Abstract
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
Similar content being viewed by others
References
Allison BZ, Kübler A, Jin J (2020) 30+ years of P300 brain–computer interfaces. Psychophysiology 57(7):e13569
Li W, Li M, Zhou H et al (2018) A dual stimuli approach combined with convolutional neural network to improve information transfer rate of event-related potential-based brain-computer interface. Int J Neural Syst 28(10):1850034
Zhang R, Wang Q, Li K et al (2017) A BCI-based environmental control system for patients with severe spinal cord injuries. IEEE Trans Biomed Eng 64(8):1959–1971
Duan F, Lin D, Li W et al (2015) Design of a multimodal EEG-based hybrid BCI system with visual servo module. IEEE Trans Auton Ment Dev 7(4):332–341
Bian Z, Zhang X (2000) Pattern recognition. Tsinghua University Press, China
Xu D, Wang Q (2021) Noninvasive human-prosthesis interfaces for locomotion intent recognition: a review. Cyborg Bionic Syst 2021:9863761
Dos Santos EM, San-Martin R, Fraga FJ (2023) Comparison of subject-independent and subject-specific EEG-based BCI using LDA and SVM classifiers. Med Biol Eng Comput 61(3):835–845. https://doi.org/10.1007/s11517-023-02769-3
Alkan A, Akben SB (2011) Use of k-means clustering in migraine detection by using EEG records under flash stimulation. Int J Phys Sci 6(4):641–650
Sunnetci KM, Alkan A (2022) Biphasic majority voting-based comparative COVID-19 diagnosis using chest X-ray images. Exp Syst Appl 216:119430. https://doi.org/10.1016/j.eswa.2022.119430
Sunnetci KM, Kaba E, Celiker FB et al (2023) Deep network-based comprehensive parotid gland tumor detection. Acad Radiol. https://doi.org/10.1016/j.acra.2023.04.028
Li M, Han J, Yang J (2021) Automatic feature extraction and fusion recognition of motor imagery EEG using multilevel multiscale CNN. Med Biol Eng Compu 59(10):2037–2050
Borhade R, Nagmode MS (2020) Modified atom search optimization-based deep recurrent neural network for epileptic seizure prediction using electroencephalogram signals. Biocybernetics Biomed Eng 40(4):1638–1653
She Q, Zou J, Luo Z et al (2020) Multi-class motor imagery EEG classification using collaborative representation-based semi-supervised extreme learning machine. Med Biol Eng Compu 58:2119–2130
Mishuhina V, Jiang X (2018) Feature weighting and regularization of common spatial patterns in EEG-based motor imagery BCI. IEEE Signal Process Lett 25(6):783–787
Abibullaev B, Zollanvari A (2021) A systematic deep learning model selection for P300-based brain–computer interfaces. IEEE Trans Syst Man Cybernet: Syst 52(5):2744–2756
Amin SU, Alsulaiman M, Muhammad G et al (2019) Deep learning for EEG motor imagery classification based on multi-layer CNNs feature fusion. Futur Gener Comput Syst 101:542–554
Pan SJ, Yang Q (2010) A survey on transfer learning. IEEE Trans Knowl Data Eng 22(10):1345–1359
Kindermans PJ, Tangermann M, Müller KR et al (2014) Integrating dynamic stopping, transfer learning and language models in an adaptive zero-training ERP speller. J Neural Eng 11(3):035005
Fahimi F, Zhang Z, Goh WB et al (2019) Inter-subject transfer learning with an end-to-end deep convolutional neural network for EEG-based BCI. J Neural Eng 16(2):026007
Zhang D, Yao L, Zhang X et al (2018) Cascade and parallel convolutional recurrent neural networks on EEG-based intention recognition for brain computer interface. Proc AAAI Conf Artif Intell 32:1
Jin J, Li S, Daly I et al (2020) The study of generic model set for reducing calibration time in P300-based brain–computer interface. IEEE Trans Neural Syst Rehabil Eng 28(1):3–12
Raza H, Cecotti H, Li Y et al (2016) Adaptive learning with covariate shift-detection for motor imagery-based brain–computer interface. Soft Comput 20:3085–3096
Li M, Lin F, Xu G (2020) A TrAdaBoost method for detecting multiple subjects’ N200 and P300 potentials based on cross-validation and an adaptive threshold. Int J Neural Syst 30(3):2050009
Roy AM (2022) Adaptive transfer learning-based multiscale feature fused deep convolutional neural network for EEG MI multiclassification in brain–computer interface. Eng Appl Artif Intell 116:105347
Kang H, Nam Y, Choi S (2009) Composite common spatial pattern for subject-to-subject transfer. IEEE Signal process Lett 16(8):683–686
Lees S (2018) A review of rapid serial visual presentation-based brain–computer interfaces. J Neural Eng 15(2):1–24
Azab AM, Mihaylova L, Ang KK et al (2019) Weighted transfer learning for improving motor imagery-based brain–computer interface. IEEE Trans Neural Syst Rehabil Eng 27(7):1352–1359
He H, Wu D (2020) Transfer learning for brain–computer interfaces: a Euclidean space data alignment approach. IEEE Trans Biomed Eng 67(2):399–410
Zhao X, Zhao J, Liu C et al (2020) Deep neural network with joint distribution matching for cross-subject motor imagery brain-computer interfaces. Biomed Res Int 2020:1–15
Lawhern VJ, Solon AJ, Waytowich NR et al (2018) EEGNet: a compact convolutional neural network for EEG-based brain–computer interfaces. J Neural Eng 15(5):1–17
Sun B, Zhao X, Zhang H et al (2020) EEG motor imagery classification with sparse spectrotemporal decomposition and deep learning. IEEE Trans Automat Sci Eng 18(2):541–551
Schirrmeister R, Springenberg J, Fiederer L et al (2017) Deep learning with convolutional neural networks for EEG decoding and visualization. Hum Brain Mapp 38(11):5391–5420
Ni Z, Xu J, Wu Y et al (2022) Improving cross-state and cross-subject visual ERP-based BCI with temporal modeling and adversarial training. IEEE Trans Neural Syst Rehabil Eng 30:369–379
Kumar S, Reddy T, Arora V et al (2020) Formulating divergence framework for multiclass motor imagery EEG brain computer interface. In: Proc. IEEE Int Conf Acoustics Speech Signal Process. IEEE, pp 1344–1348. https://doi.org/10.1109/icassp40776.2020.9053168
Lotte F, Bougrain L, Cichocki A et al (2018) A review of classification algorithms for EEG-based braincomputer interfaces: a 10 year update. J Neural Eng 15(3):031005
Kalaganis F, Laskaris N, Chatzilari E et al (2020) A Riemannian geometry approach to reduced and discriminative covariance estimation in brain computer interfaces. IEEE Trans Biomed Eng 67(1):245–255
Hussein R, Palangi H, Ward R et al (2019) Optimized deep neural network architecture for robust detection of epileptic seizures using EEG signals. Clin Neuriophysiol 130:25–37
Li F, Xia Y, Wang F et al (2020) Transfer learning algorithm of P300-EEG signal based on XDAWN spatial filter and Riemannian geometry classifier. Appl Sci 10(5):1804
Kilani S, Aghili SN, Hulea M (2023) Enhancing P300-based brain-computer interfaces with hybrid transfer learning: a data alignment and fine-tuning approach. Appl Sci 13(10):6283
Huang Z, Guo J, Zheng W et al (2022) A calibration-free approach to implementing P300-based brain–computer interface. Cogn Comput 14(2):887–899
Wu Z, She Q, Hou Z et al (2023) Multi-source online transfer algorithm based on source domain selection for EEG classification. Math Biosci Eng 20(3):4560–4573
Long M, Wang J, Ding G et al (2013) Transfer feature learning with joint distribution adaptation. In: Proceedings of the IEEE international conference on computer vision. IEEE, pp 2200–2207. https://doi.org/10.1109/iccv.2013.274
Funding
This work was funded by the Natural Science Foundation of Hebei Province (F2021202003), the Technology Nova of Hebei University of Technology (JBKYXX2007), the State Key Laboratory of Reliability and Intelligence of Electrical Equipment (EERI_OY2020004, EERI_OY202000), the National Natural Science Foundation of China (51977060), and the Key Research and Development Foundation of Hebei (19277752D, 21372002D).
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare no competing interests.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Li, M., Li, J., Song, Z. et al. EEGNet-based multi-source domain filter for BCI transfer learning. Med Biol Eng Comput 62, 675–686 (2024). https://doi.org/10.1007/s11517-023-02967-z
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11517-023-02967-z