A multi-band centroid contrastive reconstruction fusion network for motor imagery electroencephalogram signal decoding
Motor imagery (MI) brain-computer interface (BCI) assist users in establishing direct communication between their brain and external devices by decoding the movement intention of human electroencephalogram (EEG) signals. However, cerebral cortical potentials are highly rhythmic and sub-band features...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
AIMS Press
2023-11-01
|
| Series: | Mathematical Biosciences and Engineering |
| Subjects: | |
| Online Access: | https://www.aimspress.com/article/doi/10.3934/mbe.2023912?viewType=HTML |
| _version_ | 1797397610935353344 |
|---|---|
| author | Jiacan Xu Donglin Li Peng Zhou Chunsheng Li Zinan Wang Shenghao Tong |
| author_facet | Jiacan Xu Donglin Li Peng Zhou Chunsheng Li Zinan Wang Shenghao Tong |
| author_sort | Jiacan Xu |
| collection | DOAJ |
| description | Motor imagery (MI) brain-computer interface (BCI) assist users in establishing direct communication between their brain and external devices by decoding the movement intention of human electroencephalogram (EEG) signals. However, cerebral cortical potentials are highly rhythmic and sub-band features, different experimental situations and subjects have different categories of semantic information in specific sample target spaces. Feature fusion can lead to more discriminative features, but simple fusion of features from different embedding spaces leading to the model global loss is not easily convergent and ignores the complementarity of features. Considering the similarity and category contribution of different sub-band features, we propose a multi-band centroid contrastive reconstruction fusion network (MB-CCRF). We obtain multi-band spatio-temporal features by frequency division, preserving the task-related rhythmic features of different EEG signals; use a multi-stream cross-layer connected convolutional network to perform a deep feature representation for each sub-band separately; propose a centroid contrastive reconstruction fusion module, which maps different sub-band and category features into the same shared embedding space by comparing with category prototypes, reconstructing the feature semantic structure to ensure that the global loss of the fused features converges more easily. Finally, we use a learning mechanism to model the similarity between channel features and use it as the weight of fused sub-band features, thus enhancing the more discriminative features, suppressing the useless features. The experimental accuracy is 79.96% in the BCI competition Ⅳ-Ⅱa dataset. Moreover, the classification effect of sub-band features of different subjects is verified by comparison tests, the category propensity of different sub-band features is verified by confusion matrix tests and the distribution in different classes of each sub-band feature and fused feature are showed by visual analysis, revealing the importance of different sub-band features for the EEG-based MI classification task. |
| first_indexed | 2024-03-09T01:12:38Z |
| format | Article |
| id | doaj.art-622d370c3df349da9321d2f8fdb5159d |
| institution | Directory Open Access Journal |
| issn | 1551-0018 |
| language | English |
| last_indexed | 2024-03-09T01:12:38Z |
| publishDate | 2023-11-01 |
| publisher | AIMS Press |
| record_format | Article |
| series | Mathematical Biosciences and Engineering |
| spelling | doaj.art-622d370c3df349da9321d2f8fdb5159d2023-12-11T01:11:28ZengAIMS PressMathematical Biosciences and Engineering1551-00182023-11-012012206242064710.3934/mbe.2023912A multi-band centroid contrastive reconstruction fusion network for motor imagery electroencephalogram signal decodingJiacan Xu 0Donglin Li1Peng Zhou2Chunsheng Li3Zinan Wang 4Shenghao Tong 51. The College of Engineering Training and Innovation, Shenyang Jianzhu University, Shenyang 110000, China2. The College of Electrical Engineering, Shenyang University of Technology, Shenyang 110000, China1. The College of Engineering Training and Innovation, Shenyang Jianzhu University, Shenyang 110000, China2. The College of Electrical Engineering, Shenyang University of Technology, Shenyang 110000, China1. The College of Engineering Training and Innovation, Shenyang Jianzhu University, Shenyang 110000, China1. The College of Engineering Training and Innovation, Shenyang Jianzhu University, Shenyang 110000, ChinaMotor imagery (MI) brain-computer interface (BCI) assist users in establishing direct communication between their brain and external devices by decoding the movement intention of human electroencephalogram (EEG) signals. However, cerebral cortical potentials are highly rhythmic and sub-band features, different experimental situations and subjects have different categories of semantic information in specific sample target spaces. Feature fusion can lead to more discriminative features, but simple fusion of features from different embedding spaces leading to the model global loss is not easily convergent and ignores the complementarity of features. Considering the similarity and category contribution of different sub-band features, we propose a multi-band centroid contrastive reconstruction fusion network (MB-CCRF). We obtain multi-band spatio-temporal features by frequency division, preserving the task-related rhythmic features of different EEG signals; use a multi-stream cross-layer connected convolutional network to perform a deep feature representation for each sub-band separately; propose a centroid contrastive reconstruction fusion module, which maps different sub-band and category features into the same shared embedding space by comparing with category prototypes, reconstructing the feature semantic structure to ensure that the global loss of the fused features converges more easily. Finally, we use a learning mechanism to model the similarity between channel features and use it as the weight of fused sub-band features, thus enhancing the more discriminative features, suppressing the useless features. The experimental accuracy is 79.96% in the BCI competition Ⅳ-Ⅱa dataset. Moreover, the classification effect of sub-band features of different subjects is verified by comparison tests, the category propensity of different sub-band features is verified by confusion matrix tests and the distribution in different classes of each sub-band feature and fused feature are showed by visual analysis, revealing the importance of different sub-band features for the EEG-based MI classification task.https://www.aimspress.com/article/doi/10.3934/mbe.2023912?viewType=HTMLbrain computer interface (bci)motor imagery (mi)electroencephalogram (eeg)feature reconstructionfeature fusion |
| spellingShingle | Jiacan Xu Donglin Li Peng Zhou Chunsheng Li Zinan Wang Shenghao Tong A multi-band centroid contrastive reconstruction fusion network for motor imagery electroencephalogram signal decoding Mathematical Biosciences and Engineering brain computer interface (bci) motor imagery (mi) electroencephalogram (eeg) feature reconstruction feature fusion |
| title | A multi-band centroid contrastive reconstruction fusion network for motor imagery electroencephalogram signal decoding |
| title_full | A multi-band centroid contrastive reconstruction fusion network for motor imagery electroencephalogram signal decoding |
| title_fullStr | A multi-band centroid contrastive reconstruction fusion network for motor imagery electroencephalogram signal decoding |
| title_full_unstemmed | A multi-band centroid contrastive reconstruction fusion network for motor imagery electroencephalogram signal decoding |
| title_short | A multi-band centroid contrastive reconstruction fusion network for motor imagery electroencephalogram signal decoding |
| title_sort | multi band centroid contrastive reconstruction fusion network for motor imagery electroencephalogram signal decoding |
| topic | brain computer interface (bci) motor imagery (mi) electroencephalogram (eeg) feature reconstruction feature fusion |
| url | https://www.aimspress.com/article/doi/10.3934/mbe.2023912?viewType=HTML |
| work_keys_str_mv | AT jiacanxu amultibandcentroidcontrastivereconstructionfusionnetworkformotorimageryelectroencephalogramsignaldecoding AT donglinli amultibandcentroidcontrastivereconstructionfusionnetworkformotorimageryelectroencephalogramsignaldecoding AT pengzhou amultibandcentroidcontrastivereconstructionfusionnetworkformotorimageryelectroencephalogramsignaldecoding AT chunshengli amultibandcentroidcontrastivereconstructionfusionnetworkformotorimageryelectroencephalogramsignaldecoding AT zinanwang amultibandcentroidcontrastivereconstructionfusionnetworkformotorimageryelectroencephalogramsignaldecoding AT shenghaotong amultibandcentroidcontrastivereconstructionfusionnetworkformotorimageryelectroencephalogramsignaldecoding AT jiacanxu multibandcentroidcontrastivereconstructionfusionnetworkformotorimageryelectroencephalogramsignaldecoding AT donglinli multibandcentroidcontrastivereconstructionfusionnetworkformotorimageryelectroencephalogramsignaldecoding AT pengzhou multibandcentroidcontrastivereconstructionfusionnetworkformotorimageryelectroencephalogramsignaldecoding AT chunshengli multibandcentroidcontrastivereconstructionfusionnetworkformotorimageryelectroencephalogramsignaldecoding AT zinanwang multibandcentroidcontrastivereconstructionfusionnetworkformotorimageryelectroencephalogramsignaldecoding AT shenghaotong multibandcentroidcontrastivereconstructionfusionnetworkformotorimageryelectroencephalogramsignaldecoding |