EUVIMEDEUVIMED entdecken
Crossref

MTDANet: lightweight multi-scale temporal hierarchy and distance-regularized linear attention network for motor imagery brain–computer interfaces

Binbin Chen iD, Chen Deng iD, Huanmin Ge iD, Xinhua Su iD, Yanfei Shen iD

Journal of Neural Engineering · 2026

DOI: 10.1088/1741-2552/aeb330 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Neural Engineering, 2026. Verfasst von Binbin Chen und 4 weiteren Autor:innen. Beteiligte Einrichtung: School of Sport Engineering, Beijing Sport University, Beijing 100084, People's Republic of China. Hauptthemen laut MEDLINE: Brain-Computer Interfaces, Electroencephalography, Attention, Imagination. Die Arbeit nennt 45 Literaturangaben. Gefördert von: Natural Science Foundation of Beijing Municipality, Chinese Universities Scientific Fund, National Natural Science Foundation of China.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Objective . Motor-imagery (MI) electroencephalography (EEG) offers a non-invasive route to brain–computer interaction, yet reliable single-trial decoding remains difficult because task-related patterns are weak, nonstationary, and distributed over multiple temporal scales. This study investigates whether a staged local-to-global model can integrate fine temporal–spatial structure with longer-range context without relying on unrestricted self-attention. Approach. We propose MTDANet, an architecture that combines multi-scale temporal–spatial feature extraction, factorized temporal consolidation, and distance-regularized recurrent attention. The model was evaluated on three public MI datasets under subject-specific and leave-one-subject-out (LOSO) protocols. Separately initialized models were also trained on P300 and steady-state visual evoked potential (SSVEP) datasets to examine broader architectural applicability. Participant-level paired analyses used two-sided paired t tests with within-setting Holm correction. Main results. MTDANet achieved subject-specific accuracies of 85.61 ± 8.32 % , 90.85 ± 7.62 % , and 87.79 ± 13.93 % on brain–computer interface (BCI) Competition IV 2a, BCI Competition IV 2b, and BNCI2014_002, respectively. Relative to the highest-mean baseline, the paired improvements were supported after within-setting Holm correction on BCI Competition IV 2a (3.56 percentage points, adjusted p = 0.0097 ) and 2b (1.62 points, adjusted p = 0.0168 ). Under LOSO evaluation, the corresponding accuracies were 63.31 ± 15.74 % , 79.77 ± 7.57 % , and 79.96 ± 11.95 % . The 2.08-point BNCI2014_002 difference had a 95% confidence interval of 0.42–3.73 points and an adjusted p value of 0.0387. These three comparisons did not remain significant in a broader 56-comparison Holm sensitivity analysis. The P300 and SSVEP models attained the highest observed mean accuracies. Significance. The results provide within-setting evidence for staged local-to-global temporal modeling in offline MI-EEG evaluations.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Brain-Computer InterfacesElectroencephalographyAttentionImaginationHumansEvoked Potentials, VisualMaleFemaleAlgorithms

Stichwörter der Autor:innen

attention mechanismbrain–computer interfaces (BCIs)electroencephalography (EEG)lightweight neural networkmotor imagery (MI)temporal modeling

Autor:innen und Institutionen

  • School of Sport Engineering, Beijing Sport University, Beijing 100084, People's Republic of China.
  • School of Sport Science, Beijing Sport University, Beijing 100084, People's Republic of China.

ORCID-Profile: Binbin Chen, Chen Deng, Huanmin Ge, Xinhua Su, Yanfei Shen

Förderung

  • Natural Science Foundation of Beijing Municipality
  • Chinese Universities Scientific Fund
  • National Natural Science Foundation of China

Zitieren

APA

Chen, B., Deng, C., Ge, H., Su, X., & Shen, Y. (2026). MTDANet: lightweight multi-scale temporal hierarchy and distance-regularized linear attention network for motor imagery brain–computer interfaces. Journal of Neural Engineering. https://doi.org/10.1088/1741-2552/aeb330

Vancouver

Chen B, Deng C, Ge H, Su X, Shen Y. MTDANet: lightweight multi-scale temporal hierarchy and distance-regularized linear attention network for motor imagery brain–computer interfaces. Journal of Neural Engineering. 2026. doi:10.1088/1741-2552/aeb330

BibTeX
@article{Chen2026,
  author = {Chen, Binbin and Deng, Chen and Ge, Huanmin and Su, Xinhua and Shen, Yanfei},
  title = {MTDANet: lightweight multi-scale temporal hierarchy and distance-regularized linear attention network for motor imagery brain–computer interfaces},
  journal = {Journal of Neural Engineering},
  year = {2026},
  doi = {10.1088/1741-2552/aeb330}
}
Kostenlos starten →