Decoding Imagined Speech From EEG Using Transfer Learning
We present a transfer learning-based approach for decoding imagined speech from electroencephalogram (EEG). Features are extracted simultaneously from multiple EEG channels, rather than separately from individual channels. This helps in capturing the interrelationships between the cortical regions....
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2021-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9551930/ |
| _version_ | 1818834833728602112 |
|---|---|
| author | Jerrin Thomas Panachakel Ramakrishnan Angarai Ganesan |
| author_facet | Jerrin Thomas Panachakel Ramakrishnan Angarai Ganesan |
| author_sort | Jerrin Thomas Panachakel |
| collection | DOAJ |
| description | We present a transfer learning-based approach for decoding imagined speech from electroencephalogram (EEG). Features are extracted simultaneously from multiple EEG channels, rather than separately from individual channels. This helps in capturing the interrelationships between the cortical regions. To alleviate the problem of lack of enough data for training deep networks, sliding window-based data augmentation is performed. Mean phase coherence and magnitude-squared coherence, two popular measures used in EEG connectivity analysis, are used as features. These features are compactly arranged, exploiting their symmetry, to obtain a three dimensional “image-like” representation. The three dimensions of this matrix correspond to the alpha, beta and gamma EEG frequency bands. A deep network with ResNet50 as the base model is used for classifying the imagined prompts. The proposed method is tested on the publicly available ASU dataset of imagined speech EEG, comprising four different types of prompts. The accuracy of decoding the imagined prompt varies from a minimum of 79.7% for vowels to a maximum of 95.5% for short-long words across the various subjects. The accuracies obtained are better than the state-of-the-art methods, and the technique is good in decoding prompts of different complexities. |
| first_indexed | 2024-12-19T02:41:06Z |
| format | Article |
| id | doaj.art-40ec06f8977c48d9a54d0b649c0701d7 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-19T02:41:06Z |
| publishDate | 2021-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-40ec06f8977c48d9a54d0b649c0701d72022-12-21T20:39:07ZengIEEEIEEE Access2169-35362021-01-01913537113538310.1109/ACCESS.2021.31161969551930Decoding Imagined Speech From EEG Using Transfer LearningJerrin Thomas Panachakel0https://orcid.org/0000-0001-5940-9423Ramakrishnan Angarai Ganesan1https://orcid.org/0000-0002-3646-1955Department of Electrical Engineering, Indian Institute of Science, Bengaluru, IndiaDepartment of Electrical Engineering, Indian Institute of Science, Bengaluru, IndiaWe present a transfer learning-based approach for decoding imagined speech from electroencephalogram (EEG). Features are extracted simultaneously from multiple EEG channels, rather than separately from individual channels. This helps in capturing the interrelationships between the cortical regions. To alleviate the problem of lack of enough data for training deep networks, sliding window-based data augmentation is performed. Mean phase coherence and magnitude-squared coherence, two popular measures used in EEG connectivity analysis, are used as features. These features are compactly arranged, exploiting their symmetry, to obtain a three dimensional “image-like” representation. The three dimensions of this matrix correspond to the alpha, beta and gamma EEG frequency bands. A deep network with ResNet50 as the base model is used for classifying the imagined prompts. The proposed method is tested on the publicly available ASU dataset of imagined speech EEG, comprising four different types of prompts. The accuracy of decoding the imagined prompt varies from a minimum of 79.7% for vowels to a maximum of 95.5% for short-long words across the various subjects. The accuracies obtained are better than the state-of-the-art methods, and the technique is good in decoding prompts of different complexities.https://ieeexplore.ieee.org/document/9551930/Brain–computer interfacetransfer learningelectroencephalogramspeech imageryimagined speech |
| spellingShingle | Jerrin Thomas Panachakel Ramakrishnan Angarai Ganesan Decoding Imagined Speech From EEG Using Transfer Learning IEEE Access Brain–computer interface transfer learning electroencephalogram speech imagery imagined speech |
| title | Decoding Imagined Speech From EEG Using Transfer Learning |
| title_full | Decoding Imagined Speech From EEG Using Transfer Learning |
| title_fullStr | Decoding Imagined Speech From EEG Using Transfer Learning |
| title_full_unstemmed | Decoding Imagined Speech From EEG Using Transfer Learning |
| title_short | Decoding Imagined Speech From EEG Using Transfer Learning |
| title_sort | decoding imagined speech from eeg using transfer learning |
| topic | Brain–computer interface transfer learning electroencephalogram speech imagery imagined speech |
| url | https://ieeexplore.ieee.org/document/9551930/ |
| work_keys_str_mv | AT jerrinthomaspanachakel decodingimaginedspeechfromeegusingtransferlearning AT ramakrishnanangaraiganesan decodingimaginedspeechfromeegusingtransferlearning |