Personalized Adaptive Gabor Filtering with Three-Stage Semi-Supervised Domain-Adversarial Learning for Cross-Subject SSVEP Decoding
Sensors · 10.06.2026 · 26 (12) · 3694
DOI: 10.3390/s26123694 ↗Kurz gefasst
Fachartikel in der Zeitschrift Sensors, online erschienen am 10.06.2026. Verfasst von Junjun Guo und 4 weiteren Autor:innen. Beteiligte Einrichtung: School of Mathematics and Information Engineering, Longdong University, Qingyang 745000, China. Hauptthemen laut MEDLINE: Evoked Potentials, Visual, Brain-Computer Interfaces. Die Arbeit nennt 33 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Innovation Fund Project for University Teachers in Gansu Province, Key Research and Development Program of Gansu Province-Industrial Project, Scientific Research Projects of Higher Education Institutions in Gansu Province.
Abstract (Originalsprache)
Improving the decoding accuracy and information transfer rate (ITR) of steady-state visual evoked potential brain–computer interface (SSVEP-BCI) systems, while enhancing cross-subject generalization and reducing calibration cost, is essential for practical deployment. This study proposes an end-to-end framework that integrates adaptive filtering with semi-supervised domain adaptation. The framework incorporates a Gabor adaptive filter bank (G-AFB) to optimize time–frequency representations and extract features matched to individual neural responses. It also introduces a three-stage semi-supervised domain-adversarial neural network (TriS-DANN), which combines unsupervised pre-alignment and supervised fine-tuning to align cross-subject feature distributions and enable lightweight calibration. On the 1.0 s public benchmark dataset, G-AFB-tCNN achieved 89.13% accuracy, a 4.63 percentage-point improvement over its conventional filter-bank counterpart. On the 0.4 s in-house dataset, G-AFB-tCNN achieved 91.85% accuracy, a 3.22 percentage-point improvement over the conventional fixed filter bank. In transfer learning, TriS-DANN reached 86.60% accuracy using 0.4 s segments extracted from the stimulation period and only 23.07% of the available target-domain training/calibration trials, demonstrating higher efficiency and stability than conventional fine-tuning. These results support the proposed framework as a feasible route toward reliable, low-calibration SSVEP-BCI systems.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- School of Mathematics and Information Engineering, Longdong University, Qingyang 745000, China.
- School of Intelligent Manufacturing, Longdong University, Qingyang 745000, China.
- School of Life Science and Technology, Northwestern Polytechnical University, Xi'an 710072, China.
ORCID-Profile: Junjun Guo, Xiaonan Pan, Ting Huyan
Förderung
- Innovation Fund Project for University Teachers in Gansu Province
- Key Research and Development Program of Gansu Province-Industrial Project
- Scientific Research Projects of Higher Education Institutions in Gansu Province
- Qingyang City Major Science and Technology Project-Industrial Field Project
- Doctoral Fund Project of Longdong University
- Longdong University Higher Education Teaching Research 2026 Key Project
- Crossing Research Project of Longdong University
- Key Research and Development Program of Gansu Province–Industrial Project
Zitieren
APA
Guo, J., Pan, X., Mi, N., Zhang, J., & Huyan, T. (2026). Personalized Adaptive Gabor Filtering with Three-Stage Semi-Supervised Domain-Adversarial Learning for Cross-Subject SSVEP Decoding. Sensors, 26(12), 3694. https://doi.org/10.3390/s26123694
Vancouver
Guo J, Pan X, Mi N, Zhang J, Huyan T. Personalized Adaptive Gabor Filtering with Three-Stage Semi-Supervised Domain-Adversarial Learning for Cross-Subject SSVEP Decoding. Sensors. 2026;26(12):3694. doi:10.3390/s26123694
BibTeX
@article{Guo2026,
author = {Guo, Junjun and Pan, Xiaonan and Mi, Ning and Zhang, Jianrui and Huyan, Ting},
title = {Personalized Adaptive Gabor Filtering with Three-Stage Semi-Supervised Domain-Adversarial Learning for Cross-Subject SSVEP Decoding},
journal = {Sensors},
year = {2026},
volume = {26},
number = {12},
pages = {3694},
doi = {10.3390/s26123694}
}
EUVIMED