Abstract
Recently, imagined speech has become a subject of study due to its potential as an intuitive communication system. It involves registering neural responses generated by mental speaking without moving the articulators. Although it may not perform as well as other paradigms, it has multiclass scalability, making it suitable for building extensible BCI systems. Hence, our study revolves around this intuitive paradigm that decodes human speech imagery from EEG signals using Riemannian geometry and a recently introduced covariance estimation method that is based on the concept of Approximate Joint Diagonalization (AJD). The employed methodological framework approach sets its grounds on neuroscientifically sound theories and is being validated on a competition dataset consisting of multichannel EEG trials from five different imagined prompts. Despite its simplicity, the presented methodology achieves over 70% accuracy in some classes, which is on par with State-of-the-Art performance on the dataset. Our methodology performs significantly better in monosyllabic prompts (i.e., ‘yes’ and ‘stop’) which may constitute it more appropriate in immediate-response critical BCI applications. Moreover, the conducted preliminary analysis that was used for sensor selection and onset detection sheds light into the understudied neural phenomena of imagined speech as captured in EEG signals.
*This work was supported by the NeuroMkt project, co-financed by the European Regional Development Fund of the EU and Greek National Funds through the Operational Program “Competitiveness, Entrepreneurship and Innovation”, under RESEARCH CREATE INNOVATE (T2EDK-03661).
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Abdulkader, S.N., Atia, A., Mostafa, M.S.M.: Brain computer interfacing: applications and challenges. Egypt. Informa. J. 16(2), 213–230 (2015)
Bakhshali, M.A., Khademi, M., Ebrahimi-Moghadam, A., Moghimi, S.: EEG signal classification of imagined speech based on Riemannian distance of correntropy spectral density. Biomed. Signal Process. Control 59, 101899 (2020)
Cardoso, J.F., Souloumiac, A.: Jacobi angles for simultaneous diagonalization. SIAM J. Matrix Anal. Appl. 17(1), 161–164 (1996)
Chengaiyan, S., Retnapandian, A.S., Anandan, K.: Identification of vowels in consonant-vowel-consonant words from speech imagery based EEG signals. Cogn. Neurodyn. 14(1), 1–19 (2020)
Chu, Y., et al.: Decoding multiclass motor imagery EEG from the same upper limb by combining Riemannian geometry features and partial least squares regression. J. Neural Eng. 17(4), 046029 (2020)
Congedo, M., Barachant, A., Bhatia, R.: Riemannian geometry for EEG-based brain-computer interfaces; a primer and a review. Brain-Comput. Interfaces 4(3), 155–174 (2017)
Cooney, C., Korik, A., Folli, R., Coyle, D.: Evaluation of hyperparameter optimization in machine and deep learning methods for decoding imagined speech EEG. Sensors 20(16), 4629 (2020)
García-Salinas, J.S., Villaseñor-Pineda, L., Reyes-García, C.A., Torres-García, A.A.: Transfer learning in imagined speech EEG-based BCIs. Biomed. Signal Process. Control 50, 151–157 (2019)
Georgiadis, K., Laskaris, N., Nikolopoulos, S., Kompatsiaris, I.: Connectivity steered graph Fourier transform for motor imagery BCI decoding. J. Neural Eng. 16(5), 056021 (2019)
Georgiadis, K., Adamos, D.A., Nikolopoulos, S., Laskaris, N., Kompatsiaris, I.: A graph-theoretic sensor-selection scheme for covariance-based motor imagery (mi) decoding. In: 2020 28th European Signal Processing Conference (EUSIPCO), pp. 1234–1238. IEEE (2021)
Georgiadis, K., Kalaganis, F.P., Oikonomou, V.P., Nikolopoulos, S., Laskaris, N.A., Kompatsiaris, I.: \(^R\)NeuMark: a Riemannian EEG analysis framework for neuromarketing. Brain Inform. 9(1), 22 (2022)
Herff, C., de Pesters, A., Heger, D., Brunner, P., Schalk, G., Schultz, T.: Towards continuous speech recognition for BCI. Brain-Comput. Interface Res. State Art Summary 5, 21–29 (2017)
Hyvärinen, A.: Survey on independent component analysis (1999)
Jeong, J.H., et al.: 2020 international brain-computer interface competition: a review. Front. Hum. Neurosci. 16, 898300 (2022)
Kalaganis, F.P., Laskaris, N.A., Chatzilari, E., Nikolopoulos, S., Kompatsiaris, I.: A Riemannian geometry approach to reduced and discriminative covariance estimation in brain computer interfaces. IEEE Trans. Biomed. Eng. 67(1), 245–255 (2019)
Kalaganis, F.P., Laskaris, N.A., Oikonomou, V.P., Nikopolopoulos, S., Kompatsiaris, I.: Revisiting Riemannian geometry-based EEG decoding through approximate joint diagonalization. J. Neural Eng. 19(6), 066030 (2022)
Lee, S.H., Lee, M., Jeong, J.H., Lee, S.W.: Towards an EEG-based intuitive BCI communication system using imagined speech and visual imagery. In: 2019 IEEE International Conference on Systems, Man and Cybernetics (SMC), pp. 4409–4414. IEEE (2019)
Lee, Y.E., Lee, S.H.: EEG-transformer: self-attention from transformer architecture for decoding EEG of imagined speech. In: 2022 10th International Winter Conference on Brain-Computer Interface (BCI), pp. 1–4. IEEE (2022)
Li, F., Xia, Y., Wang, F., Zhang, D., Li, X., He, F.: Transfer learning algorithm of P300-EEG signal based on XDAWN spatial filter and Riemannian geometry classifier. Appl. Sci. 10(5), 1804 (2020)
Lopez-Bernal, D., Balderas, D., Ponce, P., Molina, A.: A state-of-the-art review of EEG-based imagined speech decoding. Front. Hum. Neurosci. 16, 867281 (2022)
Müller, K.R., Vigario, R., Meinecke, F., Ziehe, A.: Blind source separation techniques for decomposing event-related brain signals. Int. J. Bifurcat. Chaos 14(02), 773–791 (2004)
Nguyen, C.H., Karavas, G.K., Artemiadis, P.: Inferring imagined speech using EEG signals: a new approach using Riemannian manifold features. J. Neural Eng. 15(1), 016002 (2017)
Pennec, X., Fillard, P., Ayache, N.: A Riemannian framework for tensor computing. Int. J. Comput. Vision 66(1), 41–66 (2006)
Proix, T., et al.: Imagined speech can be decoded from low-and cross-frequency intracranial EEG features. Nat. Commun. 13(1), 48 (2022)
Si, X., Li, S., Xiang, S., Yu, J., Ming, D.: Imagined speech increases the hemodynamic response and functional connectivity of the dorsal motor cortex. J. Neural Eng. 18(5), 056048 (2021)
Ziehe, A.: Blind source separation based on joint diagonalization of matrices with applications in biomedical signal processing. Ph.D. thesis, Universität Potsdam (2005)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Kalaganis, F.P., Georgiadis, K., Oikonomou, V.P., Nikolopoulos, S., Laskaris, N.A., Kompatsiaris, I. (2023). Exploiting Approximate Joint Diagonalization for Covariance Estimation in Imagined Speech Decoding. In: Liu, F., Zhang, Y., Kuai, H., Stephen, E.P., Wang, H. (eds) Brain Informatics. BI 2023. Lecture Notes in Computer Science(), vol 13974. Springer, Cham. https://doi.org/10.1007/978-3-031-43075-6_35
Download citation
DOI: https://doi.org/10.1007/978-3-031-43075-6_35
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-43074-9
Online ISBN: 978-3-031-43075-6
eBook Packages: Computer ScienceComputer Science (R0)