A Novel 3D Approach with a CNN and Swin Transformer for Decoding EEG-Based Motor Imagery Classification
Sensors · 05.05.2025 · 25 (9) · 2922
DOI: 10.3390/s25092922 ↗Kurz gefasst
Fachartikel in der Zeitschrift Sensors, online erschienen am 05.05.2025. Verfasst von Xin Deng und 4 weiteren Autor:innen. Beteiligte Einrichtung: Chongqing Key Laboratory of Germplasm Innovation and Utilization of Native Plants, Chongqing 401329, China. Hauptthemen laut MEDLINE: Electroencephalography, Neural Networks, Computer, Imagination. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 52 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Chongqing Key Laboratory of Germplasm Innovation 755 and Utilization of Native Plants under Grant, Chongqing Key Laboratory of Germplasm Innovation and Utilization of Native Plants.
Abstract (Originalsprache)
Motor imagery (MI) is a crucial research field within the brain–computer interface (BCI) domain. It enables patients with muscle or neural damage to control external devices and achieve movement functions by simply imagining bodily motions. Despite the significant clinical and application value of MI-BCI technology, accurately decoding high-dimensional and low signal-to-noise ratio (SNR) electroencephalography (EEG) signals remains challenging. Moreover, traditional deep learning approaches exhibit limitations in processing EEG signals, particularly in capturing the intrinsic correlations between electrode channels and long-distance temporal dependencies. To address these challenges, this research introduces a novel end-to-end decoding network that integrates convolutional neural networks (CNNs) and a Swin Transformer, aiming at enhancing the classification accuracy of the MI paradigm in EEG signals. This approach transforms EEG signals into a three-dimensional data structure, utilizing one-dimensional convolutions along the temporal dimension and two-dimensional convolutions across the EEG electrode distribution for initial spatio-temporal feature extraction, followed by deep feature exploration using a 3D Swin Transformer module. Experimental results show that on the BCI Competition IV-2a dataset, the proposed method achieves 83.99% classification accuracy, which is significantly better than the existing deep learning methods. This finding underscores the efficacy of combining a CNN and Swin Transformer in a 3D data space for processing high-dimensional, low-SNR EEG signals, offering a new perspective for the future development of MI-BCI. Future research could further explore the applicability of this method across various BCI tasks and its potential clinical implementations.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Chongqing Key Laboratory of Germplasm Innovation and Utilization of Native Plants, Chongqing 401329, China.
- The Key Laboratory of Data Engineering and Visual Computing, Chongqing University of Posts and Telecommunications, Chongqing 400065, China.
ORCID-Profile: Huaxiang Huo
Förderung
- Chongqing Key Laboratory of Germplasm Innovation 755 and Utilization of Native Plants under Grant
- Chongqing Key Laboratory of Germplasm Innovation and Utilization of Native Plants
Zitieren
APA
Deng, X., Huo, H., Ai, L., Xu, D., & Li, C. (2025). A Novel 3D Approach with a CNN and Swin Transformer for Decoding EEG-Based Motor Imagery Classification. Sensors, 25(9), 2922. https://doi.org/10.3390/s25092922
Vancouver
Deng X, Huo H, Ai L, Xu D, Li C. A Novel 3D Approach with a CNN and Swin Transformer for Decoding EEG-Based Motor Imagery Classification. Sensors. 2025;25(9):2922. doi:10.3390/s25092922
BibTeX
@article{Deng2025,
author = {Deng, Xin and Huo, Huaxiang and Ai, Lijiao and Xu, Daijiang and Li, Chenhui},
title = {A Novel 3D Approach with a CNN and Swin Transformer for Decoding EEG-Based Motor Imagery Classification},
journal = {Sensors},
year = {2025},
volume = {25},
number = {9},
pages = {2922},
doi = {10.3390/s25092922}
}
EUVIMED