Biomedical Engineering Reference
In-Depth Information
8.2.3.4 Stimulator: CRT Versus LED
In an SSVEP-based BCI, the visual stimulator serves as a visual response modulator
and a virtual control panel, thus it is a crucial aspect of system design. The visual
stimulator commonly consists of flickering targets in the form of color alternating
or checkerboard reversing. Usually, the CRT/LCD monitor or flashtube/LED is
used for stimulus display. A computer monitor is convenient for target alignment
and feedback presentation by programming. But for a frequency-coded system, the
number of targets is limited due to the refresh rate of the monitor and poor timing
accuracy of the computer operating system. Therefore, an LED stimulator is prefer-
able for a multiple-target system. The flickering frequency of each LED can be con-
trolled independently by a programmable logic device. Using such a stimulator, a
48-target BCI was reported in [30].
The number of stimulation targets can be up to 64, leading to various system
performances. Generally, the system with more targets can achieve a higher infor-
mation transfer rate. For example, in tests of a 13-target system, the subjects had an
average information transfer rate of 43 bits/min [31]. However, due to the fact that
a stimulator with more targets is also more exhausting for users, the number of tar-
gets should be considered by evaluating the trade-off between system performance
and user comfort.
8.2.3.5 Optimization of Electrode Layout: Bipolar Versus Multielectrode
As we know, using a small number of electrodes can reduce the cost of hardware
while improving the convenience of system operation. The Oz, O1, and O2 elec-
trode positions of the international 10-20 system are widely used in SSVEP-based
BCI. As shown in Section 8.2.2.3, in our system, we use a subject-specific electrode
placement method to achieve a high SNR for the SSVEPs, especially for the subjects
with strong background brain activities over the area of the visual cortex [31, 36].
In the near future, more convenient electrode designs, for example, the dry elec-
trode [44], will be highly desirable as replacements for the currently used wet elec-
trode. Under this circumstance, it is acceptable to use more electrodes to acquire
more sufficient data to fulfill detection of SSVEP signals with multichannel data
analysis approaches, for example, spatial filtering techniques described in [45] and
the canonical correlation analysis method presented in [46]. An additional advan-
tage of multiple-channel recording is that no calibration for electrode selection is
needed.
8.3
Sensorimotor Rhythm-Based BCI
8.3.1 Physiological Background and BCI Paradigm
In scalp EEGs, the occipital alpha rhythm (8 to 13 Hz) is a prominent feature espe-
cially when the subject is in the resting wakeful state. This kind of spontaneous
alpha rhythm is usually called “idling” activity. Besides visual alpha rhythm, a dis-
tinct alpha-band rhythm, in some circumstance with a beta-band accompaniment
(around 20 Hz), can be measured over the sensorimotor cortex, which is called
sensorimotor rhythm (SMR) [47, 48]. The mu and beta rhythms are commonly con-
Search WWH ::
Custom Search