Recognition of MI-EEG signals using extended-LSR-based inductive transfer learning

IntroductionMotor imagery electroencephalographic (MI-EEG) signal recognition is used in various brain–computer interface (BCI) systems. In most existing BCI systems, this identification relies on classification algorithms. However, generally, a large amount of subject-specific labeled training data...

Full description

Saved in:
Bibliographic Details
Main Authors: Zhibin Jiang, Keli Hu, Jia Qu, Zekang Bian, Donghua Yu, Jie Zhou
Format: Article
Language:English
Published: Frontiers Media S.A. 2025-04-01
Series:Frontiers in Neuroinformatics
Subjects:
Online Access:https://www.frontiersin.org/articles/10.3389/fninf.2025.1559335/full
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850186118160449536
author Zhibin Jiang
Zhibin Jiang
Keli Hu
Keli Hu
Jia Qu
Zekang Bian
Zekang Bian
Donghua Yu
Donghua Yu
Jie Zhou
Jie Zhou
author_facet Zhibin Jiang
Zhibin Jiang
Keli Hu
Keli Hu
Jia Qu
Zekang Bian
Zekang Bian
Donghua Yu
Donghua Yu
Jie Zhou
Jie Zhou
author_sort Zhibin Jiang
collection DOAJ
description IntroductionMotor imagery electroencephalographic (MI-EEG) signal recognition is used in various brain–computer interface (BCI) systems. In most existing BCI systems, this identification relies on classification algorithms. However, generally, a large amount of subject-specific labeled training data is required to reliably calibrate the classification algorithm for each new subject. To address this challenge, an effective strategy is to integrate transfer learning into the construction of intelligent models, allowing knowledge to be transferred from the source domain to enhance the performance of models trained in the target domain. Although transfer learning has been implemented in EEG signal recognition, many existing methods are designed specifically for certain intelligent models, limiting their application and generalization.MethodsTo broaden application and generalization, an extended-LSR-based inductive transfer learning method is proposed to facilitate transfer learning across various classical intelligent models, including neural networks, Takagi-SugenoKang (TSK) fuzzy systems, and kernel methods.Results and discussionThe proposed method not only promotes the transfer of valuable knowledge from the source domain to improve learning performance in the target domain when target domain training data are insufficient but also enhances application and generalization by incorporating multiple classic base models. The experimental results demonstrate the effectiveness of the proposed method in MI-EEG signal recognition.
format Article
id doaj-art-893202873baa4b1386eaceea4630a0bf
institution OA Journals
issn 1662-5196
language English
publishDate 2025-04-01
publisher Frontiers Media S.A.
record_format Article
series Frontiers in Neuroinformatics
spelling doaj-art-893202873baa4b1386eaceea4630a0bf2025-08-20T02:16:29ZengFrontiers Media S.A.Frontiers in Neuroinformatics1662-51962025-04-011910.3389/fninf.2025.15593351559335Recognition of MI-EEG signals using extended-LSR-based inductive transfer learningZhibin Jiang0Zhibin Jiang1Keli Hu2Keli Hu3Jia Qu4Zekang Bian5Zekang Bian6Donghua Yu7Donghua Yu8Jie Zhou9Jie Zhou10Department of Computer Science and Engineering, Shaoxing University, Shaoxing, ChinaInstitute of Artificial Intelligence, Shaoxing University, Shaoxing, ChinaDepartment of Computer Science and Engineering, Shaoxing University, Shaoxing, ChinaInformation Technology R&D Innovation Center of Peking University, Shaoxing, ChinaDepartment of Computer Science and Artificial Intelligence, Changzhou University, Changzhou, ChinaDepartment of AI & Computer Science, Jiangnan University, Wuxi, ChinaDepartment of Taihu Jiangsu Key Construction Lab of IoT Application Technologies, Wuxi, ChinaDepartment of Computer Science and Engineering, Shaoxing University, Shaoxing, ChinaInstitute of Artificial Intelligence, Shaoxing University, Shaoxing, ChinaDepartment of Computer Science and Engineering, Shaoxing University, Shaoxing, ChinaInstitute of Artificial Intelligence, Shaoxing University, Shaoxing, ChinaIntroductionMotor imagery electroencephalographic (MI-EEG) signal recognition is used in various brain–computer interface (BCI) systems. In most existing BCI systems, this identification relies on classification algorithms. However, generally, a large amount of subject-specific labeled training data is required to reliably calibrate the classification algorithm for each new subject. To address this challenge, an effective strategy is to integrate transfer learning into the construction of intelligent models, allowing knowledge to be transferred from the source domain to enhance the performance of models trained in the target domain. Although transfer learning has been implemented in EEG signal recognition, many existing methods are designed specifically for certain intelligent models, limiting their application and generalization.MethodsTo broaden application and generalization, an extended-LSR-based inductive transfer learning method is proposed to facilitate transfer learning across various classical intelligent models, including neural networks, Takagi-SugenoKang (TSK) fuzzy systems, and kernel methods.Results and discussionThe proposed method not only promotes the transfer of valuable knowledge from the source domain to improve learning performance in the target domain when target domain training data are insufficient but also enhances application and generalization by incorporating multiple classic base models. The experimental results demonstrate the effectiveness of the proposed method in MI-EEG signal recognition.https://www.frontiersin.org/articles/10.3389/fninf.2025.1559335/fullmotor imageryEEGbrain-computer interfaceLSRinductive transfer learning
spellingShingle Zhibin Jiang
Zhibin Jiang
Keli Hu
Keli Hu
Jia Qu
Zekang Bian
Zekang Bian
Donghua Yu
Donghua Yu
Jie Zhou
Jie Zhou
Recognition of MI-EEG signals using extended-LSR-based inductive transfer learning
Frontiers in Neuroinformatics
motor imagery
EEG
brain-computer interface
LSR
inductive transfer learning
title Recognition of MI-EEG signals using extended-LSR-based inductive transfer learning
title_full Recognition of MI-EEG signals using extended-LSR-based inductive transfer learning
title_fullStr Recognition of MI-EEG signals using extended-LSR-based inductive transfer learning
title_full_unstemmed Recognition of MI-EEG signals using extended-LSR-based inductive transfer learning
title_short Recognition of MI-EEG signals using extended-LSR-based inductive transfer learning
title_sort recognition of mi eeg signals using extended lsr based inductive transfer learning
topic motor imagery
EEG
brain-computer interface
LSR
inductive transfer learning
url https://www.frontiersin.org/articles/10.3389/fninf.2025.1559335/full
work_keys_str_mv AT zhibinjiang recognitionofmieegsignalsusingextendedlsrbasedinductivetransferlearning
AT zhibinjiang recognitionofmieegsignalsusingextendedlsrbasedinductivetransferlearning
AT kelihu recognitionofmieegsignalsusingextendedlsrbasedinductivetransferlearning
AT kelihu recognitionofmieegsignalsusingextendedlsrbasedinductivetransferlearning
AT jiaqu recognitionofmieegsignalsusingextendedlsrbasedinductivetransferlearning
AT zekangbian recognitionofmieegsignalsusingextendedlsrbasedinductivetransferlearning
AT zekangbian recognitionofmieegsignalsusingextendedlsrbasedinductivetransferlearning
AT donghuayu recognitionofmieegsignalsusingextendedlsrbasedinductivetransferlearning
AT donghuayu recognitionofmieegsignalsusingextendedlsrbasedinductivetransferlearning
AT jiezhou recognitionofmieegsignalsusingextendedlsrbasedinductivetransferlearning
AT jiezhou recognitionofmieegsignalsusingextendedlsrbasedinductivetransferlearning