EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Continual-Learning-Enhanced CNN–Transformer Framework for Real-Time Motor-Imagery BCI in Virtual Environments

Chao-Jen Huang, Cheng-Fu Cao, Kuo Kai Shyu iD, Te-Min Lee, Po-Lei Lee

Bioengineering · 06.05.2026 · 13 (5) · 536

DOI: 10.3390/bioengineering13050536 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Bioengineering, online erschienen am 06.05.2026. Verfasst von Chao-Jen Huang und 4 weiteren Autor:innen. Beteiligte Einrichtung: Department of Electrical Engineering, National Central University, Taoyuan 32001, Taiwan. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Science and Technology Council.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Motor imagery (MI)-based brain–computer interfaces (BCIs) provide an intuitive pathway for neural interaction and rehabilitation, yet their practical deployment remains constrained by long calibration requirements, substantial inter-subject variability, and the non-stationary nature of EEG signals. These challenges are amplified when using dry-electrode EEG, which offers superior convenience for real-world systems but produces noisier and less stable recordings than traditional wet electrodes. As a result, online or real-time four-class MI detection—especially with dry electrodes—has been explored only in a limited number of studies, underscoring an important gap in the field and the need for adaptive, intelligent models capable of coping with continuous signal drift. In this study, we propose a real-time MI-BCI framework that integrates immersive action observation (AO) in virtual reality with a continual learning strategy to manage the evolving nature of dry-EEG features. A CNN–Transformer hybrid model is first initialized through AO-enhanced pre-training and subsequently refined via online continual adaptation during user interaction. This continual learning mechanism enables the classifier to incrementally assimilate new MI patterns while preserving previously acquired knowledge, thereby mitigating the performance degradation that typically arises in extended MI-BCI sessions. Experimental results across four motor classes demonstrate improved decoding accuracy and strengthened sensorimotor activation over time, confirming the system’s capacity for user-specific and session-to-session adaptation. By addressing the rarely studied combination of dry electrodes, online four-class MI decoding, and continual learning, the proposed approach enhances MI-BCI robustness, reduces calibration burden, and supports sustainable long-term deployment in intelligent neurotechnology applications.

Schlagwörter

Stichwörter der Autor:innen

brain–computer interfacecontinual learningmotor imagery

Autor:innen und Institutionen

  • Department of Electrical Engineering, National Central University, Taoyuan 32001, Taiwan.
  • Department of Medical Research, Cathay General Hospital, Taipei 106, Taiwan.

ORCID-Profile: Kuo Kai Shyu

Förderung

  • National Science and Technology Council

Zitieren

APA

Huang, C. J., Cao, C. F., Shyu, K. K., Lee, T. M., & Lee, P. L. (2026). Continual-Learning-Enhanced CNN–Transformer Framework for Real-Time Motor-Imagery BCI in Virtual Environments. Bioengineering, 13(5), 536. https://doi.org/10.3390/bioengineering13050536

Vancouver

Huang CJ, Cao CF, Shyu KK, Lee TM, Lee PL. Continual-Learning-Enhanced CNN–Transformer Framework for Real-Time Motor-Imagery BCI in Virtual Environments. Bioengineering. 2026;13(5):536. doi:10.3390/bioengineering13050536

BibTeX
@article{Huang2026,
  author = {Huang, Chao-Jen and Cao, Cheng-Fu and Shyu, Kuo Kai and Lee, Te-Min and Lee, Po-Lei},
  title = {Continual-Learning-Enhanced CNN–Transformer Framework for Real-Time Motor-Imagery BCI in Virtual Environments},
  journal = {Bioengineering},
  year = {2026},
  volume = {13},
  number = {5},
  pages = {536},
  doi = {10.3390/bioengineering13050536}
}
Kostenlos starten →