Automated Classification and Removal of EEG Artifacts With SVM and Wavelet-ICA
Brain electrical activity recordings by electroencephalography (EEG) are often contaminated with signal artifacts. Procedures for automated removal of EEG artifacts are frequently sought for clinical diagnostics and brain-computer interface applications. In recent years, a combination of independent...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Published: |
Institute of Electrical and Electronics Engineers
2018
|
| Subjects: |
| _version_ | 1825721571362209792 |
|---|---|
| author | Chong, Yeh Sai Mokhtar, Norrima Arof, Hamzah Cumming, Paul Iwahashi, Masahiro |
| author_facet | Chong, Yeh Sai Mokhtar, Norrima Arof, Hamzah Cumming, Paul Iwahashi, Masahiro |
| author_sort | Chong, Yeh Sai |
| collection | UM |
| description | Brain electrical activity recordings by electroencephalography (EEG) are often contaminated with signal artifacts. Procedures for automated removal of EEG artifacts are frequently sought for clinical diagnostics and brain-computer interface applications. In recent years, a combination of independent component analysis (ICA) and discrete wavelet transform has been introduced as standard technique for EEG artifact removal. However, in performing the wavelet-ICA procedure, visual inspection or arbitrary thresholding may be required for identifying artifactual components in the EEG signal. We now propose a novel approach for identifying artifactual components separated by wavelet-ICA using a pretrained support vector machine (SVM). Our method presents a robust and extendable system that enables fully automated identification and removal of artifacts from EEG signals, without applying any arbitrary thresholding. Using test data contaminated by eye blink artifacts, we show that our method performed better in identifying artifactual components than did existing thresholding methods. Furthermore, wavelet-ICA in conjunction with SVM successfully removed target artifacts, while largely retaining the EEG source signals of interest. We propose a set of features including kurtosis, variance, Shannon's entropy, and range of amplitude as training and test data of SVM to identify eye blink artifacts in EEG signals. This combinatorial method is also extendable to accommodate multiple types of artifacts present in multichannel EEG. We envision future research to explore other descriptive features corresponding to other types of artifactual components. |
| first_indexed | 2024-03-06T05:52:22Z |
| format | Article |
| id | um.eprints-20815 |
| institution | Universiti Malaya |
| last_indexed | 2024-03-06T05:52:22Z |
| publishDate | 2018 |
| publisher | Institute of Electrical and Electronics Engineers |
| record_format | dspace |
| spelling | um.eprints-208152019-04-05T08:35:48Z http://eprints.um.edu.my/20815/ Automated Classification and Removal of EEG Artifacts With SVM and Wavelet-ICA Chong, Yeh Sai Mokhtar, Norrima Arof, Hamzah Cumming, Paul Iwahashi, Masahiro TK Electrical engineering. Electronics Nuclear engineering Brain electrical activity recordings by electroencephalography (EEG) are often contaminated with signal artifacts. Procedures for automated removal of EEG artifacts are frequently sought for clinical diagnostics and brain-computer interface applications. In recent years, a combination of independent component analysis (ICA) and discrete wavelet transform has been introduced as standard technique for EEG artifact removal. However, in performing the wavelet-ICA procedure, visual inspection or arbitrary thresholding may be required for identifying artifactual components in the EEG signal. We now propose a novel approach for identifying artifactual components separated by wavelet-ICA using a pretrained support vector machine (SVM). Our method presents a robust and extendable system that enables fully automated identification and removal of artifacts from EEG signals, without applying any arbitrary thresholding. Using test data contaminated by eye blink artifacts, we show that our method performed better in identifying artifactual components than did existing thresholding methods. Furthermore, wavelet-ICA in conjunction with SVM successfully removed target artifacts, while largely retaining the EEG source signals of interest. We propose a set of features including kurtosis, variance, Shannon's entropy, and range of amplitude as training and test data of SVM to identify eye blink artifacts in EEG signals. This combinatorial method is also extendable to accommodate multiple types of artifacts present in multichannel EEG. We envision future research to explore other descriptive features corresponding to other types of artifactual components. Institute of Electrical and Electronics Engineers 2018 Article PeerReviewed Chong, Yeh Sai and Mokhtar, Norrima and Arof, Hamzah and Cumming, Paul and Iwahashi, Masahiro (2018) Automated Classification and Removal of EEG Artifacts With SVM and Wavelet-ICA. IEEE Journal of Biomedical and Health Informatics, 22 (3). pp. 664-670. ISSN 2168-2194, DOI https://doi.org/10.1109/JBHI.2017.2723420 <https://doi.org/10.1109/JBHI.2017.2723420>. https://doi.org/10.1109/JBHI.2017.2723420 doi:10.1109/JBHI.2017.2723420 |
| spellingShingle | TK Electrical engineering. Electronics Nuclear engineering Chong, Yeh Sai Mokhtar, Norrima Arof, Hamzah Cumming, Paul Iwahashi, Masahiro Automated Classification and Removal of EEG Artifacts With SVM and Wavelet-ICA |
| title | Automated Classification and Removal of EEG Artifacts With SVM and Wavelet-ICA |
| title_full | Automated Classification and Removal of EEG Artifacts With SVM and Wavelet-ICA |
| title_fullStr | Automated Classification and Removal of EEG Artifacts With SVM and Wavelet-ICA |
| title_full_unstemmed | Automated Classification and Removal of EEG Artifacts With SVM and Wavelet-ICA |
| title_short | Automated Classification and Removal of EEG Artifacts With SVM and Wavelet-ICA |
| title_sort | automated classification and removal of eeg artifacts with svm and wavelet ica |
| topic | TK Electrical engineering. Electronics Nuclear engineering |
| work_keys_str_mv | AT chongyehsai automatedclassificationandremovalofeegartifactswithsvmandwaveletica AT mokhtarnorrima automatedclassificationandremovalofeegartifactswithsvmandwaveletica AT arofhamzah automatedclassificationandremovalofeegartifactswithsvmandwaveletica AT cummingpaul automatedclassificationandremovalofeegartifactswithsvmandwaveletica AT iwahashimasahiro automatedclassificationandremovalofeegartifactswithsvmandwaveletica |