EEGNet-based multi-source domain filter for BCI transfer learning

    Deep learning has great potential on decoding EEG in brain-computer interface. While common deep learning algorithms cannot directly train models with data from multiple individuals because of the inter-individual differences in EEG. Collecting enough data for each subject to satisfy the trainin...

Full description

Saved in:
Bibliographic Details
Published in:Medical & biological engineering & computing 2024-03, Vol.62 (3), p.675-686
Main Authors: Li, Mengfan, Li, Jundi, Song, Zhiyong, Deng, Haodong, Xu, Jiaming, Xu, Guizhi, Liao, Wenzhe
Format: Article
Language:English
Subjects:
Algorithms
bags
Biomedical and Life Sciences
Biomedical Engineering and Bioengineering
Biomedicine
Brain
Brain-Computer Interfaces
Computer Applications
data collection
Deep learning
domain
EEG
Electroencephalography
filters
Human Physiology
Human-computer interface
Humans
Imaging
Implants
Invariants
learning
Machine Learning
Model accuracy
Original Article
Performance enhancement
Radiology
Transfer learning
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:    Deep learning has great potential on decoding EEG in brain-computer interface. While common deep learning algorithms cannot directly train models with data from multiple individuals because of the inter-individual differences in EEG. Collecting enough data for each subject to satisfy the training of deep learning would result in an increase in training cost. This study proposes a novel transfer learning, EEGNet-based multi-source domain filter for transfer learning (EEGNet-MDFTL), to reduce the amount of training data and improve the performance of BCI. The EEGNet-MDFTL uses bagging ensemble learning to learn domain-invariant features from the multi-source domain and utilizes model loss value to filter the multi-source domain. Compared with baseline methods, the accuracy of the EEGNet-MDFTL reaches 91.96%, higher than two state-of-the-art methods, which demonstrates source domain filter can select similar source domains to improve the accuracy of the model, and remains a high level even when the data amount is reduced to 1/8, proving that ensemble learning learns enough domain invariant features from the multi-source domain to make the model insensitive to data amount. The proposed EEGNet-MDFTL is effective in improving the decoding performance with a small amount of data, which is helpful to save the BCI training cost. Graphical Abstract
ISSN:0140-0118
1741-0444
DOI:10.1007/s11517-023-02967-z