EEG-Based Classification of Spoken Words Using Machine Learning Approaches
Amyotrophic lateral sclerosis (ALS) is a neurodegenerative disease that affects the nerve cells in the brain and spinal cord. This condition leads to the loss of motor skills and, in many cases, the inability to speak. Decoding spoken words from electroencephalography (EEG) signals emerges as an ess...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-11-01
|
| Series: | Computation |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-3197/11/11/225 |
| _version_ | 1827640191281528832 |
|---|---|
| author | Denise Alonso-Vázquez Omar Mendoza-Montoya Ricardo Caraza Hector R. Martinez Javier M. Antelis |
| author_facet | Denise Alonso-Vázquez Omar Mendoza-Montoya Ricardo Caraza Hector R. Martinez Javier M. Antelis |
| author_sort | Denise Alonso-Vázquez |
| collection | DOAJ |
| description | Amyotrophic lateral sclerosis (ALS) is a neurodegenerative disease that affects the nerve cells in the brain and spinal cord. This condition leads to the loss of motor skills and, in many cases, the inability to speak. Decoding spoken words from electroencephalography (EEG) signals emerges as an essential tool to enhance the quality of life for these patients. This study compares two classification techniques: (1) the extraction of spectral power features across various frequency bands combined with support vector machines (PSD + SVM) and (2) EEGNet, a convolutional neural network specifically designed for EEG-based brain–computer interfaces. An EEG dataset was acquired from 32 electrodes in 28 healthy participants pronouncing five words in Spanish. Average accuracy rates of 91.04 ± 5.82% for <i>Attention</i> vs. <i>Pronunciation</i>, 73.91 ± 10.04% for <i>Short words</i> vs. <i>Long words</i>, 81.23 ± 10.47% for <i>Word</i> vs. <i>Word</i>, and 54.87 ± 14.51% in the multiclass scenario (<i>All words</i>) were achieved. EEGNet outperformed the PSD + SVM method in three of the four classification scenarios. These findings demonstrate the potential of EEGNet for decoding words from EEG signals, laying the groundwork for future research in ALS patients using non-invasive methods. |
| first_indexed | 2024-03-09T16:54:30Z |
| format | Article |
| id | doaj.art-27014fd9e5504952bdc0cfaa1b2e7ba1 |
| institution | Directory Open Access Journal |
| issn | 2079-3197 |
| language | English |
| last_indexed | 2024-03-09T16:54:30Z |
| publishDate | 2023-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Computation |
| spelling | doaj.art-27014fd9e5504952bdc0cfaa1b2e7ba12023-11-24T14:36:25ZengMDPI AGComputation2079-31972023-11-01111122510.3390/computation11110225EEG-Based Classification of Spoken Words Using Machine Learning ApproachesDenise Alonso-Vázquez0Omar Mendoza-Montoya1Ricardo Caraza2Hector R. Martinez3Javier M. Antelis4Tecnologico de Monterrey, Escuela de Ingeniería y Ciencias, Monterrey 64849, MexicoTecnologico de Monterrey, Escuela de Ingeniería y Ciencias, Monterrey 64849, MexicoTecnologico de Monterrey, Escuela de Medicina y Ciencias de la Salud, Monterrey 64849, MexicoTecnologico de Monterrey, Escuela de Medicina y Ciencias de la Salud, Monterrey 64849, MexicoTecnologico de Monterrey, Escuela de Ingeniería y Ciencias, Monterrey 64849, MexicoAmyotrophic lateral sclerosis (ALS) is a neurodegenerative disease that affects the nerve cells in the brain and spinal cord. This condition leads to the loss of motor skills and, in many cases, the inability to speak. Decoding spoken words from electroencephalography (EEG) signals emerges as an essential tool to enhance the quality of life for these patients. This study compares two classification techniques: (1) the extraction of spectral power features across various frequency bands combined with support vector machines (PSD + SVM) and (2) EEGNet, a convolutional neural network specifically designed for EEG-based brain–computer interfaces. An EEG dataset was acquired from 32 electrodes in 28 healthy participants pronouncing five words in Spanish. Average accuracy rates of 91.04 ± 5.82% for <i>Attention</i> vs. <i>Pronunciation</i>, 73.91 ± 10.04% for <i>Short words</i> vs. <i>Long words</i>, 81.23 ± 10.47% for <i>Word</i> vs. <i>Word</i>, and 54.87 ± 14.51% in the multiclass scenario (<i>All words</i>) were achieved. EEGNet outperformed the PSD + SVM method in three of the four classification scenarios. These findings demonstrate the potential of EEGNet for decoding words from EEG signals, laying the groundwork for future research in ALS patients using non-invasive methods.https://www.mdpi.com/2079-3197/11/11/225electroencephalographyspeech decodingEEGNet |
| spellingShingle | Denise Alonso-Vázquez Omar Mendoza-Montoya Ricardo Caraza Hector R. Martinez Javier M. Antelis EEG-Based Classification of Spoken Words Using Machine Learning Approaches Computation electroencephalography speech decoding EEGNet |
| title | EEG-Based Classification of Spoken Words Using Machine Learning Approaches |
| title_full | EEG-Based Classification of Spoken Words Using Machine Learning Approaches |
| title_fullStr | EEG-Based Classification of Spoken Words Using Machine Learning Approaches |
| title_full_unstemmed | EEG-Based Classification of Spoken Words Using Machine Learning Approaches |
| title_short | EEG-Based Classification of Spoken Words Using Machine Learning Approaches |
| title_sort | eeg based classification of spoken words using machine learning approaches |
| topic | electroencephalography speech decoding EEGNet |
| url | https://www.mdpi.com/2079-3197/11/11/225 |
| work_keys_str_mv | AT denisealonsovazquez eegbasedclassificationofspokenwordsusingmachinelearningapproaches AT omarmendozamontoya eegbasedclassificationofspokenwordsusingmachinelearningapproaches AT ricardocaraza eegbasedclassificationofspokenwordsusingmachinelearningapproaches AT hectorrmartinez eegbasedclassificationofspokenwordsusingmachinelearningapproaches AT javiermantelis eegbasedclassificationofspokenwordsusingmachinelearningapproaches |