Feature weighting and regularization of common spatial patterns in EEG-based motor imagery BCI
Electroencephalography signals have very low spatial resolution and electrodes capture signals that are overlapping each other. To extract the discriminative features and alleviate overfitting problem for motor imagery brain-computer interface (BCI), spatial filtering is widely applied but often onl...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Journal Article |
| Language: | English |
| Published: |
2020
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/142567 |
| _version_ | 1811693764009263104 |
|---|---|
| author | Mishuhina, Vasilisa Jiang, Xudong |
| author2 | School of Electrical and Electronic Engineering |
| author_facet | School of Electrical and Electronic Engineering Mishuhina, Vasilisa Jiang, Xudong |
| author_sort | Mishuhina, Vasilisa |
| collection | NTU |
| description | Electroencephalography signals have very low spatial resolution and electrodes capture signals that are overlapping each other. To extract the discriminative features and alleviate overfitting problem for motor imagery brain-computer interface (BCI), spatial filtering is widely applied but often only very few common spatial patterns (CSP) are selected as features while ignoring all others. However, using only few CSP features, though alleviates overfitting problem, loses the discriminating information, which limits the BCI performance. This letter proposes a novel feature weighting and regularization (FWR) method that utilizes all CSP features to avoid information loss. The proposed method can be applied in all CSP-based approaches. Experiments of this letter show the effect of the proposed method applied in the standard CSP and its two extensions, common spatio-spectral patterns and regularized CSP. Results on BCI Competition III Dataset IIIa and IV Dataset IIa demonstrate that the proposed FWR method enhances the classification accuracy comparing to the conventional feature selection approaches. |
| first_indexed | 2024-10-01T06:56:51Z |
| format | Journal Article |
| id | ntu-10356/142567 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T06:56:51Z |
| publishDate | 2020 |
| record_format | dspace |
| spelling | ntu-10356/1425672020-06-24T07:16:40Z Feature weighting and regularization of common spatial patterns in EEG-based motor imagery BCI Mishuhina, Vasilisa Jiang, Xudong School of Electrical and Electronic Engineering Engineering::Electrical and electronic engineering Brain-computer Interface (BCI) Common Spatial Patterns (CSP) Electroencephalography signals have very low spatial resolution and electrodes capture signals that are overlapping each other. To extract the discriminative features and alleviate overfitting problem for motor imagery brain-computer interface (BCI), spatial filtering is widely applied but often only very few common spatial patterns (CSP) are selected as features while ignoring all others. However, using only few CSP features, though alleviates overfitting problem, loses the discriminating information, which limits the BCI performance. This letter proposes a novel feature weighting and regularization (FWR) method that utilizes all CSP features to avoid information loss. The proposed method can be applied in all CSP-based approaches. Experiments of this letter show the effect of the proposed method applied in the standard CSP and its two extensions, common spatio-spectral patterns and regularized CSP. Results on BCI Competition III Dataset IIIa and IV Dataset IIa demonstrate that the proposed FWR method enhances the classification accuracy comparing to the conventional feature selection approaches. 2020-06-24T07:16:39Z 2020-06-24T07:16:39Z 2018 Journal Article Mishuhina, V., & Jiang, X. (2018). Feature weighting and regularization of common spatial patterns in EEG-based motor imagery BCI. IEEE Signal Processing Letters, 25(6), 783-787. doi:10.1109/LSP.2018.2823683 1070-9908 https://hdl.handle.net/10356/142567 10.1109/LSP.2018.2823683 2-s2.0-85045223581 6 25 783 787 en IEEE Signal Processing Letters © 2018 IEEE. All rights reserved. |
| spellingShingle | Engineering::Electrical and electronic engineering Brain-computer Interface (BCI) Common Spatial Patterns (CSP) Mishuhina, Vasilisa Jiang, Xudong Feature weighting and regularization of common spatial patterns in EEG-based motor imagery BCI |
| title | Feature weighting and regularization of common spatial patterns in EEG-based motor imagery BCI |
| title_full | Feature weighting and regularization of common spatial patterns in EEG-based motor imagery BCI |
| title_fullStr | Feature weighting and regularization of common spatial patterns in EEG-based motor imagery BCI |
| title_full_unstemmed | Feature weighting and regularization of common spatial patterns in EEG-based motor imagery BCI |
| title_short | Feature weighting and regularization of common spatial patterns in EEG-based motor imagery BCI |
| title_sort | feature weighting and regularization of common spatial patterns in eeg based motor imagery bci |
| topic | Engineering::Electrical and electronic engineering Brain-computer Interface (BCI) Common Spatial Patterns (CSP) |
| url | https://hdl.handle.net/10356/142567 |
| work_keys_str_mv | AT mishuhinavasilisa featureweightingandregularizationofcommonspatialpatternsineegbasedmotorimagerybci AT jiangxudong featureweightingandregularizationofcommonspatialpatternsineegbasedmotorimagerybci |