Feature enhancement of P300 based brain computer interface through spatially-constrained ICA

The P300 word speller is one of the important BC' applications which detects real-time P300 waveforms and translates them into letters (and then words) within a particular BC' paradigm. However due to the poor SNR of EEG, as well as the presence of other artifacts, the identification accuracy is still not high enough for real-world application. This paper presents two slightly different 'CA approaches to improve character classification performance based on 'CA. When compared with the classification results obtained from the data, the results using these approaches show distinct improvement. Furthermore, the results indicate that it is possible to reduce the number of epochs required to perform stimulus locked averages, whilst still maintaining good performance measures. This has the potential of speeding up the word speller and has further implications for the use on similar ERP based systems.

[1]  Christopher J James,et al.  Independent component analysis for biomedical signals , 2005, Physiological measurement.

[2]  Erkki Oja,et al.  Independent Component Analysis , 2001 .

[3]  E. Donchin,et al.  Talking off the top of your head: toward a mental prosthesis utilizing event-related brain potentials. , 1988, Electroencephalography and clinical neurophysiology.

[4]  Vibha Sazawal,et al.  The Unigesture Approach , 2002, Mobile HCI.

[5]  J. Wolpaw,et al.  EEG correlates of P300-based brain–computer interface (BCI) performance in people with amyotrophic lateral sclerosis , 2012, Journal of neural engineering.

[6]  Suogang Wang,et al.  Enhancing Evoked Responses for BCI Through Advanced ICA Techniques , 2006 .

[7]  Erkki Oja,et al.  Independent component approach to the analysis of EEG and MEG recordings , 2000, IEEE Transactions on Biomedical Engineering.

[8]  Helge J. Ritter,et al.  BCI competition 2003-data set IIb: support vector machines for the P300 speller paradigm , 2004, IEEE Transactions on Biomedical Engineering.

[9]  Erkki Oja,et al.  Independent component analysis: algorithms and applications , 2000, Neural Networks.

[10]  Sriram Subramanian,et al.  GesText: accelerometer-based gestural text-entry systems , 2010, CHI.

[11]  Jonghun Baek,et al.  A sequence-action recognition applying state machine for user interface , 2008, IEEE Transactions on Consumer Electronics.

[12]  Vibha Sazawal,et al.  TiltType: accelerometer-supported text entry for very small devices , 2002, UIST '02.

[13]  Jing Yang,et al.  Beatbox music phone: gesture-based interactive mobile phone using a tri-axis accelerometer , 2005, 2005 IEEE International Conference on Industrial Technology.

[14]  Frances M. Dyro,et al.  The EEG handbook , 1989 .

[15]  Zhen Wang,et al.  uWave: Accelerometer-based Personalized Gesture Recognition and Its Applications , 2009, PerCom.

[16]  Gregory D. Abowd,et al.  No-Look Notes: Accessible Eyes-Free Multi-touch Text Entry , 2010, Pervasive.

[17]  Yasuhiro Nageishi,et al.  Endogenous Event‐Related Potentials in Obsessive Character , 1993, The Japanese journal of psychiatry and neurology.

[18]  Daniel J. Wigdor,et al.  TiltText: using tilt for text input to mobile phones , 2003, UIST '03.

[19]  Masahiko Tsukamoto,et al.  Pointing with Accelerometers for Wearable Computing , 2007, 2007 11th IEEE International Symposium on Wearable Computers.

[20]  J J Vidal,et al.  Toward direct brain-computer communication. , 1973, Annual review of biophysics and bioengineering.

[21]  Christopher J. James,et al.  On Semi-Blind Source Separation Using Spatial Constraints With Applications in EEG Analysis , 2006, IEEE Transactions on Biomedical Engineering.

[22]  S Makeig,et al.  Blind separation of auditory event-related brain responses into independent components. , 1997, Proceedings of the National Academy of Sciences of the United States of America.

[23]  Sriram Subramanian,et al.  Tilt techniques: investigating the dexterity of wrist-based input , 2009, CHI.