Subject-Independent Multi-Class Motor Imagery EEG Classification Using Hybrid CNN–BiLSTM with Channel Attention
DOI:
https://doi.org/10.31185/wjes.Vol14.Iss3.938Keywords:
Motor imagery EEG, subject-independent classification, brain-computer interface, CNN, BiLSTMAbstract
The classification of MI-EEG is complex due to the noise, non-stationarity and inter-subject variation of these signals. Some of these challenges are magnified in subject-agnostic settings, where a model must generalise to an unseen user without subject-specific tuning. This study presents a hybrid deep-learning architecture that combines convolutional neural networks (CNNs), bidirectional long short-term memory (BiLSTM), and squeeze-and-excitation feature-channel attention for subject-independent four-class MI-EEG classification. The CNN extracts local spatiotemporal representations, the BiLSTM models forward and backward temporal dependencies, and the attention module recalibrates informative feature maps before classification. The framework was evaluated on BCI Competition IV Dataset 2a using leave-one-subject-out cross-validation across nine subjects. It achieved a mean accuracy of 69.75% and Cohen’s kappa of 0.5967, exceeding EEGNet by 6.94 percentage points and ShallowConvNet by 9.72 percentage points. Ablation analysis showed that removing attention reduced accuracy by 2.47 percentage points. Although the proposed model has higher computational cost and latency than EEGNet, it provides a favorable accuracy–complexity trade-off for subject-independent multi-class BCI applications in which decoding performance is prioritized over minimal latency.
References
[1] D. J. McFarland and J. R. Wolpaw, “EEG-based brain-computer interfaces,” Current Opinion in Biomedical Engineering, vol. 4, pp. 194–200, 2017. https://doi.org/10.1016/j.cobme.2017.11.004 DOI: https://doi.org/10.1016/j.cobme.2017.11.004
[2] M. A. Khan, R. Das, H. K. Iversen, and S. Puthusserypady, “Review on motor imagery based BCI systems for upper limb post-stroke neurorehabilitation: From designing to application,” Computers in Biology and Medicine, vol. 123, art. no. 103843, 2020. https://doi.org/10.1016/j.compbiomed.2020.103843 DOI: https://doi.org/10.1016/j.compbiomed.2020.103843
[3] X. Wang, V. Liesaputra, Z. Liu, Y. Wang, and Z. Huang, “An in-depth survey on deep learning-based motor imagery electroencephalogram (EEG) classification,” Artificial Intelligence in Medicine, vol. 147, art. no. 102738, 2024. https://doi.org/10.1016/j.artmed.2023.102738 DOI: https://doi.org/10.1016/j.artmed.2023.102738
[4] H. Zhang et al., “Subject-independent EEG classification based on a hybrid neural network,” Frontiers in Neuroscience, vol. 17, art. no. 1124089, 2023. https://doi.org/10.3389/fnins.2023.1124089 DOI: https://doi.org/10.3389/fnins.2023.1124089
[5] O.-Y. Kwon, M.-H. Lee, C. Guan, and S.-W. Lee, “Subject-independent brain-computer interfaces based on deep convolutional neural networks,” IEEE Transactions on Neural Networks and Learning Systems, vol. 31, no. 10, pp. 3839–3852, 2020. https://doi.org/10.1109/TNNLS.2019.2946869 DOI: https://doi.org/10.1109/TNNLS.2019.2946869
[6] M. Tangermann et al., “Review of the BCI Competition IV,” Frontiers in Neuroscience, vol. 6, art. no. 55, 2012. https://doi.org/10.3389/fnins.2012.00055 DOI: https://doi.org/10.3389/fnins.2012.00055
[7] R. T. Schirrmeister et al., “Deep learning with convolutional neural networks for EEG decoding and visualization,” Human Brain Mapping, vol. 38, no. 11, pp. 5391–5420, 2017. https://doi.org/10.1002/hbm.23730 DOI: https://doi.org/10.1002/hbm.23730
[8] V. J. Lawhern, A. J. Solon, N. R. Waytowich, S. M. Gordon, C. P. Hung, and B. J. Lance, “EEGNet: A compact convolutional neural network for EEG-based brain-computer interfaces,” Journal of Neural Engineering, vol. 15, no. 5, art. no. 056013, 2018. https://doi.org/10.1088/1741-2552/aace8c DOI: https://doi.org/10.1088/1741-2552/aace8c
[9] Y. Hou et al., “Deep feature mining via the attention-based bidirectional long short-term memory graph convolutional neural network for human motor imagery recognition,” Frontiers in Bioengineering and Biotechnology, vol. 9, art. no. 706229, 2022. https://doi.org/10.3389/fbioe.2021.706229 DOI: https://doi.org/10.3389/fbioe.2021.706229
[10] A. A. Al Shiam, M. K. I. Molla, M. R. Islam, and T. Tanaka, “Motor imagery classification using effective channel selection of multichannel EEG,” Brain Sciences, vol. 14, no. 5, art. no. 462, 2024. https://doi.org/10.3390/brainsci14050462 DOI: https://doi.org/10.3390/brainsci14050462
[11] X. Tang, C. Yang, X. Sun, M. Zou, and H. Wang, “Motor imagery EEG decoding based on multi-scale hybrid networks and feature enhancement,” IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 31, pp. 1208–1218, 2023. https://doi.org/10.1109/TNSRE.2023.3242280 DOI: https://doi.org/10.1109/TNSRE.2023.3242280
[12] Y. Zheng, S. Wu, J. Chen, Q. Yao, and S. Zheng, “Cross-subject motor imagery electroencephalogram decoding with domain generalization,” Bioengineering, vol. 12, no. 5, art. no. 495, 2025. https://doi.org/10.3390/bioengineering12050495 DOI: https://doi.org/10.3390/bioengineering12050495
[13] Z. Otarbay et al., “Transfer learning for subject-independent motor imagery EEG classification using convolutional relational networks,” Frontiers in Neuroscience, vol. 19, art. no. 1691929, 2026. https://doi.org/10.3389/fnins.2025.1691929 DOI: https://doi.org/10.3389/fnins.2025.1691929
[14] H. Li, J. Liu, and J. Li, “Cross-subject motor imagery EEG signal classification based on meta-transfer learning,” Computer Methods in Biomechanics and Biomedical Engineering, pp. 1–12, 2026. https://doi.org/10.1080/10255842.2026.2626477 DOI: https://doi.org/10.1080/10255842.2026.2626477
[15] X. Jiang et al., “Cross-subject EEG decoding with spatiotemporal transformers for zero-calibration brain–computer interfaces,” Alexandria Engineering Journal, vol. 143, pp. 13–25, 2026. https://doi.org/10.1016/j.aej.2026.04.006 DOI: https://doi.org/10.1016/j.aej.2026.04.006
[16] T. Lotey, P. Keserwani, D. P. Dogra, and P. P. Roy, “Feature reweighting for EEG-based motor imagery classification,” Biomedical Signal Processing and Control, vol. 110, pt. A, art. no. 108215, 2025. https://doi.org/10.1016/j.bspc.2025.108215 DOI: https://doi.org/10.1016/j.bspc.2025.108215
[17] J. Hu, L. Shen, and G. Sun, “Squeeze-and-Excitation Networks,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 7132–7141, 2018. https://doi.org/10.1109/CVPR.2018.00745 DOI: https://doi.org/10.1109/CVPR.2018.00745
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Saif Rasool
This work is licensed under a Creative Commons Attribution 4.0 International License.
