Multiclass classification of imagined speech EEG using noise-assisted multivariate empirical mode decomposition and multireceptive field convolutional neural network
IntroductionIn this study, we classified electroencephalography (EEG) data of imagined speech using signal decomposition and multireceptive convolutional neural network. The imagined speech EEG with five vowels /a/, /e/, /i/, /o/, and /u/, and mute (rest) sounds were obtained from ten study particip...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2023-08-01
|
| Series: | Frontiers in Human Neuroscience |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fnhum.2023.1186594/full |
| _version_ | 1797743158166028288 |
|---|---|
| author | Hyeong-jun Park Boreom Lee Boreom Lee |
| author_facet | Hyeong-jun Park Boreom Lee Boreom Lee |
| author_sort | Hyeong-jun Park |
| collection | DOAJ |
| description | IntroductionIn this study, we classified electroencephalography (EEG) data of imagined speech using signal decomposition and multireceptive convolutional neural network. The imagined speech EEG with five vowels /a/, /e/, /i/, /o/, and /u/, and mute (rest) sounds were obtained from ten study participants.Materials and methodsFirst, two different signal decomposition methods were applied for comparison: noise-assisted multivariate empirical mode decomposition and wavelet packet decomposition. Six statistical features were calculated from the decomposed eight sub-frequency bands EEG. Next, all features obtained from each channel of the trial were vectorized and used as the input vector of classifiers. Lastly, EEG was classified using multireceptive field convolutional neural network and several other classifiers for comparison.ResultsWe achieved an average classification rate of 73.09 and up to 80.41% in a multiclass (six classes) setup (Chance: 16.67%). In comparison with various other classifiers, significant improvements for other classifiers were achieved (p-value < 0.05). From the frequency sub-band analysis, high-frequency band regions and the lowest-frequency band region contain more information about imagined vowel EEG data. The misclassification and classification rate of each vowel imaginary EEG was analyzed through a confusion matrix.DiscussionImagined speech EEG can be classified successfully using the proposed signal decomposition method and a convolutional neural network. The proposed classification method for imagined speech EEG can contribute to developing a practical imagined speech-based brain-computer interfaces system. |
| first_indexed | 2024-03-12T14:51:29Z |
| format | Article |
| id | doaj.art-3388d149d3004efe918f099756eb4d97 |
| institution | Directory Open Access Journal |
| issn | 1662-5161 |
| language | English |
| last_indexed | 2024-03-12T14:51:29Z |
| publishDate | 2023-08-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Human Neuroscience |
| spelling | doaj.art-3388d149d3004efe918f099756eb4d972023-08-15T09:04:55ZengFrontiers Media S.A.Frontiers in Human Neuroscience1662-51612023-08-011710.3389/fnhum.2023.11865941186594Multiclass classification of imagined speech EEG using noise-assisted multivariate empirical mode decomposition and multireceptive field convolutional neural networkHyeong-jun Park0Boreom Lee1Boreom Lee2Department of Biomedical Science and Engineering, Gwangju Institute of Science and Technology, Gwangju, Republic of KoreaDepartment of Biomedical Science and Engineering, Gwangju Institute of Science and Technology, Gwangju, Republic of KoreaAI Graduate School, Gwangju Institute of Science and Technology, Gwangju, Republic of KoreaIntroductionIn this study, we classified electroencephalography (EEG) data of imagined speech using signal decomposition and multireceptive convolutional neural network. The imagined speech EEG with five vowels /a/, /e/, /i/, /o/, and /u/, and mute (rest) sounds were obtained from ten study participants.Materials and methodsFirst, two different signal decomposition methods were applied for comparison: noise-assisted multivariate empirical mode decomposition and wavelet packet decomposition. Six statistical features were calculated from the decomposed eight sub-frequency bands EEG. Next, all features obtained from each channel of the trial were vectorized and used as the input vector of classifiers. Lastly, EEG was classified using multireceptive field convolutional neural network and several other classifiers for comparison.ResultsWe achieved an average classification rate of 73.09 and up to 80.41% in a multiclass (six classes) setup (Chance: 16.67%). In comparison with various other classifiers, significant improvements for other classifiers were achieved (p-value < 0.05). From the frequency sub-band analysis, high-frequency band regions and the lowest-frequency band region contain more information about imagined vowel EEG data. The misclassification and classification rate of each vowel imaginary EEG was analyzed through a confusion matrix.DiscussionImagined speech EEG can be classified successfully using the proposed signal decomposition method and a convolutional neural network. The proposed classification method for imagined speech EEG can contribute to developing a practical imagined speech-based brain-computer interfaces system.https://www.frontiersin.org/articles/10.3389/fnhum.2023.1186594/fullbrain-computer interfacesimagined speech EEGmulticlass classificationmultireceptive field convolutional neural networknoise-assisted empirical mode decomposition |
| spellingShingle | Hyeong-jun Park Boreom Lee Boreom Lee Multiclass classification of imagined speech EEG using noise-assisted multivariate empirical mode decomposition and multireceptive field convolutional neural network Frontiers in Human Neuroscience brain-computer interfaces imagined speech EEG multiclass classification multireceptive field convolutional neural network noise-assisted empirical mode decomposition |
| title | Multiclass classification of imagined speech EEG using noise-assisted multivariate empirical mode decomposition and multireceptive field convolutional neural network |
| title_full | Multiclass classification of imagined speech EEG using noise-assisted multivariate empirical mode decomposition and multireceptive field convolutional neural network |
| title_fullStr | Multiclass classification of imagined speech EEG using noise-assisted multivariate empirical mode decomposition and multireceptive field convolutional neural network |
| title_full_unstemmed | Multiclass classification of imagined speech EEG using noise-assisted multivariate empirical mode decomposition and multireceptive field convolutional neural network |
| title_short | Multiclass classification of imagined speech EEG using noise-assisted multivariate empirical mode decomposition and multireceptive field convolutional neural network |
| title_sort | multiclass classification of imagined speech eeg using noise assisted multivariate empirical mode decomposition and multireceptive field convolutional neural network |
| topic | brain-computer interfaces imagined speech EEG multiclass classification multireceptive field convolutional neural network noise-assisted empirical mode decomposition |
| url | https://www.frontiersin.org/articles/10.3389/fnhum.2023.1186594/full |
| work_keys_str_mv | AT hyeongjunpark multiclassclassificationofimaginedspeecheegusingnoiseassistedmultivariateempiricalmodedecompositionandmultireceptivefieldconvolutionalneuralnetwork AT boreomlee multiclassclassificationofimaginedspeecheegusingnoiseassistedmultivariateempiricalmodedecompositionandmultireceptivefieldconvolutionalneuralnetwork AT boreomlee multiclassclassificationofimaginedspeecheegusingnoiseassistedmultivariateempiricalmodedecompositionandmultireceptivefieldconvolutionalneuralnetwork |