EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

EEG–ShuffleFormer: A Multi-View Hybrid Network Integrating Time–Frequency and Raw Signal Representations for Few-Channel Motor Imagery EEG Classification

Kang Fan iD, Qin Gu, Yaduan Ruan iD

Bioengineering · 19.05.2026 · 13 (5) · 578

DOI: 10.3390/bioengineering13050578 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Bioengineering, online erschienen am 19.05.2026. Verfasst von Kang Fan und 2 weiteren Autor:innen. Beteiligte Einrichtung: Department of Critical Care Medicine, Nanjing Drum Tower Hospital, Affiliated Hospital of Medical School, Nanjing University, Nanjing 210008, China. Die Arbeit nennt 31 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Electroencephalogram (EEG) signals hold significant research value in brain function decoding, disease diagnosis, and brain–computer interfaces (BCIs). Few-channel EEG recording devices feature superior portability, simple operation, and facilitated real-time monitoring implementation. However, few-channel motor imagery (MI) EEG signals inherently suffer from data scarcity and limited spatial discriminative information, which pose critical challenges, including insufficient feature extraction and poor robustness in classification tasks. To address these issues, this paper presents EEG–ShuffleFormer, a hybrid network that integrates two complementary views of EEG signals: time–frequency representations obtained via continuous wavelet transform and the original raw signal representations. A lightweight ShuffleNet backbone extracts local features, followed by a Transformer encoder that models long-range temporal dependencies. Evaluated on the BCI Competition IV Dataset 2b, the proposed method achieves an average classification accuracy of 82.23%, with a substantial improvement on challenging subjects compared to the closest baseline method. Compared with existing methods, the proposed multi-view fusion strategy raises the performance floor while maintaining high accuracy on typical subjects, demonstrating its potential to enhance robustness for different subjects in few-channel scenarios.

Schlagwörter

Stichwörter der Autor:innen

ShuffleNetTransformercontinuous wavelet transformfew-channel MI EEGhybrid networkmulti-view feature fusiontransfer learning

Autor:innen und Institutionen

  • Department of Critical Care Medicine, Nanjing Drum Tower Hospital, Affiliated Hospital of Medical School, Nanjing University, Nanjing 210008, China.

ORCID-Profile: Kang Fan, Yaduan Ruan

Zitieren

APA

Fan, K., Gu, Q., & Ruan, Y. (2026). EEG–ShuffleFormer: A Multi-View Hybrid Network Integrating Time–Frequency and Raw Signal Representations for Few-Channel Motor Imagery EEG Classification. Bioengineering, 13(5), 578. https://doi.org/10.3390/bioengineering13050578

Vancouver

Fan K, Gu Q, Ruan Y. EEG–ShuffleFormer: A Multi-View Hybrid Network Integrating Time–Frequency and Raw Signal Representations for Few-Channel Motor Imagery EEG Classification. Bioengineering. 2026;13(5):578. doi:10.3390/bioengineering13050578

BibTeX
@article{Fan2026,
  author = {Fan, Kang and Gu, Qin and Ruan, Yaduan},
  title = {EEG–ShuffleFormer: A Multi-View Hybrid Network Integrating Time–Frequency and Raw Signal Representations for Few-Channel Motor Imagery EEG Classification},
  journal = {Bioengineering},
  year = {2026},
  volume = {13},
  number = {5},
  pages = {578},
  doi = {10.3390/bioengineering13050578}
}
Kostenlos starten →