Characterizing Regularization Techniques for Spatial Filter Optimization in Oscillatory EEG Regression Problems
- 388 Downloads
We report on novel supervised algorithms for single-trial brain state decoding. Their reliability and robustness are essential to efficiently perform neurotechnological applications in closed-loop. When brain activity is assessed by multichannel recordings, spatial filters computed by the source power comodulation (SPoC) algorithm allow identifying oscillatory subspaces. They regress to a known continuous trial-wise variable reflecting, e.g. stimulus characteristics, cognitive processing or behavior. In small dataset scenarios, this supervised method tends to overfit to its training data as the involved recordings via electroencephalogram (EEG), magnetoencephalogram or local field potentials generally provide a low signal-to-noise ratio. To improve upon this, we propose and characterize three types of regularization techniques for SPoC: approaches using Tikhonov regularization (which requires model selection via cross-validation), combinations of Tikhonov regularization and covariance matrix normalization as well as strategies exploiting analytical covariance matrix shrinkage. All proposed techniques were evaluated both in a novel simulation framework and on real-world data. Based on the simulation findings, we saw our expectations fulfilled, that SPoC regularization generally reveals the largest benefit for small training sets and under severe label noise conditions. Relevant for practitioners, we derived operating ranges of regularization hyperparameters for cross-validation based approaches and offer open source code. Evaluating all methods additionally on real-world data, we observed an improved regression performance mainly for datasets from subjects with initially poor performance. With this proof-of-concept paper, we provided a generalizable regularization framework for SPoC which may serve as a starting point for implementing advanced techniques in the future.
KeywordsEEG bandpower Subspace decomposition Brain-computer interface Single trial analysis Source power comodulation Brain state decoding algorithm
This work was fully supported by BrainLinks-BrainTools Cluster of Excellence funded by the German Research Foundation (DFG), grant number EXC1086. For the data analysis, the authors acknowledge support by the state of Baden-Württemberg through bwHPC and the German Research Foundation (DFG) by grant no. INST 39/963-1 FUGG. Fabien Lotte received research support from the French National Research Agency with the REBEL project (grant ANR-15-CE23-0013-01) and the European Research Council with the BrainConquest project (grant ERC-2016-STG-714567). For parts of the data analysis, the Matlab-based BBCI toolbox was utilized (Blankertz et al. 2016). The authors declare that they have no conflict of interest.
- Arvaneh, M., Guan, C., Ang, K.K., Quek, C. (2013). Optimizing spatial filters by minimizing within-class dissimilarities in electroencephalogram-based brain-computer interface. IEEE Transactions on Neural Networks and Learning Systems, 24(4), 610–619. https://doi.org/10.1109/TNNLS.2013.2239310.CrossRefGoogle Scholar
- Bai, Z, & Silverstein, JW. (2009). Spectral analysis of large dimensional random matrices. Springer Science & Business Media.Google Scholar
- Bartz, D., & Müller, K.-R. (2014). Covariance shrinkage for autocorrelated data. In Ghahramani, Z., Welling, M., Cortes, C., Lawrence, N.D., Weinberger, K.Q. (Eds.) Advances in neural information processing systems, (Vol. 27 pp. 1592–1600): Curran Associates Inc.Google Scholar
- Blankertz, B., Acqualagna, L., Dähne, S, Haufe, S., Schultze-Kraft, M., Sturm, I., Ušćumlic, M., Wenzel, M.A., Curio, G., Müller, K.-R. (2016). The berlin brain-computer interface: progress beyond communication and control. Frontiers in Neuroscience, 10, 530. https://doi.org/10.3389/fnins.2016.00530.CrossRefGoogle Scholar
- Castaño-Candamil, J.S., Meinel, A., Dähne, S., Tangermann, M. (2015). Probing meaningfulness of oscillatory EEG components with bootstrapping, label noise and reduced training sets. In 2015 37th Annual international conference of the IEEE engineering in medicine and biology society (EMBC) (pp. 5159–5162). IEEE.Google Scholar
- Castaño-Candamil, S., Meinel, A., Tangermann, M. (2017). Post-hoc labeling of arbitrary EEG recordings for data-efficient evaluation of neural decoding methods. arXiv:171108208.
- Clerc, M., Bougrain, L., Lotte, F. (2016). Brain-computer interfaces 2: technology and applications. Wiley.Google Scholar
- Dähne, S., Meinecke, F.C., Haufe, S., Höhne, J, Tangermann, M., Müller, K.-R., Nikulin, V.V. (2014). SPoC: a novel framework for relating the amplitude of neuronal oscillations to behaviorally relevant parameters. NeuroImage, 86(0), 111–122. https://doi.org/10.1016/j.neuroimage.2013.07.079.CrossRefGoogle Scholar
- De Bie, T., Cristianini, N., Rosipal, R. (2005). Eigenproblems in pattern recognition. In Handbook of geometric computing (pp. 129–167). Springer.Google Scholar
- Farina, D., Jiang, N., Rehbaum, H., Holobar, A., Graimann, B., Dietl, H., Aszmann, O.C. (2014). The extraction of neural information from the surface EMG for the control of upper-limb prostheses: emerging avenues and challenges. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 22(4), 797–809. https://doi.org/10.1109/TNSRE.2014.2305111.CrossRefGoogle Scholar
- Farquhar, J., Hill, N., Lal, T.N., Schölkopf, B. (2006). Regularised CSP for sensor selection in BCI. In Proceedings of the 3rd international BCI workshop.Google Scholar
- Frey, J., Daniel, M., Hachet, M., Castet. J., Lotte, F. (2016). Framework for electroencephalography-based evaluation of user experience. InProcedings of CHI (pp. 2283–2294).Google Scholar
- Kenney, J.F. (2013). Mathematics of statistics. Toronto: D. Van Nostrand Company Inc. Princeton; New Jersey; London; New York,; Affiliated East-West Press Pvt-Ltd; New Delhi.Google Scholar
- Lotte, F., & Guan, C. (2010). Learning from other subjects helps reducing brain-computer interface calibration time. In 2010 IEEE International conference on acoustics, speech and signal processing (pp. 614–617). https://doi.org/10.1109/ICASSP.2010.5495183.
- Lu, H., Eng, H.L., Guan, C., Plataniotis, K.N., Venetsanopoulos, A.N. (2010). Regularized common spatial pattern with aggregation for EEG classification in small-sample setting. IEEE Transactions on Biomedical Engineering, 57(12), 2936–2946. https://doi.org/10.1109/TBME.2010.2082540.CrossRefGoogle Scholar
- Meinel, A., Lotte, F., Tangermann, M. (2017). Tikhonov regularization enhances EEG-based spatial filtering for single-trial regression. In Proceedings of the 7th Graz brain-computer interface conference 2017 (pp. 308-313). https://doi.org/10.3217/978-3-85125-533-1-57.
- Millán, J.d.R., Rupp, R., Mueller-Putz, G., Murray-Smith, R., Giugliemma, C., Tangermann, M., Vidaurre, C., Cincotti, F., Kübler, A, Leeb, R., Neuper, C., Müller, K.-R, Mattia, D. (2010). Combining brain–computer interfaces and assistive technologies: state-of-the-art and challenges. Frontiers in Neuroscience, 4, 161.Google Scholar
- Nicolae, I.E., Acqualagna, L., Blankertz, B. (2017). Assessing the depth of cognitive processing as the basis for potential user-state adaptation. Frontiers in Neuroscience, 11. https://doi.org/10.3389/fnins.2017.00548.
- Reuderink, B., & Poel, M. (2008). Robustness of the common spatial patterns algorithm in the BCI-pipeline. Tech rep. HMI, University of Twente.Google Scholar
- Tikhonov, A.N. (1963). Regularization of incorrectly posed problems. Soviet Mathematics Doklady, 4, 1624–1627.Google Scholar