing regularized CSP, Principle Component Analysis (PCA) and Fisher Discriminate Analysis (FDA) in both two-class and multi-class scenarios. The simulation results showed that the proposed method outperformed conventional CSP by 6.9% in 2-class and 2.23% in multi-class problem in term of mean classification accuracy. Public Library Science 2021 Article PeerReviewed text en https://eprints.ums.edu.my/id/eprint/32837/1/Correlation-based%20common%20spatial%20pattern%20.pdf text en https://eprints.ums.edu.my/id/eprint/32837/2/Correlation-based%20common%20spatial%20pattern%201.pdf Khatereh Darvish ghanbar and Tohid Yousefi Rezaii and Ali Farzamnia and Ismail Saad (2021) Correlation-based common spatial pattern (CCSP): A novel extension of CSP for classification of motor imagery signal. PLoS ONE, 16. pp. 1-18. ISSN 1932-6203 https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0248511 https://doi.org/10.1371/journal.pone.0248511 https://doi.org/10.1371/journal.pone.0248511 |
| spellingShingle | QA75.5-76.95 Electronic computers. Computer science Khatereh Darvish ghanbar Tohid Yousefi Rezaii Ali Farzamnia Ismail Saad Correlation-based common spatial pattern (CCSP): A novel extension of CSP for classification of motor imagery signal |
| title | Correlation-based common spatial pattern (CCSP): A novel extension of CSP for classification of motor imagery signal |
| title_full | Correlation-based common spatial pattern (CCSP): A novel extension of CSP for classification of motor imagery signal |
| title_fullStr | Correlation-based common spatial pattern (CCSP): A novel extension of CSP for classification of motor imagery signal |
| title_full_unstemmed | Correlation-based common spatial pattern (CCSP): A novel extension of CSP for classification of motor imagery signal |
| title_short | Correlation-based common spatial pattern (CCSP): A novel extension of CSP for classification of motor imagery signal |
| title_sort | correlation based common spatial pattern ccsp a novel extension of csp for classification of motor imagery signal |
| topic | QA75.5-76.95 Electronic computers. Computer science |
| url | https://eprints.ums.edu.my/id/eprint/32837/1/Correlation-based%20common%20spatial%20pattern%20.pdf https://eprints.ums.edu.my/id/eprint/32837/2/Correlation-based%20common%20spatial%20pattern%201.pdf |
| work_keys_str_mv | AT khaterehdarvishghanbar correlationbasedcommonspatialpatternccspanovelextensionofcspforclassificationofmotorimagerysignal AT tohidyousefirezaii correlationbasedcommonspatialpatternccspanovelextensionofcspforclassificationofmotorimagerysignal AT alifarzamnia correlationbasedcommonspatialpatternccspanovelextensionofcspforclassificationofmotorimagerysignal AT ismailsaad correlationbasedcommonspatialpatternccspanovelextensionofcspforclassificationofmotorimagerysignal |