Multi-scale motor imagery electroencephalogram classification framework with dynamic weighted multi-source domain alignment_Journal of Biomedical Engineering

Authors：

KONG Xintang , ZHANG Pengwei ,  CHEN Jingxia

School of Electronic Information and Artificial Intelligence, Shaanxi University of Science & Technology, Xi’an 710021, P. R. China;

Corresponding?author：

CHEN Jingxia, Email: chenjingxia@sust.edu.cn

Keywords：

Electroencephalogram signal; Motor imagery; Spatiotemporal feature extraction; Multi-source domain alignment; Multi-scale convolution; Transformer

DOI：

10.7507/1001-5515.202607017

Video：

Export PDF Favorites Scan Get Citation

Abstract Full text Figures/Tables Video References Cited by

Brain–computer interface (BCI) technology facilitates information exchange between the brain and external devices, offering new approaches to human–computer interaction and medical rehabilitation. However, distribution shifts caused by inter-subject variability in electroencephalogram (EEG) signals limit classification performance and generalization. To address this issue, DWDANet, an EEG classification framework integrating multi-scale spatiotemporal feature modeling with dynamically weighted domain alignment, is developed. Multi-scale convolutions extracted spatiotemporal features, while a Transformer encoder modeled global dependencies. Multi-source domain alignment with dynamic weighting enhanced collaborative modeling across source domains and improved generalization. In within-subject evaluations, DWDANet achieved accuracies of 87.22%, 90.32%, and 68.82% on the BCI IV-2a, BCI IV-2b, and SHU-MI datasets, respectively. In cross-subject experiments, it achieved accuracies of 70.61%, 81.43%, and 64.38% on these datasets, respectively, outperforming other existing methods on all three. These results indicate that DWDANet could provide an effective approach to improving motor imagery EEG classification performance and cross-subject generalization.

Citation： KONG Xintang, ZHANG Pengwei, CHEN Jingxia. Multi-scale motor imagery electroencephalogram classification framework with dynamic weighted multi-source domain alignment. Journal of Biomedical Engineering, 2026, 43(5): 929-937. doi: 10.7507/1001-5515.202607017 Copy

1.	Otarbay Z, Kyzyrkanov A. Transfer learning for subject-independent motor imagery EEG classification using convolutional relational networks[J]. Front Neurosci, 2026, 19: 1691929.
2.	Vafaei E, Hosseini M. Transformers in EEG analysis: a review of architectures and applications in motor imagery, seizure, and emotion classification[J]. Sensors, 2025, 25(5): 1293.
3.	Zhao W, Jiang X, Zhang B, et al. CTNet: a convolutional transformer network for EEG-based motor imagery classification[J]. Sci Rep, 2024, 14: 20237.
4.	Roy Y, Banville H, Albuquerque I, et al. Deep learning-based electroencephalography analysis: a systematic review[J]. J Neural Eng, 2019, 16(5): 051001.
5.	Craik A, He Y, Contreras-Vidal J L. Deep learning for electroencephalogram (EEG) classification tasks: a review[J]. J Neural Eng, 2019, 16(3): 031001.
6.	Lotey T, Keserwani P, Dogra D P, et al. Feature reweighting for EEG-based motor imagery classification[PP/OL]. arXiv (2023-07-29) [2026-04-06]. https://arxiv.org/abs/2308.02515.
7.	肖楠, 李明愛. 基于動態多尺度卷積和多頭時間注意力的運動想象分類[J]. 生物醫學工程學雜志, 2025, 42(4): 678-685.
8.	Liao W, Miao Z, Liang S, et al. A composite improved attention convolutional network for motor imagery EEG classification[J]. Front Neurosci, 2025, 19: 1543508.
9.	Hamidi A, Kiani K. Motor imagery EEG signals classification using a Transformer-GCN approach[J]. Appl Soft Comput, 2025, 170: 112686.
10.	Liu M, Liu Y, Shi W, et al. EMPT: a sparsity transformer for EEG-based motor imagery recognition[J]. Front Neurosci, 2024, 18: 1366294.
11.	Zhang D, Li H, Xie J. Unsupervised and semi-supervised domain adaptation networks considering both global knowledge and prototype-based local class information for motor imagery classification[J]. Neural Netw, 2024, 180: 106497.
12.	Zhong X C, Wang Q, Liu D, et al. EEG-DG: a multi-source domain generalization framework for motor imagery EEG classification[J]. IEEE J Biomed Health Inform, 2025, 29(4): 2484-2495.
13.	齊琪, 李明愛. 基于時頻變換和黎曼流形的腦源空間運動想象域適應方法[J]. 生物醫學工程學雜志, 2026, 43(1): 87-96.
14.	賀嘉玲, 張鵬偉, 陳景霞. 融合動態閾值與增強反饋的半監督運動想象腦電分類方法[J]. 陜西科技大學學報, 2026, 44(2): 197-207.
15.	Liao W, Liu H, Wang W. Advancing BCI with a transformer-based model for motor imagery classification[J]. Sci Rep, 2025, 15: 23380.
16.	Altaheri H, Karray F, Karimi A H. Temporal convolutional transformer for EEG based motor imagery decoding[J]. Sci Rep, 2025, 15: 32959.
17.	Tan Y, Li B, Sun Z, et al. Multi-source self-guided domain adaptation framework for EEG-based emotion recognition[J]. Med Biol Eng Comput, 2026, 64(3): 861-871.
18.	Wang Z, Zhao M. Dynamic domain adaptive EEG emotion recognition based on multi-source selection[J]. Rev Sci Instrum, 2025, 96(1): 015103.
19.	Wu X, Ju X, Dai S, et al. Multi-source domain adaptation for EEG emotion recognition based on inter-domain sample hybridization[J]. Front Hum Neurosci, 2024, 18: 1464431.
20.	Gong Y, Shi K, Niu X, et al. Multi-source discriminant dynamic domain adaptation for cross-subject motor imagery EEG recognition[J]. IEEE J Biomed Health Inform, 2026, 30(4): 3556-3567.
21.	Brunner C, Leeb R, Müller-Putz G, et al. BCI Competition IV Dataset 2a[DB/OL]. Graz: Graz University of Technology, 2008 [2026-04-06]. https://www.bbci.de/competition/iv/.
22.	Brunner C, Leeb R, Müller-Putz G, et al. BCI Competition IV Dataset 2b[DB/OL]. Graz: Graz University of Technology, 2008 [2026-04-06]. https://www.bbci.de/competition/iv/.
23.	Ma J, Yang B, Qiu W, et al. A large EEG dataset for studying cross-session variability in motor imagery brain-computer interface[J]. Sci Data, 2022, 9: 531.
24.	Lawhern V J, Solon A J, Waytowich N R, et al. EEGNet: a compact convolutional neural network for EEG-based brain-computer interfaces[J]. J Neural Eng, 2018, 15(5): 056013.
25.	Schirrmeister R T, Springenberg J T, Fiederer L D J, et al. Deep learning with convolutional neural networks for EEG decoding and visualization[J]. Hum Brain Mapp, 2017, 38(11): 5391-5420.
26.	Zhao W, Zhang B, Zhou H, et al. Multi-scale convolutional transformer network for motor imagery brain-computer interface[J]. Sci Rep, 2025, 15: 12935.
27.	Li J, Shi J, Yu P, et al. Feature-aware domain invariant representation learning for EEG motor imagery decoding[J]. Sci Rep, 2025, 15: 10664.
28.	Keutayeva A, Fakhrutdinov N, Abibullaev B. Compact convolutional transformer for subject-independent motor imagery EEG-based BCIs[J]. Sci Rep, 2024, 14: 25775.
29.	Habashi A G, Azab A M, Eldawlatly S, et al. TFOC-Net: a short-time Fourier transform-based deep learning approach for enhancing cross-subject motor imagery classification[PP/OL]. arXiv (2025-07-03) [2026-09-10]. https://arxiv.org/abs/2507.02510.
30.	Qamar W U R, Abibullaev B. Multi-scale EEG feature decoding with Swin transformers for subject-independent motor imagery BCIs[J]. Sci Rep, 2026, 16: 2503.
31.	Van der Maaten L, Hinton G. Visualizing data using t-SNE[J]. J Mach Learn Res, 2008, 9(11): 2579-2605.
32.	Bach S, Binder A, Montavon G, et al. On pixel-wise explanations for non-linear classifier decisions by layer-wise relevance propagation[J]. PLoS One, 2015, 10(7): e0130140.
33.	Sturm I, Lapuschkin S, Samek W, et al. Interpretable deep neural networks for single-trial EEG classification[J]. J Neurosci Methods, 2016, 274: 141-145.
34.	Pfurtscheller G, Neuper C. Motor imagery activates primary sensorimotor area in humans[J]. Neurosci Lett, 1997, 239(2-3): 65-68.
35.	Pfurtscheller G, Neuper C. Motor imagery and direct brain-computer communication[J]. Proc IEEE, 2001, 89(7): 1123-1134.

1. Otarbay Z, Kyzyrkanov A. Transfer learning for subject-independent motor imagery EEG classification using convolutional relational networks[J]. Front Neurosci, 2026, 19: 1691929.
2. Vafaei E, Hosseini M. Transformers in EEG analysis: a review of architectures and applications in motor imagery, seizure, and emotion classification[J]. Sensors, 2025, 25(5): 1293.
3. Zhao W, Jiang X, Zhang B, et al. CTNet: a convolutional transformer network for EEG-based motor imagery classification[J]. Sci Rep, 2024, 14: 20237.
4. Roy Y, Banville H, Albuquerque I, et al. Deep learning-based electroencephalography analysis: a systematic review[J]. J Neural Eng, 2019, 16(5): 051001.
5. Craik A, He Y, Contreras-Vidal J L. Deep learning for electroencephalogram (EEG) classification tasks: a review[J]. J Neural Eng, 2019, 16(3): 031001.
6. Lotey T, Keserwani P, Dogra D P, et al. Feature reweighting for EEG-based motor imagery classification[PP/OL]. arXiv (2023-07-29) [2026-04-06]. https://arxiv.org/abs/2308.02515.
7. 肖楠, 李明愛. 基于動態多尺度卷積和多頭時間注意力的運動想象分類[J]. 生物醫學工程學雜志, 2025, 42(4): 678-685.
8. Liao W, Miao Z, Liang S, et al. A composite improved attention convolutional network for motor imagery EEG classification[J]. Front Neurosci, 2025, 19: 1543508.
9. Hamidi A, Kiani K. Motor imagery EEG signals classification using a Transformer-GCN approach[J]. Appl Soft Comput, 2025, 170: 112686.
10. Liu M, Liu Y, Shi W, et al. EMPT: a sparsity transformer for EEG-based motor imagery recognition[J]. Front Neurosci, 2024, 18: 1366294.
11. Zhang D, Li H, Xie J. Unsupervised and semi-supervised domain adaptation networks considering both global knowledge and prototype-based local class information for motor imagery classification[J]. Neural Netw, 2024, 180: 106497.
12. Zhong X C, Wang Q, Liu D, et al. EEG-DG: a multi-source domain generalization framework for motor imagery EEG classification[J]. IEEE J Biomed Health Inform, 2025, 29(4): 2484-2495.
13. 齊琪, 李明愛. 基于時頻變換和黎曼流形的腦源空間運動想象域適應方法[J]. 生物醫學工程學雜志, 2026, 43(1): 87-96.
14. 賀嘉玲, 張鵬偉, 陳景霞. 融合動態閾值與增強反饋的半監督運動想象腦電分類方法[J]. 陜西科技大學學報, 2026, 44(2): 197-207.
15. Liao W, Liu H, Wang W. Advancing BCI with a transformer-based model for motor imagery classification[J]. Sci Rep, 2025, 15: 23380.
16. Altaheri H, Karray F, Karimi A H. Temporal convolutional transformer for EEG based motor imagery decoding[J]. Sci Rep, 2025, 15: 32959.
17. Tan Y, Li B, Sun Z, et al. Multi-source self-guided domain adaptation framework for EEG-based emotion recognition[J]. Med Biol Eng Comput, 2026, 64(3): 861-871.
18. Wang Z, Zhao M. Dynamic domain adaptive EEG emotion recognition based on multi-source selection[J]. Rev Sci Instrum, 2025, 96(1): 015103.
19. Wu X, Ju X, Dai S, et al. Multi-source domain adaptation for EEG emotion recognition based on inter-domain sample hybridization[J]. Front Hum Neurosci, 2024, 18: 1464431.
20. Gong Y, Shi K, Niu X, et al. Multi-source discriminant dynamic domain adaptation for cross-subject motor imagery EEG recognition[J]. IEEE J Biomed Health Inform, 2026, 30(4): 3556-3567.
21. Brunner C, Leeb R, Müller-Putz G, et al. BCI Competition IV Dataset 2a[DB/OL]. Graz: Graz University of Technology, 2008 [2026-04-06]. https://www.bbci.de/competition/iv/.
22. Brunner C, Leeb R, Müller-Putz G, et al. BCI Competition IV Dataset 2b[DB/OL]. Graz: Graz University of Technology, 2008 [2026-04-06]. https://www.bbci.de/competition/iv/.
23. Ma J, Yang B, Qiu W, et al. A large EEG dataset for studying cross-session variability in motor imagery brain-computer interface[J]. Sci Data, 2022, 9: 531.
24. Lawhern V J, Solon A J, Waytowich N R, et al. EEGNet: a compact convolutional neural network for EEG-based brain-computer interfaces[J]. J Neural Eng, 2018, 15(5): 056013.
25. Schirrmeister R T, Springenberg J T, Fiederer L D J, et al. Deep learning with convolutional neural networks for EEG decoding and visualization[J]. Hum Brain Mapp, 2017, 38(11): 5391-5420.
26. Zhao W, Zhang B, Zhou H, et al. Multi-scale convolutional transformer network for motor imagery brain-computer interface[J]. Sci Rep, 2025, 15: 12935.
27. Li J, Shi J, Yu P, et al. Feature-aware domain invariant representation learning for EEG motor imagery decoding[J]. Sci Rep, 2025, 15: 10664.
28. Keutayeva A, Fakhrutdinov N, Abibullaev B. Compact convolutional transformer for subject-independent motor imagery EEG-based BCIs[J]. Sci Rep, 2024, 14: 25775.
29. Habashi A G, Azab A M, Eldawlatly S, et al. TFOC-Net: a short-time Fourier transform-based deep learning approach for enhancing cross-subject motor imagery classification[PP/OL]. arXiv (2025-07-03) [2026-09-10]. https://arxiv.org/abs/2507.02510.
30. Qamar W U R, Abibullaev B. Multi-scale EEG feature decoding with Swin transformers for subject-independent motor imagery BCIs[J]. Sci Rep, 2026, 16: 2503.
31. Van der Maaten L, Hinton G. Visualizing data using t-SNE[J]. J Mach Learn Res, 2008, 9(11): 2579-2605.
32. Bach S, Binder A, Montavon G, et al. On pixel-wise explanations for non-linear classifier decisions by layer-wise relevance propagation[J]. PLoS One, 2015, 10(7): e0130140.
33. Sturm I, Lapuschkin S, Samek W, et al. Interpretable deep neural networks for single-trial EEG classification[J]. J Neurosci Methods, 2016, 274: 141-145.
34. Pfurtscheller G, Neuper C. Motor imagery activates primary sensorimotor area in humans[J]. Neurosci Lett, 1997, 239(2-3): 65-68.
35. Pfurtscheller G, Neuper C. Motor imagery and direct brain-computer communication[J]. Proc IEEE, 2001, 89(7): 1123-1134.

Journal of Biomedical Engineering

Multi-scale motor imagery electroencephalogram classification framework with dynamic weighted multi-source domain alignment

Abstract Full text Figures/Tables Video References Cited by

Previous Article

Next Article

Format

Content