Vous êtes sur la page 1sur 24

NON INVASIVE METHODS FOR

BRAIN COMPUTER INTERFACE

Under the guidance of :


Prof. (Dr.) JYOTI SAXENA and
Er SUKHJINDER SINGH
1314284

Presented by:
Ramandeep Kaur
Uni. Roll No.
M.Tech. ECE

(FT)

Brain Computer
Interface
Brain Computer Interface is

a direct
brain and

communication pathway between


computer.
A BCI is a technology which allows a human to
control a computer, peripheral or other
electronic device with thought.
It does so by using electrodes to detect
electric signals in the brain which are sent to a
computer.
The computer then translates these electric
signals into data which is used to control a

Types of BCI
BCI can be classified into three following

categories:
Brain
invasive
computer
BCIinterface (BCI)

Invasive BCI
Invasive

strategies are the brain computer


interfaces which are implanted inside the brain of
the subject surgically which are based on the use of
array of microelectrodes implanted into visual
cortex. These implants provides highest quality
signals and capable to give better temporal and
spatial resolution. Invasive BCI mostly provide
useful functionality to paralyzed people. These can
provide restoration of vision by concentrating or
linking brain with external cameras and can also
restore use of limbs by using brain controlled
robotic legs and arms.

Partial BCI
The

partially invasive mechanisms are


implanted inside the skull but rest outside the
brain slightly within the grey matter. When we
compare it with Invasive BCI it is one bit
weaker than Invasive. Electrocoticography
(ECoG) is example of partially Invasive BCI
technique.

Non Invasive BCI


Non-invasive strategies deal externally with the brain

signals. Non-invasive BCI do not employ surgical


implants. Non-invasive BCI devices are considered to be
safest and are low cost. These have the slightest or least
signal clarity when communicating with skull distort
signals. The signals are detected by placing some
electrodes
at
specific
positions
of
the
skull.
Electroencelography
(EEG),
Functional
magnetoresonance imaging (fMRI), Magnetoencelography (MEG),
magnetic resonance imaging (MRI), computerized
tomography (CT), near infrared spectroscopy (NIRS) etc.
are some of the examples of non-invasive BCI.

Non Invasive Methods for Brain


monitoring
Electroencelography (EEG)
Funtional Magnetic Resonance Imaging (fMRI)
Magnetoencelography (MEG)
Near Infrared Spectroscopy (NIRS)

Electroencelography
(EEG)
Electroencephalography

is a non-invasive
method of measuring the bioelectrical activity
of the brain. Signals are acquired through
electrodes placed on the surface of the scalp
which detect potential changes caused by the
activity of neurons of the cerebral cortex. EEG
is very useful to monitor and diagnose
epilepsy, sleep disorders, head trauma, brain
tumours disorders of consciousness and other
brain conditions.

Functional Magnetic Resonance


Imaging (fMRI)
Functional Magneto-Resonance Imaging is a non-

invasive neuroimaging technique which detects


changes in local cerebral blood volume, cerebral
blood flow and oxygenation levels during neural
activation by means of electromagnetic fields. fMRI
is generally performed using MRI scanners which
apply electromagnetic fields of strength in the
order of 3T or 7T. The main advantage of the use
of fMRI is high space resolution. fMRI appears
unsuitable for rapid communication in BCI systems
and is highly susceptible to head motion artefacts.

Methodology
brain signals data

Current and Future


Applications
Rehabilitation: The Rehabilitation idealize to

enhance life quality of human who are


paralyzed completely or partially and suffers
difficulties due to Amyotrophic Lateral Sclerosis
(ALS),
Cranium
Traumatism,
Locked
in
Syndrome, severe cerebral paralysis etc. after
losing all facilities or potentials of voluntary
movements, these patients go into state where
communication is impossible. These human
can communicate with world and external
devices.

Communication & control : BCIs provide options

for communication and control for people with


devastating neuromuscular disorders (such as
amyotrophic lateral sclerosis, or ALS, brainstem
stroke, cerebral palsy, and spinal cord injury).
Gaming: The future gamers will be able to drive
cars and steer just by thinking "go left" or "go
right" or "brake". BCI also can be used to enjoy
your home theatre by handling an entertainment
system with your thoughts which will make a
remote control obsolete.

Military Defence System: DARPA (Defence

Advanced Research Projects Agency) has been


interested in Brain-Machine-Interfaces from
many
years for military applications like
wiring fighter pilots directly to their planes to
allow autonomous flight from the safety of the
ground.

Advantages
Help people with severe debilitating muscle

wasting diseases, or with the so-called


locked-in syndrome, to communicate.
Help in direct brain communication in
completely paralyzed patients.
Help to those who suffers from disorders like
ALS
(Amyotrophic
lateral
sclerosis)
,
Brainstem stroke, brain or spinal cord injury.

Limitations
Electrodes: Electrodes to be attached inside the skull or even

implanted in the brain. Dr James admitted that this opened up


many ethical problems.
Liability: Most people would agree that, under normal
circumstances, we are fully responsible for our actions.
However, if our intent was affected by a braincomputer
interface, incorrect actions may be produced simply by
incorrect detection of correct intent.
Privacy: The capacity to induce information into the brain
may provide us with the ability to base our actions on a better
assessment of the environment. Because this information is
provided by a computer, it could be accessed and modified by
third parties, which may allow them to influence our actions.

Literature Survey
F Lotte, M Congedo, A Lecuyer, F Lamarche and B

Arnaldi surveyed the classification algorithm used to design


BCI system based on EEG. The algorithms were divided into
five categories: linear classifiers, nearest neighbour classifier,
combinations of classifiers, neural networks and non-linear
Bayesian classifiers. The result obtained, in BCI content were
analysed and compound providing readers with guidelines to
choose or design a classifier for a BCI system. It seemed that
SVM are greatly efficient for synchronous BCI because of
their regularization property and their immunity to curse of
dimensionality. For synchronous experiments, combinations
of classifiers and dynamic classifiers seems to be very
effective. They present that band power (BP) features to be
efficient for classification of motor imagery.

Mohamed

Mostapa Fouad, Khalid Mohamed Amin,


Nashwa EL-Bendary and Aboul Ella Hassain reviewed brain
computer interfaces. A BCI system can allow the encephalic
activity to control the computers and external devices such as
prosthetics, BCI aims to provide aid for people buffering
neuromuscular diseases as computer could permit them to
perform different tasks, from accessing computer-based games
to communication. Advancements in neural prosthetics have led
interest in the use of BCI .the research aims at the development
of devices that could be controlled by brain signals. Noninvasive method ranges were defined in the research of
electroencephalogram (EEG) and electrocorticographic (ECoG)
electrode array, the information regarding the state of art in
neuroimaging based approaches and their applications were
provided. Evaluation of different classification methodology
applied to the brains captured signals was done, BCI is directed
towards robot industry so that solutions could come up where
robot could sense and act like human.

Rabie A Ramadan, S Refat, Marwa AElshahed and

Rasha A Ali surveyed that BCI reads the waves generated


from the brain at different locations in the human head
translates these into actions and commands that can
control the computer. The classification of BCI is alone into
three main categories depending on the way the electrical
signals obtained from neuron cells in the brain. There are
plenty of sig. which can be used for BCI and these signals
can be divided into two categories field potentials and
spikes component of particular interest to BCI could be
divided into four classes which are oscillatory EEG activity,
Event related potentials (ERPS), slow cortical potentials
(SCP), and neural potentials. Many techniques have been
employed to monitor brain activities and each technique
have their own characteristics and related pros and cons.

Objective of Thesis
The objective of this dissertation is to carry

out the comparison of the existing techniques


associated with non-invasive brain computer
interfaces. In this thesis work, EEG and fMRI
techniques
shall
be
studied
for
the
comparison of non-invasive brain computer
interfaces. The study of these techniques will
provide a broader perspective in better
understanding of the brain computer interface
for further research in future applications.

References
Minakshi and Peter Gill, (2014), Review on: Brain

Computing Interface, International


Journal of
Engineering Research, 2(6), pp. 131-135.
Garima Singh and Manju Kaushik (2012), Brain to
Brain Communication: Without any Interface
images, thoughts can be exchanged between
minds, International Journal of Computer Science
& Engineering Technology (IJCSET), 3(9), pp. 411414.
Luis Fernando Nicolas-Alonso and Jaime Gomez-Gil
(2012), Brain Computer Interfaces, a Review,
Sensors, 12(10), pp. 1211-1279.

Remigiusz J.Rak, Marcin Kolodeziej and Andrej

Majkowski (2012), Brain-Computer Interface as


Measurement and Control System the Review
Paper, Metrology andMeasurement Systems,
19(3), pp. 427-444.
Tushar
Kanti
Bera
(2015),
Non-invasive
Electromagnetic Methods for Brain Monitoring : A
Technical
Review,
Springer
International
Publishing, Switzerland, 74(3), pp. 51-95.
Rabie A. Ramadan, S. Refat, Marwa A. Elshahed
and Rasha A. Ali (2015), Basics of Brain
Computer Interface, Springer International
Publishing, Switzerland, 74(3), pp. 31-50.

Mohamed Mostafa Fouad, Khalid Mohamed

Amin, Nashwa El-Bendary and Aboul Ella


Hassanien (2015), Brain Computer Interface:
A Review, Springer International Publishing,
Switzerland, 74(3), pp. 3-30.
F. Lotte, M. Congedo, A. Lcuyer, F. Lamarche,
and B. Arnaldi (2007), A review of
Classification algorithms for EEG-based braincomputer interfaces, Journal of Neural
Engineering, 4(7),pp. 1-13.

THANK YOU

Vous aimerez peut-être aussi