A Multi-Source Transfer Joint Matching Method for Inter-Subject Motor Imagery Decoding
Individual differences among different subjects pose a great challenge to motor imagery (MI) decoding. Multi-source transfer learning (MSTL) is one of the most promising ways to reduce individual differences, which can utilize rich information and align the data distribution among different subjects...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2023-01-01
|
| Series: | IEEE Transactions on Neural Systems and Rehabilitation Engineering |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10040708/ |
| _version_ | 1827927009957773312 |
|---|---|
| author | Fulin Wei Xueyuan Xu Tianyuan Jia Daoqiang Zhang Xia Wu |
| author_facet | Fulin Wei Xueyuan Xu Tianyuan Jia Daoqiang Zhang Xia Wu |
| author_sort | Fulin Wei |
| collection | DOAJ |
| description | Individual differences among different subjects pose a great challenge to motor imagery (MI) decoding. Multi-source transfer learning (MSTL) is one of the most promising ways to reduce individual differences, which can utilize rich information and align the data distribution among different subjects. However, most MSTL methods in MI-BCI combine all data in the source subjects into a single mixed domain, which will ignore the effect of important samples and the large differences in multiple source subjects. To address these issues, we introduce transfer joint matching and improve it to multi-source transfer joint matching (MSTJM) and weighted MSTJM (wMSTJM). Different from previous MSTL methods in MI, our methods align the data distribution for each pair of subjects, and then integrate the results by decision fusion. Besides that, we design an inter-subject MI decoding framework to verify the effectiveness of these two MSTL algorithms. It mainly consists of three modules: covariance matrix centroid alignment in the Riemannian space, source selection in the Euclidean space after tangent space mapping to reduce negative transfer and computation overhead, and further distribution alignment by MSTJM or wMSTJM. The superiority of this framework is verified on two common public MI datasets from BCI competition IV. The average classification accuracy of the MSTJM and wMSTJ methods outperformed other state-of-the-art methods by at least 4.24% and 2.62% respectively. It’s promising to advance the practical applications of MI-BCI. |
| first_indexed | 2024-03-13T05:45:57Z |
| format | Article |
| id | doaj.art-969ed14d91a2426095f241806cf31ef1 |
| institution | Directory Open Access Journal |
| issn | 1558-0210 |
| language | English |
| last_indexed | 2024-03-13T05:45:57Z |
| publishDate | 2023-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Transactions on Neural Systems and Rehabilitation Engineering |
| spelling | doaj.art-969ed14d91a2426095f241806cf31ef12023-06-13T20:10:45ZengIEEEIEEE Transactions on Neural Systems and Rehabilitation Engineering1558-02102023-01-01311258126710.1109/TNSRE.2023.324325710040708A Multi-Source Transfer Joint Matching Method for Inter-Subject Motor Imagery DecodingFulin Wei0https://orcid.org/0000-0003-1962-0675Xueyuan Xu1Tianyuan Jia2Daoqiang Zhang3https://orcid.org/0000-0002-5658-7643Xia Wu4https://orcid.org/0000-0002-2377-6093School of Artificial Intelligence, Beijing Normal University, Beijing, ChinaFaculty of Information Technology, Beijing University of Technology, Beijing, ChinaSchool of Artificial Intelligence, Beijing Normal University, Beijing, ChinaMIIT Key Laboratory of Pattern Analysis and Machine Intelligence, College of Computer Science and Technology, Nanjing University of Aeronautics and Astronautics, Nanjing, ChinaSchool of Artificial Intelligence, Beijing Normal University, Beijing, ChinaIndividual differences among different subjects pose a great challenge to motor imagery (MI) decoding. Multi-source transfer learning (MSTL) is one of the most promising ways to reduce individual differences, which can utilize rich information and align the data distribution among different subjects. However, most MSTL methods in MI-BCI combine all data in the source subjects into a single mixed domain, which will ignore the effect of important samples and the large differences in multiple source subjects. To address these issues, we introduce transfer joint matching and improve it to multi-source transfer joint matching (MSTJM) and weighted MSTJM (wMSTJM). Different from previous MSTL methods in MI, our methods align the data distribution for each pair of subjects, and then integrate the results by decision fusion. Besides that, we design an inter-subject MI decoding framework to verify the effectiveness of these two MSTL algorithms. It mainly consists of three modules: covariance matrix centroid alignment in the Riemannian space, source selection in the Euclidean space after tangent space mapping to reduce negative transfer and computation overhead, and further distribution alignment by MSTJM or wMSTJM. The superiority of this framework is verified on two common public MI datasets from BCI competition IV. The average classification accuracy of the MSTJM and wMSTJ methods outperformed other state-of-the-art methods by at least 4.24% and 2.62% respectively. It’s promising to advance the practical applications of MI-BCI.https://ieeexplore.ieee.org/document/10040708/Brain–computer interfaceinter-subject variabilitytransfer joint matchingmulti-source transfer learning |
| spellingShingle | Fulin Wei Xueyuan Xu Tianyuan Jia Daoqiang Zhang Xia Wu A Multi-Source Transfer Joint Matching Method for Inter-Subject Motor Imagery Decoding IEEE Transactions on Neural Systems and Rehabilitation Engineering Brain–computer interface inter-subject variability transfer joint matching multi-source transfer learning |
| title | A Multi-Source Transfer Joint Matching Method for Inter-Subject Motor Imagery Decoding |
| title_full | A Multi-Source Transfer Joint Matching Method for Inter-Subject Motor Imagery Decoding |
| title_fullStr | A Multi-Source Transfer Joint Matching Method for Inter-Subject Motor Imagery Decoding |
| title_full_unstemmed | A Multi-Source Transfer Joint Matching Method for Inter-Subject Motor Imagery Decoding |
| title_short | A Multi-Source Transfer Joint Matching Method for Inter-Subject Motor Imagery Decoding |
| title_sort | multi source transfer joint matching method for inter subject motor imagery decoding |
| topic | Brain–computer interface inter-subject variability transfer joint matching multi-source transfer learning |
| url | https://ieeexplore.ieee.org/document/10040708/ |
| work_keys_str_mv | AT fulinwei amultisourcetransferjointmatchingmethodforintersubjectmotorimagerydecoding AT xueyuanxu amultisourcetransferjointmatchingmethodforintersubjectmotorimagerydecoding AT tianyuanjia amultisourcetransferjointmatchingmethodforintersubjectmotorimagerydecoding AT daoqiangzhang amultisourcetransferjointmatchingmethodforintersubjectmotorimagerydecoding AT xiawu amultisourcetransferjointmatchingmethodforintersubjectmotorimagerydecoding AT fulinwei multisourcetransferjointmatchingmethodforintersubjectmotorimagerydecoding AT xueyuanxu multisourcetransferjointmatchingmethodforintersubjectmotorimagerydecoding AT tianyuanjia multisourcetransferjointmatchingmethodforintersubjectmotorimagerydecoding AT daoqiangzhang multisourcetransferjointmatchingmethodforintersubjectmotorimagerydecoding AT xiawu multisourcetransferjointmatchingmethodforintersubjectmotorimagerydecoding |