Boosting bit rates in noninvasive EEG single-trial classifications by feature combination and multiclass paradigms

By Guido Dornhege, Benjamin Blankertz, Gabriel Curio, K. Müller. IEEE Transactions on Biomedical Engineering, 2004

662 citations recorded by the source corpus.

Subfields: Cellular and Molecular Neuroscience, Cognitive Neuroscience

View Paper: https://www.semanticscholar.org/paper/0360067259a8755036949324c0a2d3ff76d1c374

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Cellular and Molecular Neuroscience

In Cognitive Neuroscience

Citations in the corpus

The Berlin brain-computer interface: EEG-based communication without subject training (2006, methodology): “Further investigation with more subjects will include ERD/ERS effects in the frequency range, systematic comparison of the discriminability of different features, and classification analysis using combined RP+ERD features, see [15].”
Optimizing the Channel Selection and Classification Accuracy in EEG-Based BCI (2011, methodology): “2) Dataset IVa [24] from BCI competition III [28]: This dataset contains EEG data of ﬁve subjects recorded using 118 channels.”
Filter Bank Common Spatial Pattern Algorithm on BCI Competition IV Datasets 2a and 2b (2012, methodology): “…p ( x j | ) in 1997) given by (cid:19) (cid:20) The classiﬁcation rule of the NBPW classiﬁer is given by single-trial EEG (Ramoser et al., 2000), and several multi-class extensions of the CSP algorithm have been proposed (Dornhege al., 2004a; Dornhege et al., 2004b; Grosse-Wentrup and Buss, 2008).”
The non-invasive Berlin Brain-Computer Interface: Fast acquisition of effective performance in untrained subjects (2007, methodology): “The RP feature was used in combination with co-localized ERD features and showed encouraging results in offline BCI classification studies (Dornhege et al., 2004a,b, 2007c).”
A review of channel selection algorithms for EEG signal processing (2015, methodology): “…was evaluated using two datasets, Dataset IIa [49] from BCI competition IV with 22 channels (four motor imagery tasks; left hand, right hand, feet, or tongue) and Dataset IVa [50] from BCI competition III with 118 channels (two motor imagery tasks; right hand and foot) recorded from 14 subjects.”
The BCI competition III: validating alternative approaches to actual BCI problems (2006, background): “(3) Several of the winning algorithms incorporated the concept of combining oscillatory (ERD) and non-oscillatory (ERP) features (data sets I, IIIb, IVa), first proposed in [34], [35].”
Towards a Cure for BCI Illiteracy (2009, neutral): “Machine Learning based BCIs use EEG features of larger complexity that can be fitted better to the individual characteristics of brain patterns of each user (see Blankertz et al. 2007, 2008b; Dornhege et al. 2004, 2007; Müller et al. 2003, 2008).”
Feature Selection via Dependence Maximization (2012, methodology): “The result is then passed to a normal CSP method (Dornhege et al., 2004) for feature extraction and then classified using a linear SVM.”
A survey of signal processing algorithms in brain–computer interfaces based on electrical brain signals (2007, background): “( Cheng et al 2004, Dornhege et al 2003, 2004, Krauledat et al 2004, Mahmoudi and Erfanian 2002, Muller et al 2003b, Yom-Tov and Inbar 2001, 2002) LRP features (Blankertz et al 2002a, 2003, Krauledat et al 2004) Other (Coyle et al 2004, Huggins et al 2003, Hung et al 2005, LaCourse and Wilson 2003,…”
Multisubject Learning for Common Spatial Patterns in Motor-Imagery BCI (2011, methodology): “…with the basic CSP algorithm in Section 5 on one simulated data set and two experimental data sets, one of which is publicly available on the website of the third BCI competition [13] and one which includes data of 14 subjects recorded at the Max Planck Institute for Biological Cybernetics.”
Seperability of four-class motor imagery data using independent components analysis (2006, methodology): “For details about the extension of the CSP algorithm to multiclass problems see, e.g., Dornhege et al (2004).”
Towards adaptive classification for BCI (2006, methodology): “Other physiological paradigms implemented in the BBCI focus on the use of the lateralized readiness potential [4, 15, 16] and the combination of this feature with ERD/ERS [7, 8].”
Machine learning for real-time single-trial EEG-analysis: from brain-computer interfacing to mental state monitoring. (2008, neutral): “First of all, the use of state-of-the-art learning machines enables us to achieve high decision accuracies for BCI (e.g. (Blankertz et al., 2002, 2003; Dornhege et al., 2004)).”
BCILAB: a platform for brain–computer interface development (2013, methodology): “1) is available as dataset IVa(3) of the BCI competition III, recorded at the Berlin Institute of Technology [46].”