MSEI-ENet: A Multi-Scale EEG-Inception Integrated Encoder Network for Motor Imagery EEG Decoding
Background: Due to complex signal characteristics and distinct individual differences, the decoding of a motor imagery electroencephalogram (MI-EEG) is limited by the unsatisfactory performance of suboptimal traditional models. Methods: A subject-independent model named MSEI-ENet is proposed for mul...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-01-01
|
| Series: | Brain Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3425/15/2/129 |
| _version_ | 1826582946352463872 |
|---|---|
| author | Pengcheng Wu Keling Fei Baohong Chen Lizheng Pan |
| author_facet | Pengcheng Wu Keling Fei Baohong Chen Lizheng Pan |
| author_sort | Pengcheng Wu |
| collection | DOAJ |
| description | Background: Due to complex signal characteristics and distinct individual differences, the decoding of a motor imagery electroencephalogram (MI-EEG) is limited by the unsatisfactory performance of suboptimal traditional models. Methods: A subject-independent model named MSEI-ENet is proposed for multiple-task MI-EEG decoding. It employs a specially designed multi-scale structure EEG-inception module (MSEI) for comprehensive feature learning. The encoder module further helps to detect discriminative information by its multi-head self-attention layer with a larger receptive field, which enhances feature representation and improves recognition efficacy. Results: The experimental results on Competition IV dataset 2a showed that our proposed model yielded an overall accuracy of 94.30%, MF1 score of 94.31%, and Kappa of 0.92. Conclusions: A performance comparison with state-of-the-art methods demonstrated the effectiveness and generalizability of the proposed model on challenging multi-task MI-EEG decoding. |
| first_indexed | 2025-03-14T15:14:02Z |
| format | Article |
| id | doaj.art-ced498844c7a4c4babe903c211089bd9 |
| institution | Directory Open Access Journal |
| issn | 2076-3425 |
| language | English |
| last_indexed | 2025-03-14T15:14:02Z |
| publishDate | 2025-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Brain Sciences |
| spelling | doaj.art-ced498844c7a4c4babe903c211089bd92025-02-25T13:16:42ZengMDPI AGBrain Sciences2076-34252025-01-0115212910.3390/brainsci15020129MSEI-ENet: A Multi-Scale EEG-Inception Integrated Encoder Network for Motor Imagery EEG DecodingPengcheng Wu0Keling Fei1Baohong Chen2Lizheng Pan3School of Mechanical Engineering and Rail Transit, Changzhou University, Changzhou 213164, ChinaSchool of Mechanical Engineering and Rail Transit, Changzhou University, Changzhou 213164, ChinaSchool of Mechanical Engineering and Rail Transit, Changzhou University, Changzhou 213164, ChinaSchool of Mechanical Engineering and Rail Transit, Changzhou University, Changzhou 213164, ChinaBackground: Due to complex signal characteristics and distinct individual differences, the decoding of a motor imagery electroencephalogram (MI-EEG) is limited by the unsatisfactory performance of suboptimal traditional models. Methods: A subject-independent model named MSEI-ENet is proposed for multiple-task MI-EEG decoding. It employs a specially designed multi-scale structure EEG-inception module (MSEI) for comprehensive feature learning. The encoder module further helps to detect discriminative information by its multi-head self-attention layer with a larger receptive field, which enhances feature representation and improves recognition efficacy. Results: The experimental results on Competition IV dataset 2a showed that our proposed model yielded an overall accuracy of 94.30%, MF1 score of 94.31%, and Kappa of 0.92. Conclusions: A performance comparison with state-of-the-art methods demonstrated the effectiveness and generalizability of the proposed model on challenging multi-task MI-EEG decoding.https://www.mdpi.com/2076-3425/15/2/129multi-scale structureinceptiontransformermotor imagerybrain–computer interface |
| spellingShingle | Pengcheng Wu Keling Fei Baohong Chen Lizheng Pan MSEI-ENet: A Multi-Scale EEG-Inception Integrated Encoder Network for Motor Imagery EEG Decoding Brain Sciences multi-scale structure inception transformer motor imagery brain–computer interface |
| title | MSEI-ENet: A Multi-Scale EEG-Inception Integrated Encoder Network for Motor Imagery EEG Decoding |
| title_full | MSEI-ENet: A Multi-Scale EEG-Inception Integrated Encoder Network for Motor Imagery EEG Decoding |
| title_fullStr | MSEI-ENet: A Multi-Scale EEG-Inception Integrated Encoder Network for Motor Imagery EEG Decoding |
| title_full_unstemmed | MSEI-ENet: A Multi-Scale EEG-Inception Integrated Encoder Network for Motor Imagery EEG Decoding |
| title_short | MSEI-ENet: A Multi-Scale EEG-Inception Integrated Encoder Network for Motor Imagery EEG Decoding |
| title_sort | msei enet a multi scale eeg inception integrated encoder network for motor imagery eeg decoding |
| topic | multi-scale structure inception transformer motor imagery brain–computer interface |
| url | https://www.mdpi.com/2076-3425/15/2/129 |
| work_keys_str_mv | AT pengchengwu mseienetamultiscaleeeginceptionintegratedencodernetworkformotorimageryeegdecoding AT kelingfei mseienetamultiscaleeeginceptionintegratedencodernetworkformotorimageryeegdecoding AT baohongchen mseienetamultiscaleeeginceptionintegratedencodernetworkformotorimageryeegdecoding AT lizhengpan mseienetamultiscaleeeginceptionintegratedencodernetworkformotorimageryeegdecoding |