Design and Implementation of a Real-Time Brain-Computer Interface

sciencediscussionAI and Robotics

Oct 20, 2013 (3 years and 10 months ago)

93 views


1


Design and Implementation of a
Real
-
Time Brain
-
Computer Interface


A brain
-
computer interface (BCI) aims to build a communication system that is ca
pable of translating a subject’s intention,
reflected by suitable brain signals, into a control signal.
The
Electroensephalogram (EEG) reflects brain function as brainwaves
and it has been the
primary means of devising BCIs.

The BCI technology has the potential to enable severely disabled people to
drive computers directly by brain activity rather than by physic
al means.

It is specially useful for those who are
totally
paralysed
or have other severe motor disabilities such as people suffering from motor neurone diseases (MNDs) and spinal cord injury
[1]
[2]
.
Nearly two million people in USA alone
[2]

are affected from neuro
-
muscular disorders. A conservative estimate of
overall prevalence is that 1 in 3500 of the world population ma
y be expected to have a disabling inherited neuro
-
muscular
disorder presenting in childhood or in later life
[1]
.
In UK, there are over 5000 patients with some type of
neuro
-
muscular
disorders

[4]
.
In many cases such patients may have no control over any of their muscles.
An appropriate BCI technology can
therefore help improve the standard of living for these people.

A BCI mainly involves

a feature extraction procedur
e (F
E
P), a
translation algorithm (TA), and a
n optional

feedback mechanism

[5]
-
[8]
. The F
E
P obtains the dominant discriminating features
from the EEG activity during t
wo or more specific thought processes. The TA translates

or classifies

these features into a
specific control signal. The optional feedback mechanism helps the user become more efficient in controlling their EEG for
specific tasks (e.g. cursor movement).

A

state
-
of
-
the
-
art
EEG
-
based BCI
experimental

set
-
up

has recently been
bought

and
installed in a custom
-
built
screened lab at the Intelligent Systems Engineering Centre (ISEC)

at Magee campus
.
Using this setup, t
he project
would involve
planning and
conduc
ting experiment on both able
-
bodied and disabled subjects to record data from
sufficient number of trials
.
The recorded data
is to

be used to
desi
gn

a

BCI with an appropriate feature vector and

a
classifier
.
The designed BCI will be implemented in the BCI
experimental setup and applied for on
-
line cursor
control. An appropriate feedback mechanism will be incorporated
and
further extended experimentation

will be
carried out to assess the variability in the designed BCI characteristics. An appropriate adaptat
ion mechanism will
then be devised to account for the variability.


Hardware/software systems involved:

PC under Windows XP, MATLAB, C/C++.


Bibliography

[1]

Kuebler A., Kotchoubey, B., Kaiser, J., Wolpa, J.R., Birbaumer, N., “

Brain
-
computer communication

: u
nlocking the
locked in
”, Psychology Bulletin, 2001, Vol. 127, pp 358
-
375.

[2]

Wolpa
w

J.R., Birbaumer, N., McFarland DJ, Pfurtscheller, G., Vaughan, TM, “
Brain
-
computer interfaces for
c
ommunication and control

“, Clinical Neurophysiology, 2002, Vol. 113, pp 767
-
791.

[3]

Emery, A.E.H., “Population frequencies of inherited neuro
-
muscular diseases


A world survey”, Neuro
-
muscular
disorders, Vol. 1, No. 1, 1991, pp 19
-
29.

[4]

MNDs:
http://news.bbc.co.uk/1/hi/health/1588329.stm
.

[5]

Pfurtscheller G. , Neuper, C., “
Motor imagery and direct brain computer communication
”, Proc. IEEE, 2001; Vol. 89(7),
pp 1123
-
1134.

[6]

Coyle, D.H., Prasad, G., McGi
nnity T.M. (2005)
,

A Time
-
Series Prediction Approach to Extracting Features for a Brain
-
Computer Interface
”, (to appear in) IEEE Trans. On Neural systems and rehabilitation engine
ering.

[7]

Coyle, D.H., Prasad, G., McGinnity T.M. (2005), “
A time
-
frequency approach to feature extraction for a brain
-
computer
interface with a comparative analysis of performance mea
sures
", (to appear in) Applied
Signal Processing
, The
International Journal of the European Association for Signal Processing (EURASIP), (special issue
-

Trends in Brain
-
Computer Interfaces).

[8]

Herman, P., Prasad, G., McGinnity T.M. (2005) “
Investigation of the Type
-
2 Fuzzy Logic Approach to Classification in
an EEG
-
based Brain
-
Computer Interface
”,
(accepted in)
27th International IEEE EMBS Conference, Sept., 2005,
Shanghai, China.

[9]

Sc
herer, R, Muller, GR, Neuper C, Graimann, B., Pfurtscheller, G., “
An Asynchronously controlled EEG
-
based virtual
keyboard: Improvement of the spelling rate
”, IEEE Trans. on Neura
l Systems and Rehabilitation Engineering, 2004, Vol.
12, No.6, pp 979
-
984.