Real-time brain-computer interfacing: A preliminary study using Bayesian learning

Article

Abstract

Preliminary results from real-time ‘brain-computer interface’ experiments are presented. The analysis is based on autoregressive modelling of a single EEG channel coupled with classification and temporal smoothing under a Bayesian paradigm. It is shown that uncertainty in decisions is taken into account under such a formalism and that this may be used to reject uncertain samples, thus dramatically improving system performance. Using the strictest rejection method, a classification performance of 86.5±6.9% is achieved over a set of seven subjects in two-way cursor movement experiments.

Keywords

Brain-computer interfacing Real-time EEG analysis Biosignal analysis Bayesian learning 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Bishop, C. (1995): “Neural networks for pattern recognition’, (Oxford University Press, Oxford)Google Scholar
  2. Fernandez, T. (1995): ‘EEG activation patterns during the performance of tasks involving different components of mental calculation’,Electroenceph. Clin. Neuro-physiol.,94, pp. 175–182Google Scholar
  3. Lee, P. (1994): ‘Bayesian Statistics: An Introduction’, (Edward Arnold)Google Scholar
  4. MacKay, D. (1992): ‘The evidence framework applied to classification networks’,Neural Computat.,4, pp. 720–736Google Scholar
  5. McFarland, D., McCane, L., Miner, L., Vaughan, T., andWolpaw, J. (1997). ‘EEG Mu and beta rhythm topographies with movement imagery and actual movement’,Soc. Neuroscience Abstr., p. 1277Google Scholar
  6. McFarland, D., Neat, G., Read, R., andWolpaw, J. (1993): An EEG-based method for graded cursor control’,Psychobiol.,21, pp. 77–81Google Scholar
  7. O'Ruanaidth, J., andFitzgerald, W. (1996): ‘Numerical Bayesian methods applied to signal processing’ (Springer)Google Scholar
  8. Pardey, J., Roberts, S., andTarassenko, L. (1996): ‘A review of parametric modelling techniques for EEG analysis’,Med. Eng. Phys.,18, pp. 2–11CrossRefGoogle Scholar
  9. Peltoranta, M., andPfurtscheller, G. (1994): ‘Neural network based classification of non-averaged event-related EEG responses’,Med. Biol. Eng. Comput.,32, pp. 189–196Google Scholar
  10. Penny, W., andRoberts, S. (1998): ‘Imagined hand movements identified from the EEG Mu-Rhythm’.Technical report, Imperial College, University of London. Available via http://www.ee.ic.ac.uk.Google Scholar
  11. Penny, W., andRoberts, S. (1999a): ‘Dynamic models for nonstationary signal segmentation’, to be published inComput. Biomed. Research Google Scholar
  12. Penny, W., andRoberts, S. (1999b): ‘Experiments with an EEG-based computer interface’.Technical report, Imperial College, University of London. Available via http://www.ee.ic.ac.uk.Google Scholar
  13. Penny, W., Roberts, S., andStokes, M. (1999): ‘EEG-based communication: a pattern recognition approach’, to be published inIEEE Trans. Rehabil. Eng. Google Scholar
  14. Pfurtscheller, G., Flotzinger, D., andKalcher, J. (1993): ‘Brain-computer interface-a new communication device for handicapped people’,J. Microcomput. Appl.,16, pp. 293–299Google Scholar
  15. Pfurtscheller, G., Flotzinger, D., andNeuper, C. (1994): ‘Differentiation between finger, toe and tongue movement in man based on 40 Hz EEG’,Electroenceph. Clin. Neur.,90, pp. 456–460Google Scholar
  16. Pfurtscheller, G., Neuper, C., Schloegl, A., andLugger, K. (1998): ‘Separability of EEG signals recorded during right and left motor imagery using adaptive autoregressive parameters’,IEEE Trans. Rehabil. Eng.,6, pp. 316–325CrossRefGoogle Scholar
  17. Press, W., Flannery, B., Teukolsky, S., andVetterling, W. (1991): ‘Numerical recipes in C’ (Cambridge University Press)Google Scholar
  18. Puskorius, G. V., andFeldkamp, L. A. (1994). ‘Neurocontrol of nonlinear dynamical systems with Kalman filter-trained recurrent networks’,IEEE Trans. Neural Netw.,5, pp. 279–297CrossRefGoogle Scholar
  19. Roberts, S., andPenny, W. (1997): ‘Maximum certainty approach to feedforward neural networks’,Electron. Lett.,33, pp. 306–307Google Scholar
  20. Roberts, S., Penny, W., andRezek, I. (1998): ‘Temporal and spatial complexity measures for EEG-based brain-computer interfacing’,Med. Biol. Eng. Comput.,37, pp. 93–99Google Scholar
  21. Spiegelhalter, D., andLauritzen, S. (1990): ‘Sequential updating of conditional probabilities on directed graphical structures’,Networks,20, pp. 579–605MathSciNetGoogle Scholar
  22. Stam, C., vanWoerkom, T., andPritchard, W. (1996): ‘Use of nonlinear EEG measures to characterize EEG changes during mental activity’.Elecdtroenceph. Clin. Neurophysiol.,99, pp. 214–224Google Scholar
  23. Wolpaw, J., andMcFarland, D. (1994). ‘Multichannel EEG-based brain-computer communication’,Electroenceph. Clin. Neurophysiol.,90, pp. 444–449CrossRefGoogle Scholar
  24. World Precision Instruments Inc. (n.d.).ISO-DAM: Isolated low-noise pre-amplifier: Instruction manual, Sarasota, FL, USAGoogle Scholar

Copyright information

© IFMBE 2000

Authors and Affiliations

  1. 1.Robotics Research Group, Department of Engineering ScienceUniversity of OxfordUK

Personalised recommendations