Optimizing Spatial filters for Robust EEG Single-Trial Analysis
The DOI matches its Crossref record; 59 of the 62 references it cites were found in a registry or an index.
Abstract
Due to the volume conduction multichannel electroencephalogram (EEG) recordings give a rather blurred image of brain activity. Therefore spatial filters are extremely useful in single-trial analysis in order to improve the signal-to-noise ratio. There are powerful methods from machine learning and signal processing that permit the optimization of spatio-temporal filters for each subject in a data dependent fashion beyond the fixed filters based on the sensor geometry, e.g., Laplacians. Here we elucidate the theoretical background of the common spatial pattern (CSP) algorithm, a popular method in brain-computer interface (BCD research. Apart from reviewing several variants of the basic algorithm, we reveal tricks of the trade for achieving a powerful CSP performance, briefly elaborate on theoretical aspects of CSP, and demonstrate the application of CSP-type preprocessing in our studies of the Berlin BCI (BBCI) project.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“Unmixing of these global patterns using a number of spatial filters is therefore typically applied to the whole set of relevant electrodes as a basic step in many successful examples of EEG decoding [Ang et al., 2008; Blankertz et al., 2008; Rivet et al., 2009].”
From Deep learning with convolutional neural networks for EEG decoding and visualization (2017) · cites this paper as Blankertz et al., 2008
Citations
Cites
The Elements of Statistical Learning
Hastie, T., Tibshirani, R., & Friedman, J. (2009) · Springer Series in Statistics · doi:10.1007/978-0-387-84858-7 · Cited by 25,970
Advances in Neural Information Processing Systems 19
(2007) · The MIT Press eBooks · doi:10.52202/085713 · Cited by 15,539
The Elements of Statistical Learning: Data Mining, Inference, and Prediction
John Lu, Z. Q. (2010) · Journal of the Royal Statistical Society Series A: Statistics in Society · doi:10.1111/j.1467-985x.2010.00646_6.x · Cited by 12,390
Introduction to Statistical Pattern Recognition
(1990) · Elsevier eBooks · doi:10.1016/c2009-0-27872-x · Cited by 11,031
Brain–computer interfaces for communication and control
Wolpaw, J. R., Birbaumer, N., McFarland, D. J., Pfurtscheller, G., & Vaughan, T. M. (2002) · Clinical Neurophysiology · doi:10.1016/s1388-2457(02)00057-3 · Cited by 7,927
The Fourier Transform and Its Applications
Bracewell, R., & Kahn, P. B. (1966) · American Journal of Physics · doi:10.1119/1.1973431 · Cited by 5,737
Über das Elektrenkephalogramm des Menschen
Berger, H. (1929) · Archiv für Psychiatrie und Nervenkrankheiten · doi:10.1007/bf01797193 · Cited by 4,235
Introduction to statistical pattern recognition (2nd ed.)
Fukunaga, K. (1990) · Academic Press Professional, Inc. eBooks · Cited by 3,797
An introduction to kernel-based learning algorithms
Muller, K.-R., Mika, S., Ratsch, G., Tsuda, K., & Scholkopf, B. (2001) · IEEE Transactions on Neural Networks · doi:10.1109/72.914517 · Cited by 3,473
The Fourier Transform and Its Applications
Bracewell (1999) · Cited by 2,102
Research works printed but not linked: 7
- hill, 2007, An evidence-based approach to optimizing feature extraction in EEG signal classification
- dornhege, 2006, Increasing information transfer rates for brain-computer interfacing
- 2007, Toward Brain-Computer Interfacing
- blankertz, 2007, Universal Access in HCI, 4555, 759
- blankertz, 2008, Advances in Neural Information Processing Systems 20
- tomioka, 2007, Advances in Neural Information Processing Systems 19, 1377
- schl�gl, 2007, Towards Brain-Computer Interfacing, 297
Cited by
Items 601–700 of 1,002
Sreeja, S. R., Himanshu, & Samanta, D. (2020) · Multimedia Tools and Applications · doi:10.1007/s11042-019-08602-0 · Cited by 31
Liu, T., Huang, G., Jiang, N., Yao, L., & Zhang, Z. (2020) · Journal of Neural Engineering · doi:10.1088/1741-2552/ab914d · Cited by 31
Mohdiwale, S., Sahu, M., Sinha, G. R., & Bhateja, V. (2020) · IEEE Sensors Journal · doi:10.1109/jsen.2020.3026172 · Cited by 36
Performance of EEG Motor-Imagery based spatial filtering methods: A BCI study on Stroke patients
Miladinović, A., Ajčević, M., Jarmolowska, J., Marušič, U., Silveri, G., Battaglini, P. P., & Accardo, A. (2020) · Procedia Computer Science · doi:10.1016/j.procs.2020.09.270 · Cited by 33
Hehenberger, L., Kobler, R. J., Lopes-Dias, C., Srisrisawang, N., Tumfart, P., Uroko, J. B., Torke, P. R., & Müller-Putz, G. (2021) · Frontiers in Human Neuroscience · doi:10.3389/fnhum.2021.635777 · Cited by 38
“The equalized epochs were used to fit the parameters of common spatial pattern (CSP) spatial filters (Ramoser et al., 2000; Blankertz et al., 2008) for each of the four frequency bands (Ang et al., 2008).”
MethodTowards Identifying Optimal Biased Feedback for Various User States and Traits in Motor Imagery BCI
Mladenović, J., Frey, J., Pramij, S., Mattout, J., & Lotte, F. (2021) · IEEE Transactions on Biomedical Engineering · doi:10.1109/tbme.2021.3113854 · Cited by 38
“filter, whose frequency band was optimized in a subject specific-way as in [36]; (2) Common Spatial Patterns (CSP) spatial filters to reduce the 32 original channels down to 6 “virtual” channels that maximize the differences between the two motor imagery classes [37]; (3) Features computed as the EEG band power after…”
MethodRelevance-based channel selection in motor imagery brain–computer interface
Nagarajan, A., Robinson, N., & Guan, C. (2023) · Journal of Neural Engineering · doi:10.1088/1741-2552/acae07 · Cited by 32
Pan, L., Wang, K., Xu, L., Sun, X., Yi, W., Xu, M., & Ming, D. (2023) · Journal of Neural Engineering · doi:10.1088/1741-2552/ad0a01 · Cited by 36
“A more pronounced ERD of alpha and beta band can be found in the sensorimotor area contralateral to the imagined limb compared to the ipsilateral side when people imagine the movements of their limb [12].”
Sensorimotor learning with stereo auditory feedback for a brain–computer interface
McCreadie, K. A., Coyle, D., & Prasad, G. (2012) · Medical & Biological Engineering & Computing · doi:10.1007/s11517-012-0992-7 · Cited by 34
Wang, H., & Xu, D. (2012) · IEEE Transactions on Biomedical Engineering · doi:10.1109/tbme.2012.2205383 · Cited by 33
The Berlin Brain-Computer Interface
Blankertz, B., Tangermann, M., Popescu, F., Krauledat, M., Fazli, S., Dónaczy, M., Curio, G., & Müller, K. (2008) · Lecture notes in computer science · doi:10.1007/978-3-540-68860-0_4 · Cited by 35
“For details on the technique of CSP analysis and its extensions we refer to ([28,36,37,38,39]).”
BackgroundMethodSelf-citationHossain, I., Khosravi, A., & Nahavandhi, S. (2016) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn.2016.7727726 · Cited by 35
“C1 and C2 can be obtained by C(c) = 1 nc ∑ i Ic XiX T i c = [1, 2] (2) Here, Ic denotes all indices of trials corresponding to class c and nc is the total number of trials from class c. CSP filter matrix W can be obtained by solving C1W = (C1 + C2)WD (3) Here, D denotes diagonal matrix which consists of eigenvalues fo…”
MethodInfluentialDecoding Subjective Emotional Arousal during a Naturalistic VR Experience from EEG Using LSTMs
Hofmann, S., Klotzsche, F., Mariola, A., Nikulin, V. V., Villringer, A., & Gaebler, M. (2018) · IEEE International Conference on Artificial Intelligence and Virtual Reality (AIVR) · doi:10.1109/aivr.2018.00026 · Cited by 33
“Since the LSTM was trained on neural features extracted with SSD and SPoC, the topological distribution of oscillations could be analyzed by reprojecting the corresponding filter matrices [23], as it was done with the CSP approach [21].”
ResultCompact convolutional transformer for subject-independent motor imagery EEG-based BCIs
Keutayeva, A., Fakhrutdinov, N., & Abibullaev, B. (2024) · Scientific Reports · doi:10.1038/s41598-024-73755-4 · Cited by 47
Learning optimal spatial filters by discriminant analysis for brain–computer-interface
Pang, Y., Yuan, Y., & Wang, K. (2011) · Neurocomputing · doi:10.1016/j.neucom.2011.07.016 · Cited by 31
Nguyen, C. H., Karavas, G. K., & Artemiadis, P. (2019) · PLoS ONE · doi:10.1371/journal.pone.0212620 · Cited by 31
Zhao, X., Zhao, J., Liu, C., & Cai, W. (2020) · BioMed Research International · doi:10.1155/2020/7285057 · Cited by 40
“Frequency bands should be carefully chosen, the reason is that the performance of the CSP algorithm depends much on the frequency bands [8, 18, 45].”
BackgroundResultInfluentialRiemannian Geometry-Based Decoding of the Directional Focus of Auditory Attention Using EEG
Geirnaert, S., Francart, T., & Bertrand, A. (2021) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp39728.2021.9413404 · Cited by 30
Neural decoding of semantic concepts: a systematic literature review
Rybář, M., & Daly, I. P. (2022) · Journal of Neural Engineering · doi:10.1088/1741-2552/ac619a · Cited by 42
“…the projected data, independent component analysis decomposes the data into statistically independent components, and common spatial patterns [144, 145] (used to aid binary classification) projects the data into a space that maximizes the signal variance for one class, while simultaneously…”
Zhong, Y., Yao, L., Wang, J., & Wang, Y. (2022) · IEEE Transactions on Biomedical Engineering · doi:10.1109/tbme.2022.3201241 · Cited by 31
“Traditional machine learning approaches, such as common spatial pattern (CSP) [9] and deep learning [10] have been shown to significantly improve MI decoding accuracy, however, the further improvement is limited [11].”
BackgroundModel based generalization analysis of common spatial pattern in brain computer interfaces
Huang, G., Liu, G., Meng, J., Zhang, D., & Zhu, X. (2010) · Cognitive Neurodynamics · doi:10.1007/s11571-010-9117-x · Cited by 32
“[5], a 2-D toy example was used to make a good understanding on CSP.”
MethodAssessment of new spectral features for eeg-based emotion recognition
Conneau, A.-C., & Essid, S. (2014) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp.2014.6854493 · Cited by 36
“We refer the interested reader to [20] for more details.”
BackgroundRaza, H., Cecotti, H., & Prasad, G. (2015) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn.2015.7280737 · Cited by 34
“To date, the low accuracy of classification has been one of the main concerns of the developed BCI systems, which directly affects the decision made by the BCI output [13].”
BackgroundMethodInfluentialWang, Z., Lu, Y., Zhou, Y., Chen, L., Gu, B., Liu, S., Xu, M., He, F., & Ming, D. (2023) · IEEE Transactions on Neural Systems and Rehabilitation Engineering · doi:10.1109/tnsre.2023.3281855 · Cited by 31
Lee, M., Park, H.-Y., Park, W., Kim, K.-T., Kim, Y., & Jeong, J.-H. (2024) · IEEE Transactions on Neural Systems and Rehabilitation Engineering · doi:10.1109/tnsre.2024.3395133 · Cited by 33
“We calculated a transformation matrix using spatial patterns consisting of the logarithmic variances of the first and last three columns of the inverse weight matrix [35], [36].”
MethodDelisle-Rodríguez, D., Villa-Parra, A. C., Bastos-Filho, T., López-Delis, A., Frizera, A., Krishnan, S., & Rocón, E. (2017) · Sensors · doi:10.3390/s17122725 · Cited by 31
“Notice also that RK, CSP and FBCSP have been successfully used in BCIs to recognize imagery motor and mental tasks [15,18], however, RK presented better performance than CSP and FBCSP, showing values of ACC = 62.76 ± 1.28%, TPR = 59.95 ± 1.63%, FPR = 36.0± 1.79%, F1 = 51.66± 1.47%.”
BackgroundMethodInfluentialRobotic orthosis compared to virtual hand for Brain–Computer Interface feedback
Cantillo-Negrete, J., Carino-Escobar, R. I., Carrillo‐Mora, P., Madrigal, J. A. B., & Arias-Carrión, Ó. (2018) · Journal of Applied Biomedicine · doi:10.1016/j.bbe.2018.12.002 · Cited by 29
Kirar, J. S., & Agrawal, R. K. (2019) · Applied Soft Computing · doi:10.1016/j.asoc.2019.105519 · Cited by 30
Zhang, S., Zhu, Z., Zhang, B., Feng, B., Yu, T., & Li, Z. (2020) · IEEE Sensors Journal · doi:10.1109/jsen.2020.3016402 · Cited by 30
“Therefore, the 8-30 Hz band-pass filter is usually used to filter the motor imagery signals, and then feature extraction is conducted [19].”
BackgroundMethodThe performance impact of data augmentation in CSP-based motor-imagery systems for BCI applications
Gubert, P. H., Costa, M. H., Silva, C. D., & Trofino, A. (2020) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2020.102152 · Cited by 30
Dehghani, M., Mobaien, A., & Boostani, R. (2021) · Brain-Computer Interfaces · doi:10.1080/2326263x.2021.1943955 · Cited by 33
Performance Variation of a Somatosensory BCI Based on Imagined Sensation: A Large Population Study
Yao, L., Jiang, N., Mrachacz‐Kersting, N., Zhu, X., Farina, D., & Wang, Y. (2022) · IEEE Transactions on Neural Systems and Rehabilitation Engineering · doi:10.1109/tnsre.2022.3198970 · Cited by 30
“We used Common Spatial Pattern (CSP) to enhance the feature discrimination between different tasks [56], [57], and the linear discriminative analysis (LDA) was utilized for classification.”
MethodRNeuMark: A Riemannian EEG Analysis Framework for Neuromarketing
Georgiadis, K., Kalaganis, F. P., Oikonomou, V. P., Nikolopoulos, S., Laskaris, N., & Kompatsiaris, I. Y. (2022) · Brain Informatics · doi:10.1186/s40708-022-00171-7 · Cited by 34
“One such approach could be the use of spatial filters [52], resulting in the selection of a predefined number of sensors.”
BackgroundDouble Stage Transfer Learning for Brain–Computer Interfaces
Gao, Y., Li, M., Peng, Y. F., Fang, F. Z., & Zhang, Y. (2023) · IEEE Transactions on Neural Systems and Rehabilitation Engineering · doi:10.1109/tnsre.2023.3241301 · Cited by 31
“Second, the aligned data in the source domain was re-weighted by the proposed distance-based reweighting method, and then after feature extracted by CSP [8], [25], [26], [27], TCA was used to reduce the inter-domain difference again.”
BackgroundMethodTowards a cure for BCI illiteracy: machine learning based co-adaptive learning
Blankertz, B., & Vidaurre, C. (2009) · BMC Neuroscience · doi:10.1186/1471-2202-10-s1-p85 · Cited by 29
“For the subsequent 3 runs, a classifier was trained on a more complex composed band-power feature in a subject-specific narrow band from optimized Common Spatial Pattern (CSP [2]) filters and in 6 Laplacian channels.”
MethodSelf-citationNeuromuscular electrical stimulation induced brain patterns to decode motor imagery
Vidaurre, C., Pascual, J., Ramos‐Murguialday, A., Lorenz, R., Blankertz, B., Birbaumer, N., & Müller, K. R. (2013) · Clinical Neurophysiology · doi:10.1016/j.clinph.2013.03.009 · Cited by 28
Self-citationClosing the sensorimotor loop: Haptic feedback facilitates decoding of arm movement imagery
Gomez-Rodriguez, M., Peters, J., Hill, J., Schölkopf, B., Gharabaghi, A., & Grosse‐Wentrup, M. (2010) · Conference proceedings/Conference proceedings - IEEE International Conference on Systems, Man, and Cybernetics · doi:10.1109/icsmc.2010.5642217 · Cited by 33
“In the past two decades, research on Brain-Computer Inter faces (BCls) has evolved from basic feasibility studies [1; 2; 3] to a state in which basic communication can be routinely performed after only brief calibration periods with healthy subjects [4; 5] as well as with subjects in early stages of amyotrophic later…”
BackgroundTime-shift denoising source separation
Alain De Cheveigné, A. (2010) · Journal of Neuroscience Methods · doi:10.1016/j.jneumeth.2010.03.002 · Cited by 29
Meng, J., Edelman, B. J., Olsoe, J., Jacobs, G., Zhang, S., Beyko, A., & He, B. (2018) · Frontiers in Neuroscience · doi:10.3389/fnins.2018.00227 · Cited by 31
“The selection of channel configuration, more often entangled with frequency/spectral optimization, and associated computer algorithms have been studied by various offline analyses (Lal et al., 2004; Blankertz et al., 2008b; Arvaneh et al., 2011; Meng et al., 2013).”
BackgroundMethodInfluentialMeisheri, H., Ramrao, N., & Mitra, S. K. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1802.09046 · Cited by 33
“Selection of m is done using following method as proposed by Blankertz et al. (2008) where, ι in a set of trials belonging to a particular class.”
BackgroundMethodRegularized Filters for L1-Norm-Based Common Spatial Patterns
Wang, H., & Li, X. (2015) · IEEE Transactions on Neural Systems and Rehabilitation Engineering · doi:10.1109/tnsre.2015.2474141 · Cited by 30
“It is critically beneficial to study the robust modelling for CSP, as represented by [5], [6], [13], [14], [31], [36], [39].”
BackgroundMethodInfluentialAlgorithm based on the short-term Rényi entropy and IF estimation for noisy EEG signals analysis
Lerga, J., Saulig, N., & Mozetič, V. (2016) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2016.11.002 · Cited by 32
Real-time detection and discrimination of visual perception using electrocorticographic signals
Kapeller, C., Ogawa, H., Schalk, G., Kunii, N., Coon, W. G., Scharinger, J., Guger, C., & Kamada, K. (2018) · Journal of Neural Engineering · doi:10.1088/1741-2552/aaa9f6 · Cited by 32
“For further processing only the four most relevant filters of each paired condition were used (i.e. the filters that corresponded to the two highest and the two lowest eigenvalues (Blankertz et al 2008)), resulting in twelve feature channels in total.”
MethodAuditory Attention Detection via Cross-Modal Attention
Su, E., Cai, S., Li, P., Xie, L., & Li, H. (2021) · Frontiers in Neuroscience · doi:10.3389/fnins.2021.652058 · Cited by 38
“Previous studies have demonstrated that classification performed on the CSP features generally yields better accuracy in motor imagery BCI systems (Blankertz et al., 2007a; Zhang et al., 2017).”
BackgroundResultTriana-Guzman, N., Orjuela-Cañón, Á. D., Jutinico, A. L., Mendoza-Montoya, O., & Antelis, J. M. (2022) · Frontiers in Neuroinformatics · doi:10.3389/fninf.2022.961089 · Cited by 35
Anders, C., & Arnrich, B. (2022) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2022.106088 · Cited by 40
Subject-Invariant Deep Neural Networks Based on Baseline Correction for EEG Motor Imagery BCI
Kwak, Y., Kong, K., Song, W., & Kim, S. (2023) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2023.3238421 · Cited by 34
“TheCSPmethodextractsmaximallydiscrimi-native band power for different classes, and classifies them with classical machine learning algorithms [8], [9].”
BackgroundA new feature and associated optimal spatial filter for EEG signal classification: Waveform Length
Lotte, F., & Bordeaux, I. (2012) · HAL · Cited by 38
Motion Estimation Based on EMG and EEG Signals to Control Wearable Robots
Kiguchi, K., & Hayashi, Y. (2013) · Conference proceedings/Conference proceedings - IEEE International Conference on Systems, Man, and Cybernetics · doi:10.1109/smc.2013.718 · Cited by 28
Novel Classification System for Classifying Cognitive Workload Levels Under Vague Visual Stimulation
Mahmoud, R., Shanableh, T., Bodala, I. P., Thakor, N. V., & Al-Nashash, H. A. (2017) · IEEE Sensors Journal · doi:10.1109/jsen.2017.2727539 · Cited by 32
“BCSP is a method based on conventional Common Spatial Patterns (CSP) method [16].”
MethodMotor imagery based brain–computer interfaces
Scherer, R., & Vidaurre, C. (2018) · Elsevier eBooks · doi:10.1016/b978-0-12-812892-3.00008-x · Cited by 46
An adaptive CSP filter to investigate user independence in a 3-class MI-BCI paradigm
Costa, A. P., Møller, J. S., Iversen, H. K., & Puthusserypady, S. K. (2018) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2018.09.021 · Cited by 29
Bai, D., Chen, S., & Yang, J. (2019) · Journal of Healthcare Engineering · doi:10.1155/2019/3958029 · Cited by 37
“One approach is to combine multiple two-class CSP methods based on one class versus all other class (OvR) or one class versus another class (OvO) scheme [28] and another method based on joint approximation diagonalization with diagonalizable simultaneously multiclass covariance matrices (JAD) [21].”
MethodEEG-Based Brain-Computer Interfacing (BCI)
Bansal, D., & Mahajan, R. (2019) · Elsevier eBooks · doi:10.1016/b978-0-12-814687-3.00002-8 · Cited by 37
Kumar, K. S., Mani, A. S. R., Sundaresan, S. S. K., & Kumar, T. A. (2021) · doi:10.1002/9781119761846.ch6 · Cited by 28
Huang, W., Wang, W., Li, Y., & Wu, W. (2024) · IEEE Transactions on Affective Computing · doi:10.1109/taffc.2024.3385651 · Cited by 30
“These filters are particularly useful in single-trial analysis for improving the signal-to-noise ratio [22].”
Yang, Y., Luo, S., Wang, W., Gao, X., Yao, X., & Wu, T. (2024) · NeuroImage Clinical · doi:10.1016/j.nicl.2024.103608 · Cited by 42
“Time-frequency methods mainly include methods based on common spatial patterns (Blankertz et al., 2007) and discriminative spatial patterns, and those used to analyze the components of correlation (Liao et al., 2007; Tanaka and Miyakoshi, 2019; Zhang et al., 2018).”
Channel reflection: Knowledge-driven data augmentation for EEG-based brain–computer interfaces
Wang, Z., Li, S., Luo, J., Liu, J., & Wu, D. (2024) · Neural Networks · doi:10.1016/j.neunet.2024.106351 · Cited by 38
Gruenwald, J., Znobishchev, A., Kapeller, C., Kamada, K., Scharinger, J., & Guger, C. (2019) · Frontiers in Neuroscience · doi:10.3389/fnins.2019.00901 · Cited by 32
“Common spatial patterns (CSPs) are the de-facto standard for dimension reduction in EEG signal processing (Blankertz et al., 2008; Lotte et al., 2015) and are also popular in ECoG signal processing (Onaran et al., 2011; Kapeller et al., 2014; Xie et al., 2015; Jiang et al., 2017).”
BackgroundMethodDecoding Imagined Speech and Computer Control using Brain Waves
Singh, A., & Gumaste, A. (2021) · Journal of Neuroscience Methods · doi:10.1016/j.jneumeth.2021.109196 · Cited by 29
Achanccaray, D., Izumi, S., & Hayashibe, M. (2021) · Computational Intelligence and Neuroscience · doi:10.1155/2021/8832686 · Cited by 29
Mobile Electroencephalography for Studying Neural Control of Human Locomotion
Song, S., & Nordin, A. D. (2021) · Frontiers in Human Neuroscience · doi:10.3389/fnhum.2021.749017 · Cited by 31
“Because BCI studies have fewer trials for real-time control, preprocessing is essential (Wolpaw and McFarland, 2004; Kübler et al., 2005; Blankertz et al., 2007), with data-driven supervised decomposition algorithms typically used as a spatial filter (Blankertz et al.”
BackgroundMethodInfluentialMidfrontal theta dynamics index the monitoring of postural stability
Stokkermans, M., Solis‐Escalante, T., Cohen, M. X., & Weerdesteyn, V. (2022) · Cerebral Cortex · doi:10.1093/cercor/bhac283 · Cited by 29
“This generalized eigenvalue equation solves the Rayleigh quotient and is often used in machine learning and brain–computer interface research (Blankertz et al. 2008; Haufe et al. 2014; Cohen 2021).”
MethodDelisle-Rodríguez, D., Silva, L., & Bastos-Filho, T. (2023) · Journal of Neural Engineering · doi:10.1088/1741-2552/acb73b · Cited by 27
Improving Generalization of CNN-Based Motor-Imagery EEG Decoders via Dynamic Convolutions
Barmpas, K., Panagakis, Y., Bakas, S., Adamos, D. A., Laskaris, N., & Zafeiriou, S. P. (2023) · IEEE Transactions on Neural Systems and Rehabilitation Engineering · doi:10.1109/tnsre.2023.3265304 · Cited by 33
“The technique of common spatial pattern (CSP) algorithm [19] and its various extensions, like the Filter-Bank CSP (FBCSP) [20], are among the most popular methods of this category due to their simplicity in design and computational efficiency in implementation.”
BackgroundMethodOptimizing spatial filter pairs for EEG classification based on phase-synchronization
Caramia, N., Lotte, F., & Ramat, S. (2014) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp.2014.6853959 · Cited by 30
“, the signal band power) recorded under one condition while minimizing the variance of the band-pass filtered signal during a second condition [4].”
BackgroundCommon spatial pattern patches - An optimized filter ensemble for adaptive brain-computer interfaces
Sannelli, C., Vidaurre, C., Müller, K., & Blankertz, B. (2010) · Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) · doi:10.1109/iembs.2010.5626227 · Cited by 32
“Spatial filters which are optimized specifically for each user by Common Spatial Pattern (CSP) analysis ([8]) are particularly successful for this application, see [ 1 ] for a review.”
BackgroundMethodSelf-citationRobust EEG channel selection across sessions in brain-computer interface involving stroke patients
Arvaneh, M., Guan, C., Ang, K. K., & Quek, C. H. (2012) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn.2012.6252687 · Cited by 28
Decoding the ERD/ERS: influence of afferent input induced by a leg assistive robot
Lisi, G., Noda, T., & Morimoto, J. (2014) · Frontiers in Systems Neuroscience · doi:10.3389/fnsys.2014.00085 · Cited by 32
“CSP is applied in a One-vs-Rest (OvR) fashion, in order to cope with the multi-class nature of the problem, separately (Dornhege et al., 2004) for three different frequency bands 8–13 Hz (μ), 15–25 Hz (β), 8–30 Hz (μ and β), in the time segment starting 1 s after the cue (Blankertz et al., 2008b).”
BackgroundMethodInfluentialA novel BCI-controlled pneumatic glove system for home-based neurorehabilitation
Coffey, A. L., Leamy, D. J., & Ward, T. E. (2014) · Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) · doi:10.1109/embc.2014.6944407 · Cited by 32
“tec software [14] was used to remove artifact-affected trials and noisy channels, carry out CSP analysis, produce CSP filters and train a Linear Discriminant”
MethodReddy, T. K., Arora, V., Behera, L., Wang, Y., & Lin, C. (2019) · IEEE Transactions on Fuzzy Systems · doi:10.1109/tfuzz.2019.2892921 · Cited by 28
“In [9], a supervised spatial filter called common spatial patterns (CSP) has been designed for oscillatory activity based”
BackgroundBrain–computer interfacing under distraction: an evaluation study
Brandl, S., Frølich, L., Höhne, J., Müller, K., & Samek, W. (2016) · Journal of Neural Engineering · doi:10.1088/1741-2560/13/5/056012 · Cited by 32
“All classi fi cation approaches throughout this paper are based on feature extraction using common spatial patterns ( CSP ) [ 43 ] and RLDA [ 9, 42 ] .”
MethodA CNN-based comparing network for the detection of steady-state visual evoked potential responses
Xing, J., Qiu, S., Ma, X., Wu, C., Li, J., Wang, S., & He, H. (2020) · Neurocomputing · doi:10.1016/j.neucom.2020.03.048 · Cited by 25
Increase performance of four-class classification for motor-imagery based brain-computer interface
Thang, L. Q., & Temiyasathit, C. (2014) · International Conference on Computer, Information and Telecommunication Systems · doi:10.1109/cits.2014.6878959 · Cited by 32
“their command if they receive the incorrect result from the feedback [8].”
BackgroundMethodInfluentialMotor imagery EEG signal classification on DWT and crosscorrelated signal features
Verma, N. K., Rao, L. S. V. S., & Sharma, S. K. (2014) · IEEE International Conference on Industrial and Information Systems (ICIIS) · doi:10.1109/iciinfs.2014.7036473 · Cited by 26
“Motor Imagery (MI) is a mental process wherein the subject imagines that he is performing a specific motor action such as a hand or foot movement without otherwise performing it in reality [2].”
BackgroundNo somatotopy of sensorimotor alpha-oscillation responses to differential finger stimulation
Nierula, B., Hohlefeld, F. U., Curio, G., & Nikulin, V. V. (2013) · NeuroImage · doi:10.1016/j.neuroimage.2013.03.025 · Cited by 28
“No ICA was applied as a pre-processing step in order to avoid a possible deterioration in the performance of CSP with prior application of another spatial filtering (Blankertz et al., 2008).”
Mixture of feature specified experts
Kheradpisheh, S. R., Sharifizadeh, F., Nowzari-Dalini, A., Ganjtabesh, M., & Ebrahimpour, R. (2014) · Information Fusion · doi:10.1016/j.inffus.2014.02.006 · Cited by 27
“Brain Computer Interface (BCI) research aims to create a direct pathway between the brain and an external device.”
MethodInfluentialKawanabe, M., & Vidaurre, C. (2009) · World Congress on Medical Physics and Biomedical Engineering, September 7 - 12, 2009, Munich, Germany · doi:10.1007/978-3-642-03889-1_75 · Cited by 26
“Data driven approaches that optimize spatial filters for each subject individually have been proven useful [1], in particular in Brain-Computer Interfaces (BCIs), which translate the users intent (coded by a small set of mental tasks) into control actions such as computer applications or…”
BackgroundMethodSelf-citationEvolutionary perspective for optimal selection of EEG electrodes and features
Lahiri, R., Rakshit, P., & Konar, A. (2017) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2017.03.022 · Cited by 28
Enhancing the Hybrid BCI Performance With the Common Frequency Pattern in Dual-Channel EEG
Ko, L.-W., Komarov, O., & Lin, S.-C. (2019) · IEEE Transactions on Neural Systems and Rehabilitation Engineering · doi:10.1109/tnsre.2019.2920748 · Cited by 27
“over the EEG channels [45], [46].”
Classification of left and right foot kinaesthetic motor imagery using common spatial pattern
Tariq, M., Trivailo, P. M., & Simić, M. (2019) · Biomedical Physics & Engineering Express · doi:10.1088/2057-1976/ab54ad · Cited by 31
“It is therefore common to retain e eigenvectors from both ends of the eigenvalue spectrum [16].”
BackgroundMethodInfluentialImproving BCI performance by task-related trial pruning
Sannelli, C., Braun, M. L., & Müller, K. (2009) · Neural Networks · doi:10.1016/j.neunet.2009.08.006 · Cited by 28
Dreyer, P., Roc, A., Pillette, L., Rimbert, S., & Lotte, F. (2023) · Scientific Data · doi:10.1038/s41597-023-02445-z · Cited by 56
“29 for Fig.”
InfluentialUsing NIRS as a predictor for EEG-based BCI performance
Fazli, S., Mehnert, J., Steinbrink, J. M., & Blankertz, B. (2012) · Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) · doi:10.1109/embc.2012.6347095 · Cited by 30
“Filter-banks have been proven useful for SMR-based BCIs [12], [13] in this sense, but also heuristics [11] are able to model this variability appropriately, among others.”
BackgroundSelf-citationA real-time BCI with a small number of channels based on CSP
Li, Y., & Koike, Y. (2010) · Neural Computing and Applications · doi:10.1007/s00521-010-0481-6 · Cited by 27
“Detailed CSP implementation can be referred to [3, 7].”
Stationary Common Spatial Patterns: Towards robust classification of non-stationary EEG signals
Wojcikiewicz, W., Vidaurre, C., & Kawanabe, M. (2011) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp.2011.5946469 · Cited by 30
“Brain-Computer Interface (BCI) systems aim to provide users control over a computer application by their brain activity.”
BackgroundSelf-citationSadatnejad, K., & Ghidary, S. S. (2015) · Neurocomputing · doi:10.1016/j.neucom.2015.11.065 · Cited by 27
“Several techniques are available for extracting features from EEG signals (Li et al., 2015; Brodu et al., 2012; Blankertz et al., 2008; Barachant et al., 2012).”
BackgroundMethodInfluentialXu, R., Jiang, N., Mrachacz‐Kersting, N., Dremstrup, K., & Farina, D. (2016) · Frontiers in Neuroscience · doi:10.3389/fnins.2015.00527 · Cited by 29
“The results presented here are of practical relevance for designing BCI systems for motor function rehabilitation.”
BackgroundKirar, J. S., & Agrawal, R. K. (2016) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2016.09.014 · Cited by 29
Kang, M.-K., Cho, H., Park, H.-M., Jun, S. C., & Yoon, K. (2017) · Applied Ergonomics · doi:10.1016/j.apergo.2017.02.022 · Cited by 27
Classification of Motor Imagery Based EEG Signals Using Sparsity Approach
Sreeja, S. R., Rabha, J., Samanta, D., Mitra, P., & Sarma, M. (2017) · Lecture notes in computer science · doi:10.1007/978-3-319-72038-8_5 · Cited by 25
“The available MI-based BCI systems makes use of spatial filtering and a powerful classification methods such as support vector machine (SVM) [17,18], logistic regression (LR) [13], linear discriminant analysis (LDA) [3] to attain good accuracy.”
MethodDecoding of Multi-directional Reaching Movements for EEG-Based Robot Arm Control
Jeong, J.-H., Kim, K.-T., Kim, D., & Lee, S. (2018) · Conference proceedings/Conference proceedings - IEEE International Conference on Systems, Man, and Cybernetics · doi:10.1109/smc.2018.00096 · Cited by 26
“The selected frequency band is [8-12] Hz (mu-band).”
BackgroundInfluentialLi, X., Fan, H., Wang, H., & Wang, L. (2019) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2019.04.034 · Cited by 28
Zuo, C., Jin, J., Xu, R., Wu, L., Liu, C., Miao, Y., & Wang, X. (2021) · Journal of Neural Engineering · doi:10.1088/1741-2552/abe20f · Cited by 26
A Bayesian optimization approach for rapidly mapping residual network function in stroke
Lorenz, R., Johal, M., Dick, F., Hampshire, A., Leech, R., & Geranmayeh, F. (2021) · Brain · doi:10.1093/brain/awab109 · Cited by 24
“Beyond optimizing for brain network activation/dissociation as we have done here, neuroadaptive Bayesian optimization is highly versatile, allowing us to target any clinically promising functional brain state that can be estimated in real-time, for example based on functional 42,54 or effective connectivity, 55 multiv…”
BackgroundGeorge, L., Lotte, F., Viciana-Abad, R., & Lécuyer, A. (2011) · Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) · doi:10.1109/iembs.2011.6091554 · Cited by 25
“A common spatial pattern (CSP) method [14] was then used to compute spatial filters for each of them.”
Miao, M., Zhang, W., Hu, W., & Wang, R. (2020) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2020.101994 · Cited by 29
Long, J., Li, Y., & Yu, Z. (2010) · Cognitive Neurodynamics · doi:10.1007/s11571-010-9114-0 · Cited by 27
“Note that U can be obtained through a standard procedure for joint diagonalization of two matrices (Blankertz et al. 2008).”
BackgroundMethodInfluentialTime–frequency transform based EEG data augmentation for brain–computer interfaces
Wang, Z., Li, S., Chen, X., & Wu, D. (2025) · Knowledge-Based Systems · doi:10.1016/j.knosys.2025.113074 · Cited by 36
EEG signal enhancement using multi-channel wiener filter with a spatial correlation prior
Maki, H., Toda, T., Sakti, S., Neubig, G., & Nakamura, S. (2015) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp.2015.7178449 · Cited by 27
“Thus, techniques of artifact removal or feature extraction from a single-trial EEG data have become a highly active research topic in neuroscience, engineering and signal processing [8,9].”
Background
Metadata
- Authors
- Blankertz, B., Tomioka, R., Lemm, S., Kawanabe, M., & Muller, K.-R.
- Affiliations
- University of Florida (US)
- Conexant (United States) (US)
- Fraunhofer Institute for Open Communication Systems (DE)
- Venue
- IEEE Signal Processing Magazine
- Published
- 2008
- Volume
- 25
- Issue
- 1
- Pages
- 41-56
- ISSN
- 1053-5888 (ISSN-L)
- OpenAlex
- W2099509424
- Cited by
- 2,080
- References
- 62
- Citations by year
- 2026: 96 · 2025: 116 · 2024: 109 · 2023: 126 · 2022: 121 · 2021: 154 · 2020: 159 · 2019: 140 · 2018: 130 · 2017: 124 · 2016: 117 · 2015: 120 · 2014: 121 · 2013: 135 · 2012: 89
- FWCI
- 37.89
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- IEEE Signal Processing Magazine (publishedVersion)
- Publikationsdatenbank der Fraunhofer-Gesellschaft (Fraunhofer-Gesellschaft) (submittedVersion)
- Fraunhofer-Publica (Fraunhofer-Gesellschaft) (submittedVersion)
- Funders
- Bundesministerium für Bildung und Forschung
- Topics
- Life Sciences › Neuroscience › Cognitive Neuroscience › EEG and Brain-Computer Interfaces
- Physical Sciences › Computer Science › Artificial Intelligence › Emotion and Mood Recognition
- Physical Sciences › Computer Science › Signal Processing › Blind Source Separation Techniques
- Keywords
- common spatial pattern, spatial filtering, brain-computer interface, signal-to-noise ratio, volume conduction, EEG preprocessing
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.