EEG based Brain-Computer Interface (BCI) measures and analyzes the
electrical brain activity (electroencephalogram, EEG) in order to
convert the EEG into control commands. These commands are used to
control external devices like wheelchairs or robots, spelling
applications or smart environment like smart homes. BCIs are based -
depending on the type of application - on slow cortical potentials, EEG
oscillations in the alpha
beta band, the P300 response or steady-state visual evoked potentials
(SSVEP). For example, BCI systems based on slow cortical potentials or
oscillatory EEG components with 1-5 degrees of freedom were realized up
to now. However, high information transfer rates were reached based on
2 degrees of freedom as otherwise the accuracy of the BCI systems
dropped down. SSVEP based systems allow selecting up to 48 different
targets and are limited by the number of distinct frequency responses
that can be analyzed in the EEG. With P300 response based BCIs users
can select commands from a rather large command set reliably. Recent
advances in usability and reliability of BCI systems made it possible
to demonstrate its usefulness for persons with disabilities without
significant training effort. In this session different approaches based
on demonstrators shall be
introduced and vividly discussed.
(presentation order under discussion).
1) Tomohiko Igasaki,
Nobuki Murayama [Kumamoto University, Japan]
examination of a synchronous-type BCI using the fast Fourier transform for
Kanoh [Tohoku University,
"A brain-computer interface based on Auditory Stream
Kansaku [National Rehab Center, Japan]
interfaces for environmental control."
Edlinger [g.tec medical
engineering GmbH, Austria]
"Brain-Computer Interfaces for control of
virtual environments and smart homes."
5) Hideaki Touyama
University, Japan] (presented by G. Edlinger)
"Brain-Computer Interfaces with Virtual Reality
6) Sercan Taha Ahi,
Yasuharu Koike [Tokyo Institute of Technology, Japan]
"Dimension reduction techniques for the P300 response"
7) Arne Ewald, Guido
Nolte [Fraunhofer Institute