A Densely Connected Multi-Branch 3D Convolutional Neural Network for Motor Imagery EEG Decoding
Motor imagery (MI) is a classical method of brain–computer interaction (BCI), in which electroencephalogram (EEG) signal features evoked by imaginary body movements are recognized, and relevant information is extracted. Recently, various deep-learning methods are being focused on in finding an easy-...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-02-01
|
| Series: | Brain Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3425/11/2/197 |
| _version_ | 1797414381143719936 |
|---|---|
| author | Tianjun Liu Deling Yang |
| author_facet | Tianjun Liu Deling Yang |
| author_sort | Tianjun Liu |
| collection | DOAJ |
| description | Motor imagery (MI) is a classical method of brain–computer interaction (BCI), in which electroencephalogram (EEG) signal features evoked by imaginary body movements are recognized, and relevant information is extracted. Recently, various deep-learning methods are being focused on in finding an easy-to-use EEG representation method that can preserve both temporal information and spatial information. To further utilize the spatial and temporal features of EEG signals, an improved 3D representation of the EEG and a densely connected multi-branch 3D convolutional neural network (dense M3D CNN) for MI classification are introduced in this paper. Specifically, as compared to the original 3D representation, a new padding method is proposed to pad the points without electrodes with the mean of all the EEG signals. Based on this new 3D presentation, a densely connected multi-branch 3D CNN with a novel dense connectivity is proposed for extracting the EEG signal features. Experiments were carried out on the WAY-EEG-GAL and BCI competition IV 2a datasets to verify the performance of this proposed method. The experimental results show that the proposed framework achieves a state-of-the-art performance that significantly outperforms the multi-branch 3D CNN framework, with a 6.208% improvement in the average accuracy for the BCI competition IV 2a datasets and 6.281% improvement in the average accuracy for the WAY-EEG-GAL datasets, with a smaller standard deviation. The results also prove the effectiveness and robustness of the method, along with validating its use in MI-classification tasks. |
| first_indexed | 2024-03-09T05:32:15Z |
| format | Article |
| id | doaj.art-602899bfd44041ecbe5ade6a7cd14244 |
| institution | Directory Open Access Journal |
| issn | 2076-3425 |
| language | English |
| last_indexed | 2024-03-09T05:32:15Z |
| publishDate | 2021-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Brain Sciences |
| spelling | doaj.art-602899bfd44041ecbe5ade6a7cd142442023-12-03T12:31:14ZengMDPI AGBrain Sciences2076-34252021-02-0111219710.3390/brainsci11020197A Densely Connected Multi-Branch 3D Convolutional Neural Network for Motor Imagery EEG DecodingTianjun Liu0Deling Yang1College of Engineering and Technology, Northeast Forestry University, Harbin 150040, ChinaCollege of Engineering and Technology, Northeast Forestry University, Harbin 150040, ChinaMotor imagery (MI) is a classical method of brain–computer interaction (BCI), in which electroencephalogram (EEG) signal features evoked by imaginary body movements are recognized, and relevant information is extracted. Recently, various deep-learning methods are being focused on in finding an easy-to-use EEG representation method that can preserve both temporal information and spatial information. To further utilize the spatial and temporal features of EEG signals, an improved 3D representation of the EEG and a densely connected multi-branch 3D convolutional neural network (dense M3D CNN) for MI classification are introduced in this paper. Specifically, as compared to the original 3D representation, a new padding method is proposed to pad the points without electrodes with the mean of all the EEG signals. Based on this new 3D presentation, a densely connected multi-branch 3D CNN with a novel dense connectivity is proposed for extracting the EEG signal features. Experiments were carried out on the WAY-EEG-GAL and BCI competition IV 2a datasets to verify the performance of this proposed method. The experimental results show that the proposed framework achieves a state-of-the-art performance that significantly outperforms the multi-branch 3D CNN framework, with a 6.208% improvement in the average accuracy for the BCI competition IV 2a datasets and 6.281% improvement in the average accuracy for the WAY-EEG-GAL datasets, with a smaller standard deviation. The results also prove the effectiveness and robustness of the method, along with validating its use in MI-classification tasks.https://www.mdpi.com/2076-3425/11/2/197motor imagery (MI)electroencephalogram (EEG)dense connectivity3D convolutional neural network (3D CNN) |
| spellingShingle | Tianjun Liu Deling Yang A Densely Connected Multi-Branch 3D Convolutional Neural Network for Motor Imagery EEG Decoding Brain Sciences motor imagery (MI) electroencephalogram (EEG) dense connectivity 3D convolutional neural network (3D CNN) |
| title | A Densely Connected Multi-Branch 3D Convolutional Neural Network for Motor Imagery EEG Decoding |
| title_full | A Densely Connected Multi-Branch 3D Convolutional Neural Network for Motor Imagery EEG Decoding |
| title_fullStr | A Densely Connected Multi-Branch 3D Convolutional Neural Network for Motor Imagery EEG Decoding |
| title_full_unstemmed | A Densely Connected Multi-Branch 3D Convolutional Neural Network for Motor Imagery EEG Decoding |
| title_short | A Densely Connected Multi-Branch 3D Convolutional Neural Network for Motor Imagery EEG Decoding |
| title_sort | densely connected multi branch 3d convolutional neural network for motor imagery eeg decoding |
| topic | motor imagery (MI) electroencephalogram (EEG) dense connectivity 3D convolutional neural network (3D CNN) |
| url | https://www.mdpi.com/2076-3425/11/2/197 |
| work_keys_str_mv | AT tianjunliu adenselyconnectedmultibranch3dconvolutionalneuralnetworkformotorimageryeegdecoding AT delingyang adenselyconnectedmultibranch3dconvolutionalneuralnetworkformotorimageryeegdecoding AT tianjunliu denselyconnectedmultibranch3dconvolutionalneuralnetworkformotorimageryeegdecoding AT delingyang denselyconnectedmultibranch3dconvolutionalneuralnetworkformotorimageryeegdecoding |