Structural Feature Selection in Common Spatial Patterns Using Adaptive Sparse Group Lasso

Abstract

With the advancement of brain–computer interfaces (BCI), motor imagery (MI) electroencephalogram (EEG) decoding can greatly benefit from spatial filtering features derived from common spatial patterns (CSP). However, CSP-based features often exhibit high redundancy and intersubject variability. These limitations make the feature selection methods based on sparse learning difficult to effectively balance the heterogeneous contributions of different temporal and spatial components. Moreover, these models tend to prioritise features with larger coefficients, potentially overlooking intrinsic feature importance and compromising the quality of the selected feature subset. To address these issues, we propose an Adaptive Sparse Group Lasso (ASGL) method for structured feature selection, designed to enhance discriminative CSP features whilst suppressing irrelevant components. The proposed method partitions EEG signals into consecutive segments using a sliding window, treating each as a separate feature group. Benefiting from this, the importance of features at both the group level and the within-group level can be effectively quantified through mutual information and copula mutual information, thereby assigning adaptive weights for selective penalisation within the model. This weight construction strategy preserves important features from relevant time intervals and frequency bands. The resulting optimization problem is solved efficiently via the alternating direction method of multipliers (ADMM). Evaluations on simulated and real-world datasets demonstrate that the proposed ASGL outperforms existing methods.

Keywords

adaptive sparse group lasso / brain-computer interfaces / common spatial patterns / motor imagery EEG / structural feature selection

Acknowledgements

This work is supported by grants from the Henan Province Science Foundation of Excellent Young Scholars (Grant 242300421171), the National Natural Science Foundation of China (Grants 62106066, 62506109 and 62576128) and the Zhejiang Provincial Natural Science Foundation of China under (Grant LMS26F020035).

Disclosure

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

The datasets generated and analysed during the current study are available from the corresponding author upon reasonable request.

References

[1] A. Tiwari, “A Logistic Binary Jaya Optimization-Based Channel Selection Scheme for Motor-Imagery Classification in Brain-Computer Interface,” Expert Systems With Applications 223 (2023): 119921, https://doi.org/10.1016/j.eswa.2023.119921.

[2] Ç Yılmaz and B. Yilmaz, “Advancements in Image Feature-Based Classification of Motor Imagery EEG Data: A Comprehensive Review,” Traitement du Signal 40, no. 5 (2023), https://doi.org/10.18280/ts.400507.

[3] S. Phadikar, N. Sinha, and R. Ghosh, “Unsupervised Feature Extraction With Autoencoders for EEG Based Multiclass Motor Imagery BCI,” Expert Systems With Applications 213 (2023): 118901, https://doi.org/10.1016/j.eswa.2022.118901.

[4] C. Gao, W. Liu, and X. Yang, “Convolutional Neural Network and Riemannian Geometry Hybrid Approach for Motor Imagery Classification,” Neurocomputing 507 (2022): 180-190, https://doi.org/10.1016/j.neucom.2022.08.024.

[5] B. Blankertz, G. Dornhege, M. Krauledat, K. R. Müller, and G. Curio, “The Non-Invasive Berlin Brain-Computer Interface: Fast Acquisition of Effective Performance in Untrained Subjects,” NeuroImage 37, no. 2 (2007): 539-550, https://doi.org/10.1016/j.neuroimage.2007.01.051.

[6] B. Blankertz, F. Losch, M. Krauledat, G. Dornhege, G. Curio, and K. R. Müller, “The Berlin Brain-Computer Interface: Accurate Performance From First-Session in BCI-Naive Subjects,” IEEE Transactions on Biomedical Engineering 55, no. 10 (2008): 2452-2462, https://doi.org/10.1109/TBME.2008.923152.

[7] R. Tang, Z. Li, and X. Xie, “Motor Imagery EEG Signal Classification Using Upper Triangle Filter Bank Auto-Encode Method,” Biomedical Signal Processing and Control 68 (2021): 102608, https://doi.org/10.1016/j.bspc.2021.102608.

[8] J. X. Mi, R. F. Li, K. Liu, and W. Li, “Exploring Multi-Scale Time Group for Common Spatial Pattern Feature Based Motor Imagery EEG Classification,” Biomedical Signal Processing and Control 112 (2026): 108591, https://doi.org/10.1016/j.bspc.2025.108591.

[9] K. K. Ang, Z. Y. Chin, H. Zhang, and C. Guan, “Filter Bank Common Spatial Pattern (FBCSP) in Brain-Computer Interface,” in IEEE, (2008), 2390-2397.

[10] Y. Zhang, G. Zhou, J. Jin, X. Wang, and A. Cichocki, “Optimizing Spatial Patterns With Sparse Filter Bands for Motor-Imagery Based Brain-Computer Interface,” Journal of Neuroscience Methods 255 (2015): 85-91, https://doi.org/10.1016/j.jneumeth.2015.08.004.

[11] B. Lu, F. Wang, S. Wang, J. Chen, G. Wen, and R. Fu, “Improvement of Motor Imagery Electroencephalogram Decoding by Iterative Weighted Sparse-Group Lasso,” Expert Systems With Applications 238 (2024): 122286, https://doi.org/10.1016/j.eswa.2023.122286.

[12] jT. Luo, “Selective Regularized Spatial Features Representation Learning for Motor Imagery EEG Based on Alternating Cascaded Model,” Applied Soft Computing 165 (2024): 112087, https://doi.org/10.1016/j.asoc.2024.112087.

[13] J. Yang, Z. Wang, G. Wang, Y. Liu, Y. He, and D. Wu, “OSFS-Vague: Online Streaming Feature Selection Algorithm Based on Vague Set,” CAAI Transactions on Intelligence Technology 9, no. 6 (2024): 1451-1466, https://doi.org/10.1049/cit2.12327.

[14] N. Singh Malan and S. Sharma, “Time Window and Frequency Band Optimization Using Regularized Neighbourhood Component Analysis for Multi-View Motor Imagery EEG Classification,” Biomedical Signal Processing and Control 67 (2021): 102550, https://doi.org/10.1016/j.bspc.2021.102550.

[15] M. Radman, A. Chaibakhsh, N. Nariman-zadeh, and H. He, “Feature Fusion for Improving Performance of Motor Imagery Brain-Computer Interface System,” Biomedical Signal Processing and Control 68 (2021): 102763, https://doi.org/10.1016/j.bspc.2021.102763.

[16] J. Shi, Y. Chen, A. A. Heidari, Z. Cai, H. Chen, and G. Liang, “Topological Search and Gradient Descent Boosted Runge-Kutta Optimiser With Application to Engineering Design and Feature Selection,” CAAI Transactions on Intelligence Technology 10, no. 2 (2025): 557-614, https://doi.org/10.1049/cit2.12387.

[17] Y. Zhang, Y. Hu, X. Gao, et al., An Embedded Vertical-Federated Feature Selection Algorithm Based on Particle Swarm Optimisation,” CAAI Transactions on Intelligence Technology 8, no. 3 (2023): 734-754, https://doi.org/10.1049/cit2.12122.

[18] Y. Jiao, Y. Zhang, X. Chen, et al., Sparse Group Representation Model for Motor Imagery EEG Classification,” IEEE Journal of Biomedical and Health Informatics 23, no. 2 (2019): 631-641, https://doi.org/10.1109/JBHI.2018.2832538.

[19] Y. Zhang, C. S. Nam, G. Zhou, J. Jin, X. Wang, and A. Cichocki, “Temporally Constrained Sparse Group Spatial Patterns for Motor Imagery BCI,” IEEE Transactions on Cybernetics 49, no. 9 (2019): 3322-3332, https://doi.org/10.1109/TCYB.2018.2841847.

[20] H. Zou, “The Adaptive Lasso and Its Oracle Properties,” Journal of the American Statistical Association 101, no. 476 (2006): 1418-1429, https://doi.org/10.1198/016214506000000735.

[21] J. Li, W. Dong, and D. Meng, “Grouped Gene Selection of Cancer via Adaptive Sparse Group Lasso Based on Conditional Mutual Information,” IEEE/ACM Transactions on Computational Biology and Bioinformatics 15, no. 6 (2018): 2028-2038, https://doi.org/10.1109/TCBB.2017.2761871.

[22] M. Sehhati, A. Mehridehnavi, H. Rabbani, and M. Pourhossein, “Stable Gene Signature Selection for Prediction of Breast Cancer Recurrence Using Joint Mutual Information,” IEEE/ACM Transactions on Computational Biology and Bioinformatics 12, no. 6 (2015): 1440-1448, https://doi.org/10.1109/TCBB.2015.2407407.

[23] J. Meng, L. Yao, X. Sheng, D. Zhang, and X. Zhu, “Simultaneously Optimizing Spatial Spectral Features Based on Mutual Information for EEG Classification,” IEEE Transactions on Biomedical Engineering 62, no. 1 (2015): 227-240, https://doi.org/10.1109/TBME.2014.2345458.

[24] K. Che, X. Chen, M. Guo, C. Wang, and X. Liu, “Genetic Variants Detection Based on Weighted Sparse Group Lasso,” Frontiers in Genetics 11 (2020): 155, https://doi.org/10.3389/fgene.2020.00155.

[25] Y. Wang, X. Li, and R. Ruiz, “Weighted General Group Lasso for Gene Selection in Cancer Classification,” IEEE Transactions on Cybernetics 49, no. 8 (2019): 2860-2873, https://doi.org/10.1109/TCYB.2018.2829811.

[26] X. Song, K. Liang, and J. Li, “WGRLR: A Weighted Group Regularized Logistic Regression for Cancer Diagnosis and Gene Selection,” IEEE/ACM Transactions on Computational Biology and Bioinformatics 20, no. 2 (2023): 1563-1573, https://doi.org/10.1109/TCBB.2022.3203167.

[27] S. Boyd, N. Parikh, E. Chu, B. Peleato, and J. Eckstein, “Distributed Optimization and Statistical Learning via the Alternating Direction Method of Multipliers,” Foundations and Trends® in Machine Learning 3, no. 1 (2011): 1-122, https://doi.org/10.1561/2200000016.

[28] 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 15, no. 5 (2018): 056013, https://doi.org/10.1088/1741-2552/aace8c.

[29] R. T. Schirrmeister, J. T. Springenberg, L. D. J. Fiederer, et al., Deep Learning With Convolutional Neural Networks for EEG Decoding and Visualization,” Human Brain Mapping 38, no. 11 (2017): 5391-5420, https://doi.org/10.1002/hbm.23730.

[30] H. Peng, F. Long, and C. Ding, “Feature Selection Based on Mutual Information Criteria of Max-Dependency, Max-Relevance, and Min-Redundancy,” IEEE Transactions on Pattern Analysis and Machine Intelligence 27, no. 8 (2005): 1226-1238, https://doi.org/10.1109/TPAMI.2005.159.

[31] Q. Han, L. Hu, and W. Gao, “Feature Relevance and Redundancy Coefficients for Multi-View Multi-Label Feature Selection,” Information Sciences 652 (2024): 119747, https://doi.org/10.1016/j.ins.2023.119747.

[32] S. Lall, D. Sinha, A. Ghosh, D. Sengupta, and S. Bandyopadhyay, “Stable Feature Selection Using Copula Based Mutual Information,” Pattern Recognition 112 (2021): 107697, https://doi.org/10.1016/j.patcog.2020.107697.

[33] Q. Zhong, J. Shang, Q. Ren, F. Li, C. N. Jiao, and J. X. Liu, “FSCME: A Feature Selection Method Combining Copula Correlation and Maximal Information Coefficient by Entropy Weights,” IEEE Journal of Biomedical and Health Informatics 28, no. 9 (2024): 5638-5648, https://doi.org/10.1109/JBHI.2024.3409628.

[34] F. Ye, J. Ding, K. Chen, and X. Xi, “Investigation of Corticomuscular Functional Coupling During Hand Movements Using Vine Copula,” Brain Sciences 12, no. 6 (2022): 754, https://doi.org/10.3390/brainsci12060754.

[35] R. Tibshirani, “Regression Shrinkage and Selection via the Lasso,” Journal of the Royal Statistical Society: Series B 58, no. 1 (2018): 267-288, https://doi.org/10.1111/j.2517-6161.1996.tb02080.x.

[36] M. Yuan and Y. Lin, “Model Selection and Estimation in Regression With Grouped Variables,” Journal of the Royal Statistical Society Series B: Statistical Methodology 68, no. 1 (2005): 49-67, https://doi.org/10.1111/j.1467-9868.2005.00532.x.

[37] L. Yang, S. Ding, F. Zhou, X. Yang, and Y. Xiao, “Robust EEG Feature Learning Model Based on an Adaptive Weight and Pairwise-Fused LASSO,” Biomedical Signal Processing and Control 68 (2021): 102728, https://doi.org/10.1016/j.bspc.2021.102728.

[38] X. Wang, H. Shangguan, F. Huang, S. Wu, and W. Jia, “MEL: Efficient Multi-Task Evolutionary Learning for High-Dimensional Feature Selection,” IEEE Transactions on Knowledge and Data Engineering 36, no. 8 (2024): 4020-4033, https://doi.org/10.1109/TKDE.2024.3366333.

[39] B. Blankertz, R. Tomioka, S. Lemm, M. Kawanabe, and Kr Muller, “Optimizing Spatial Filters for Robust EEG Single-Trial Analysis,” IEEE Signal Processing Magazine 25, no. 1 (2008): 41-56, https://doi.org/10.1109/MSP.2008.4408441.

[40] M. Tangermann, K. R. Müller, A. Aertsen, et al., Review of the BCI Competition IV,” Frontiers in Neuroscience 6 (2012): 55, https://doi.org/10.3389/fnins.2012.00055.

[41] E. Dagdevir and M. Tokmakci, “Determination of Effective Signal Processing Stages for Brain Computer Interface on BCI Competition IV Data Set 2b: A Review Study,” IETE Journal of Research 69, no. 6 (2023): 3144-3155, https://doi.org/10.1080/03772063.2021.1914204.

[42] G. Pfurtscheller and C. Neuper, “Motor Imagery and Direct Brain-Computer Communication,” Proceedings of the IEEE 89, no. 7 (2001): 1123-1134, https://doi.org/10.1109/5.939829.