BCI competition 2003-data set IIb: support vector machines for the P300 speller paradigm

By Matthias Kaper, Peter Meinicke, Ulf Großekathöfer, Thomas Lingner, Helge J. Ritter. IEEE Transactions on Biomedical Engineering, 2004

538 citations recorded by the source corpus.

We propose an approach to analyze data from the P300 speller paradigm using the machine-learning technique support vector machines. In a conservative classification scheme, we found the correct solution after five repetitions. While the classification within the competition is designed for offline analysis, our approach is also well-suited for a real-world online solution: It is fast, requires only 10 electrode positions and demands only a small amount of preprocessing.

Subfields: Cognitive Neuroscience, Electrical and Electronic Engineering, Electrochemistry

View Paper: https://www.semanticscholar.org/paper/3e961a86ed2b7422d5dc7ba32c795f2e2530691d

How researchers see this paper

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

In Cognitive Neuroscience

In Electrical and Electronic Engineering

In Electrochemistry

Citations in the corpus

A comparison of classification techniques for the P300 Speller (2006, methodology): “Several classification techniques have demonstrated notable performance for the P300 Speller, including stepwise linear discriminant analysis [2, 5], support vector machines [8, 10, 11], wavelets [1] and matched filtering [14].”
Brain Computer Interfaces, a Review (2012, methodology): “The Gaussian SVM has been applied in BCIs to classify P300 evoked potentials [241–243].”
A novel P300-based brain-computer interface stimulus presentation paradigm: moving beyond rows and columns (2010, methodology): “Our analyses focused on four electrodes, (Cz, Pz, Po7, and Po8), since these reliably capture most of the P300 energy in BCI applications (Kaper et al., 2004; Krusienski et al., 2008).”
An improved P300-based brain-computer interface (2005, background): “[23], [24], [26] that presents very promising results and an online BCI by Gao et al.”
A survey of signal processing algorithms in brain–computer interfaces based on electrical brain signals (2007, background): “…and Pineda 2003, 2005, Bayliss et al 2004, Farwell and Donchin 1988) None (Guan et al 2004, Jansen et al 2004, Kaper and Ritter 2004a, 2004b, Kaper et al 2004, Thulasidas et al 2004) VEP Spectral parameters (Cheng and Gao 1999, Cheng et al 2001, 2002, 2005, Gao et al 2003b, Kelly et al…”
Review of the BCI Competition IV (2012, methodology): “This worked very well in the BCI competition II where all winner articles have been published in volume 51 of IEEE Trans Biomed Eng (Blanchard and Blankertz, 2004; Bostanov, 2004; Kaper et al., 2004; Lemm et al., 2004; Mensh et al., 2004; Wang et al., 2004; Xu et al., 2004).”
Brain–computer interface systems: progress and prospects (2007, methodology): “For example, spectral activity might change reliably with selective attention and, thus, might be used to improve classification in a P300 or SSVEP BCI [17,69,89,157,158] .”
A P300-based brain-computer interface for people with amyotrophic lateral sclerosis. (2008, background): “…adults (5.5 bits/min; Donchin et al., 2000).1 Several previous P300-based BCI studies have presented bit rate with the time between characters selections removed (e.g., Donchin et al., 2000; Kaper et al., 2004; Meinicke et al., 2002; Serby et al., 2005); therefore, we have presented it both ways.”
Current Status, Challenges, and Possible Solutions of EEG-Based Brain-Computer Interface: A Comprehensive Review (2020, methodology): “In addition, the algorithm is shown to be robust with a high-dimensional dataset, which implies that a large training set is not required for good outcomes, even with high-dimensional feature vectors (Kaper et al., 2004).”
A review of classification algorithms for EEG-based brain–computer interfaces (2007, methodology): “A great variety of features have been attempted to design BCI such as amplitude values of EEG signals [10], band powers (BP) [11], power spectral density (PSD) values [12, 13], autoregressive (AR) and adaptive autoregressive (AAR) parameters [8, 14], timefrequency features [15] and inverse model-bas”
A P300-based brain-computer interface: initial tests by ALS patients. (2006, background): “Ritter and colleagues have demonstrated that support vector machines produce very robust classification rates as well, approaching 100% (Kaper et al., 2004; Meinicke et al., 2002); however, the method has not been implemented online, and it is not clear how this would be achieved.”
Toward Brain-Computer Interfacing (2009, background): “, Oz, PO7, and PO8) have previously been included in matrix speller data classification (Kaper et al. (2004); Meinicke et al.”
Toward enhanced P300 speller performance. (2008, unknown): “While recent offline evaluations suggest that the use of additional locations, particularly posterior sites, may improve classification accuracy (BCI, 2003; BCI, 2005; Blankertz et al., 2004, 2006; Kaper et al., 2004; Spencer et al., 2001; Vaughan et al., 2003), this possibility has not been formall”
A P300 event-related potential brain-computer interface (BCI): the effects of matrix size and inter stimulus interval on performance. (2006, methodology): “and independent components analysis (Serby et al., 2005), and in offline analyses using support vector machines ( Kaper et al., 2004; Meinicke et al., 2002).”
(C)overt attention and visual speller design in an ERP-based brain-computer interface (2010, result): “This confirms earlier conjectures [38,49] and it is also in accordance with [50] who showed that classifying on posterior electrodes in addition to the classical P3 sites improves classification performance.”