Sequential Selection of Window Length for Improved SSVEP-Based BCI Classification
Erik C. Johnson, James J. S. Norton, David Jun, Timothy Bretl, Douglas L. Jones

Citation
Erik C. Johnson, James J. S. Norton, David Jun, Timothy Bretl, Douglas L. Jones. "Sequential Selection of Window Length for Improved SSVEP-Based BCI Classification". Proceedings of International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) 2013, July, 2013.

Abstract
Brain-computer interfaces (BCI) utilizing steadystate visually evoked potentials (SSVEP) recorded by electroencephalography(EEG) have exciting potential to enable new systems for disabled individuals and novel controls for robotic and computer systems. To interact with SSVEP-based BCIs, users attend to visual stimuli modulated at predetermined frequencies. A key problem for SSVEP-based BCIs is to classify which modulation frequency the user is attending, for which there is an inherent trade-off between speed and accuracy. As SSVEP signals vary with time and stimulation frequency, a fixed-length data window does not necessarily optimize this trade-off. The authors propose a strategy, developed from sequential analysis, to vary the window-length used for classification. The proposed technique adapts to the data, continuing to collect data until it is confident enough to make a classification decision. The strategy was compared to a fixed window-length method using a simple experiment involving five frequencies presented individually to three participants. Using a canonical correlation analysis classifier to compare the proposed variablelength scheme to a standard fixed-length scheme, the variablelength approach improved the classifier information transfer rate by an average of 43%.

Electronic downloads

Internal. This publication has been marked by the author for TerraSwarm-only distribution, so electronic downloads are not available without logging in.

Citation formats

• HTML

  Erik C. Johnson, James J. S. Norton, David Jun, Timothy
  Bretl, Douglas L. Jones. <a
  href="http://www.terraswarm.org/pubs/69.html"
  >Sequential Selection of Window Length for Improved
  SSVEP-Based BCI Classification</a>, Proceedings of
  International Conference of the IEEE Engineering in Medicine
  and Biology Society (EMBC) 2013, July, 2013.

• Plain text

  Erik C. Johnson, James J. S. Norton, David Jun, Timothy
  Bretl, Douglas L. Jones. "Sequential Selection of
  Window Length for Improved SSVEP-Based BCI
  Classification". Proceedings of International
  Conference of the IEEE Engineering in Medicine and Biology
  Society (EMBC) 2013, July, 2013.

• BibTeX

  @inproceedings{JohnsonNortonJunBretlJones13_SequentialSelectionOfWindowLengthForImprovedSSVEPBased,
      author = {Erik C. Johnson and James J. S. Norton and David
                Jun and Timothy Bretl and Douglas L. Jones},
      title = {Sequential Selection of Window Length for Improved
                SSVEP-Based BCI Classification},
      booktitle = {Proceedings of International Conference of the
                IEEE Engineering in Medicine and Biology Society
                (EMBC) 2013},
      month = {July},
      year = {2013},
      abstract = {Brain-computer interfaces (BCI) utilizing
                steadystate visually evoked potentials (SSVEP)
                recorded by electroencephalography(EEG) have
                exciting potential to enable new systems for
                disabled individuals and novel controls for
                robotic and computer systems. To interact with
                SSVEP-based BCIs, users attend to visual stimuli
                modulated at predetermined frequencies. A key
                problem for SSVEP-based BCIs is to classify which
                modulation frequency the user is attending, for
                which there is an inherent trade-off between speed
                and accuracy. As SSVEP signals vary with time and
                stimulation frequency, a fixed-length data window
                does not necessarily optimize this trade-off. The
                authors propose a strategy, developed from
                sequential analysis, to vary the window-length
                used for classification. The proposed technique
                adapts to the data, continuing to collect data
                until it is confident enough to make a
                classification decision. The strategy was compared
                to a fixed window-length method using a simple
                experiment involving five frequencies presented
                individually to three participants. Using a
                canonical correlation analysis classifier to
                compare the proposed variablelength scheme to a
                standard fixed-length scheme, the variablelength
                approach improved the classifier information
                transfer rate by an average of 43%.},
      URL = {http://terraswarm.org/pubs/69.html}
  }
  

Posted by Mila MacBain on 14 May 2013.