Analysis of Window-Delay Score for Data Augmentation Methods in Brain-Computer Interfaces

• João Stephan S. Maurício UFJF
• Marcelo M. Amorim UFJF
• Alex Borges UFJF
• Heder Bernardino UFJF
• Gabriel de Souza UFJF

DOI: https://doi.org/10.5753/eniac.2024.245225

Resumo

Post-stroke motor rehabilitation is a challenging problem in the medical field. Considering this, Brain-Computer Interfaces (BCI) have proven to obtain positive results, especially for chronic stroke. However, as electroencephalogram data collection for BCI can be challenging, Data Augmentation (DA) methods can reduce data collection and simplify training. This study proposes analyzing the temporal behavior of the accuracy instead of analyzing it in fixed intervals, as it is commonly done. Six DA methods and five classification models were evaluated for different scenarios. Results show Filter Bank Common Spatial Pattern is consistent while EEGNet peaks at 2.5 seconds. Sliding Window DA improves response time by 16% and enhances model robustness.

Palavras-chave: Electroencephalogram, Motor Imagery, Brain-Machine Interface, Medical applications

Referências

Ang, K. K., Chin, Z. Y., Wang, C., Guan, C., and Zhang, H. (2012). Filter bank common spatial pattern algorithm on bci competition iv datasets 2a and 2b. Frontiers in Neuroscience.

Choi, H., Park, J., and Yang, Y.-M. (2022). A novel quick-response eigenface analysis scheme for brain–computer interfaces. Sensors.

de Souza, G. H., Bernardino, H. S., and Vieira, A. B. (2021). Single electrode energy on clinical brain–computer interface challenge. Biomedical Signal Processing and Control.

de Souza, G. H., dos Santos, D. E., Bernardino, H., Vieira, A. B., and Motta, L. P. (2023). Window-delay analysis on eegnet. In Proceeding of 2023 10th International Conference on Soft Computing & Machine Intelligence.

Fahimi, F., Dosen, S., Ang, K. K., Mrachacz-Kersting, N., and Guan, C. (2021). Generative adversarial networks-based data augmentation for brain-computer interface. IEEE Transactions on Neural Networks and Learning Systems.

Faria, G., De Souza, G. H., Bernardino, H., Motta, L., and Vieira, A. (2022). Analyzing data augmentation methods for convolutional neural network-based brain-computer interfaces. In Proceedings of the International Joint Conference on Neural Networks.

Freer, D. and Yang, G.-Z. (2020). Data augmentation for self-paced motor imagery classification with c-lstm. Journal of Neural Engineering.

Huang, N. E., Shen, Z., Long, S. R., Wu, M. C., Snin, H. H., Zheng, Q., Yen, N.-C., Tung, C. C., and Liu, H. H. (1998). The empirical mode decomposition and the hubert spectrum for nonlinear and non-stationary time series analysis. In Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences.

Huang, W., Wang, L., Yan, Z., and Liu, Y. (2020). Classify motor imagery by a novel cnn with data augmentation. In Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS.

Kim, S.-J., Lee, D.-H., and Choi, Y.-W. (2023). Cropcat: Data augmentation for smoothing the feature distribution of eeg signals. In Proceedings of the International Winter Conference on Brain-Computer Interface, BCI.

Lashgari, E., Ott, J., Connelly, A., Baldi, P., and Maoz, U. (2021). An end-to-end cnn with attentional mechanism applied to raw eeg in a bci classification task. Journal of Neural Engineering.

Lawhern, V. J., Solon, A. J., Waytowich, N. R., Gordon, S. M., Hung, C. P., and Lance, B. J. (2018). Eegnet: a compact convolutional neural network for eeg-based brain–computer interfaces. Journal of Neural Engineering.

Li, B., Hou, Y., and Che, W. (2022a). Data augmentation approaches in natural language processing: A survey. AI Open.

Li, R., Wang, L., Suganthan, P., and Sourina, O. (2022b). Sample-based data augmentation based on electroencephalogram intrinsic characteristics. IEEE Journal of Biomedical and Health Informatics.

Lotte, F. (2015). Signal processing approaches to minimize or suppress calibration time in oscillatory activity-based brain-computer interfaces. Proceedings of the IEEE.

Luo, J., Wang, Y., Xu, R., Liu, G., Wang, X., and Gong, Y. (2021). Channel drop out: A simple way to prevent cnn from overfitting in motor imagery based bci. Communications in Computer and Information Science.

Mumuni, A. and Mumuni, F. (2022). Data augmentation: A comprehensive survey of modern approaches. Array.

Pacheco-Barrios, K., Giannoni-Luza, S., Navarro-Flores, A., Rebello-Sanchez, I., Parente, J., Balbuena, A., de Melo, P. S., Otiniano-Sifuentes, R., Rivera-Torrejón, O., Abanto, C., Alva-Diaz, C., Musolino, P. L., and Fregni, F. (2022). Burden of stroke and population-attributable fractions of risk factors in latin america and the caribbean. Journal of the American Heart Association.

Qin, C., Yang, R., Huang, M., Liu, W., and Wang, Z. (2023). Spatial variation generation algorithm for motor imagery data augmentation: Increasing the density of sample vicinity. IEEE Transactions on Neural Systems and Rehabilitation Engineering.

Tangermann, M., Müller, K.-R., Aertsen, A., Birbaumer, N., Braun, C., Brunner, C., Leeb, R., Mehring, C., Miller, K., Mueller-Putz, G., Nolte, G., Pfurtscheller, G., Preissl, H., Schalk, G., Schlögl, A., Vidaurre, C., Waldert, S., and Blankertz, B. (2012). Review of the bci competition iv. Frontiers in Neuroscience.

Wolpaw, J., Birbaumer, N., Heetderks, W., McFarland, D., Peckham, P., Schalk, G., Donchin, E., Quatrano, L., Robinson, C., and Vaughan, T. (2000). Brain-computer interface technology: a review of the first international meeting. IEEE Transactions on Rehabilitation Engineering.

Yang, L., Song, Y., Ma, K., and Xie, L. (2021). Motor imagery eeg decoding method based on a discriminative feature learning strategy. IEEE Transactions on Neural Systems and Rehabilitation Engineering.

Zhang, K., Xu, G., Han, Z., Ma, K., Zheng, X., Chen, L., Duan, N., and Zhang, S. (2020). Data augmentation for motor imagery signal classification based on a hybrid neural network. Sensors (Switzerland).