| |
| |
| 4.
BCI - Brain–computer interfaces |
| |
4.1.
Definition and features of a BCI
| |
4.1.3. The parts of a BCI
| |
4.1.3.3. Signal processing: the translation
algorithm |
|
|
The
first part of signal processing simply extracts specific signal
features. The next stage, the translation algorithm, translates
these signal features into device commandsorders that carry
out the user’s intent.
This algorithm might use:
1) linear methods (e.g. classical statistical
analyses)
or
2) nonlinear methods (e.g. neural networks).
Whatever its nature, each algorithm changes independent variables
(i.e. signal features) into dependent variables (i.e. device
control commands).
Effective algorithms adapt to each user on 3 levels:
1) First level of adaptation: when a new user
first accesses the BCI the algorithm adapts to that user’s signal
features.
2) Second level of adaptation: periodic online
adjustments to reduce the impact of spontaneous variations (time
of day, hormonal levels, immediate environment, recent events,
fatigue, illness, and other factors).
3) The third level of adaptation accommodates
and engages the adaptive capacities of the brain.
Like activity in the brain’s conventional neuromuscular communication
and control channels, BCI signal features will be affected by
the device commands they are translated into: the results of
BCI operation will affect future BCI input. In the most desirable
(and hopefully typical) case, the brain will modify signal features
so as to improve BCI operation.
Source:
Brain–computer
interfaces for communication and control, Clinical Neurophysiology
113 (2002) 767–791, Jonathan R. Wolpaw, Niels Birbaumer, Dennis
J. McFarland, Gert Pfurtscheller, Theresa M. Vaughan |
|
|
