Document Type : Original Article
Authors
Department of Industrial Electronics and Control Engineering , Faculty of Electronic Engineering, Menoufia University, Egypt.
Abstract
Highlights
Results of this paper prove that imagination of right and
left hand movement may alter activity of neural in primary
sensorimotor regions, resulting in the changes of the μ
and b rhythms. Moreover, the results show that the
distinction between electrode c4and c3 is more clear in
case of right hand imagery. The results of this work can be
used as a base in the classification stage for BCI and
device control. Work in the future will use the results of
this study to directly extract quantitative features to
characterize between right and left -hand imagery and
monitor embedded rehabilitation robot to assist patients
with extreme paralysis in controlling their environmental
and communicating with the outside world.
Keywords
Main Subjects
EEG signals are measure for brain neural activity,
which change according to task performed by a person [1].
The EEG has several medical applications. For example,
EEG has been used as a brain diseases diagnostic method
[2] Also, neuroscience advances, brain imaging
technologies and computing have provided us with the
opportunity to directly interface the brain with a computer,
thereby creating a control and communication system that
bypasses peripheral muscles and nerves to enable
interaction through brain activity alone , this is known as
brain-computer interface[3][4][5] . BCI depends on
voluntarily modifying user mental state to control a device
whereas a pattern analysis method concurrently tries to
determine the parallel change of EEG signals[6]. fig.
1shows the common structure for BCI system. acquired
brain signals at electrodes located on the scalp or in the
head are processed to obtain determined signal features
.These features are converted to control signals to control
machines. EEG methods commonly executed for BCIs
depend on activity of sensorimotor rhythms (SMR)
measured during movement imagery and permit
establishing sensorimotor rhythm-based BCI (SMR BCI)or
motor-imagery BCI (MI BCI) [7]. experimentally MI-BCI
can control devices like wheelchairs[8]. Wolpaw and
McFarland [9] showed that depending on imagination of
left and right hand movement a cursor two-dimensional
control was possible. The work was increased to a cursor
three-dimensional control depending on imagination of
foot, left, and right hand movement[10]. The introduction
of four-class BCIs enabled users to fly a virtual
helicopter[11] and a robotic quad-copter in three
dimensions [12]. imagining moving and relax the two
hands made the helicopter fly forward and reverse and
imagining moving left and right hand made the helicopter
turn around left and right. The work in [11] was extended
in [13] to a six-class BCI. feet and tongue movement
imagery made the exceed dimensional control. Previous
mentioned studies depend on more than two electrodes .
As shown in [7], handedness has an influence on
sensorimotor rhythm (SMR) distribution and BCI function.
During imagery of left-hand movements, left-handers have
lower precision and poorer SMR suppression in the alpha
band (8–13 Hz). similar results have been observed in our
work where the difference is clearer in case of right than
left hand imagery. In [14] quantitative EEG changes were
extracted from various combinations of channels due to
movement imagination. To determine frequency bands
specific to the subject, a characteristic filter bank popular
spatial algorithm was introduced in [15] to show how EEG
activity changes with left and right hand imagery
movements depending on two channels and this require
further computation.
In this paper, we show a study of how EEG activity
changes with imagery of left and right hand movements
depending on only two electrodes as a step toward
developing BCI based assistive devices and toward
portable BCI. The analysis techniques implemented in our
work include power spectrum, event related potentials and
the EEG signal time frequency analysis. Those analysis
techniques successfully facilitate classification between
imagination of left and right hand movements in a multidimension
way depending on two electrodes. The result
showed that we can classify right and left hand imagery
movements in either frequency, time , or combined timefrequency
domain.
Fig.1. Architecture of an EEG-based BCI
Remainder of paper is arranged in the following
manner: Section II explains methods and materials used.
Section III introduces results and its discussion. The
conclusions and future work are presented in section IV.
)
View on SCiNiTO