BRAIN COMPUTER INTERFACE BY: PRIYANSHI PANDEY B.Tech (I.T) 5 th Sem. 1
BRAIN COMPUTER
INTERFACE
BY:
PRIYANSHI PANDEY
B.Tech (I.T) 5th Sem.
1
What is BCI?
BCI approaches
Implementation of BCI
BCI based real time control of wheelchair
using EEG
BCI in India
BCI in the global market
Conclusion
References
2
• Brain-computer interface (BCI) is a fast-growing emergent technology, in which researchers aim to build a direct channel between the human brain and the computer.
• A Brain Computer Interface (BCI) is a collaboration in which a brain accepts and controls a mechanical device as a natural part of its representation of the body.
• Computer-brain interfaces are designed to operate external devices.
3
4
BCI TYPES
INVASIVENON
INVASIVE
SEMI
INVASIVE
Invasive BCIs are implanted directly into the grey matter of the brain during neurosurgery.
As they rest in the grey matter, invasive devices produce the highest quality signals among
BCI devices.
They are prone to scar- tissue build-up, causing the signal to become weaker.
Even lost as the body reacts to a foreign object in the brain.
5
Invasive BCI
implant
• Easy to wear.
• Do not give rise to any
scar tissue formation.
• Produce poor signal
resolution.
• The skull dampens the
brain waves or signals,
deflecting and blurring
them.
6
Electrodes placed
on scalp
7
Semi -invasive BCI devices are
implanted inside the skull but rest
outside the brain.
They produce signals with better
resolution than those produced in
non-invasive BCI.
As compared to invasive technique
they have lower risk of scar tissue
formation.
Electrodes are embedded in a thin
plastic pad that is placed above the
cortex beneath the Dura mater.
Semi-invasive BCI implant
Over the years invasive technique has been
implemented in the form of BCI device implanted in
the brain of humans and animals.
But the risk of scar tissue formation is always there.
Besides the idea of implanting a device inside a
normal brain is itself disturbing.
Over the years there has been a shift in focus from
invasive techniques to non-invasive techniques.
With improvement in signal processing systems
non-invasive techniques give better result.
Nowadays most research work is in the field of non-
invasive BCI implementation.
8
Implementation of BCI involves six stages: 1)User Training2)Signal Acquisition3)Digitization of signals4)Feature Extraction5)Signal Translation6)Feedback
9
Signal AcquisitionDigitization
Feature Extraction Translation algo
Signal Processing
Device CommandFeedback
The user is trained by instructing them to perform specific cognitive tasks.
The mental tasks to be performed can be:- imaginary motor movements- non-trivial mental arithmetic- visualizing rotation of 3D object- trivial mental arithmetic
The signals produced during these tasks are recorded and the user is trained to focus on performing a specific task to produce required signal.
10
fig: BCI user
14+69=?
The most important element in a bci model is signal acquisition.
Different thought processes give rise to different types of signals in the brain.
The purpose of signal acquisition is to acquire these brain signals.
The device used to acquire these signals is called an EEG device.
11
fig: Signal
acquisition
12
•EEG or electroencephalogram is
a record of the oscillations of
brain electric potentials recorded
from 20 to 256 electrodes
attached to the scalp.
•Key role of EEG is signal
amplification.
•The signals received from
electrodes are minute and to
generate a usable signal they
must be amplified.
fig: An EEG Machine
It uses an array of electrodes attached to the subject’s scalp.
User’s scalp is first prepared with an abrasive paste to remove any dead skin and sweat which may interfere with the signal.
The electrodes used are either gold or silver.
Each electrode has a small amount of conductive paste applied to it , which is then placed underneath a cap .
13
fig: Conductive
paste
fig: EEG cap
Alpha waves(8-13Hz):associated with relaxed state of brain.
Beta waves(13-40Hz):associated with alertness , problem solving and concentration.
Theta waves(4-7Hz):associated with sleep but can also be associated with anxiety, epilepsy, traumatic brain injury.
Delta waves(0-4Hz):associated with deep sleep .
Mu waves(7-11Hz):Mu rhythms are associated with motor cortex and can be used to recognize motor movement.
14
Mu and Beta rhythms
Visual evoked potentials
P300 evoked potentials
EEG signal differences during
different mental calculations
15
The signals acquired are in analog
form.
For use in controlling external devices
these signals need to be digitized.
Before digitization, the signals are
amplified and passed through filtering
circuits that filters out signals in the
required frequency range.
The signals are then converted to
digital form using analog to digital
converters.
16
Analog Signals
Digital Signals
Digitized signals are subjected to a variety of extraction processes.
This analysis extracts the signal features that correspond to user’s message.
BCIs can use signal features that are in the time domain(e.g. P300)or the frequency domain(e.g. mu or beta rhythm amplitudes)
17
Digitized signals
Extracted Signal
Features
The translation algorithm
translates these signal
features into device
commands or orders that
carry out the user’s intent.
Effective algorithms adapt to
user on 3 levels:
-First level of adaptation
-Second level of adaptation
-Third level of adaptation
18
Signal Features
Device
Commands
BCI system provides feedback and
interacts in a productive fashion with
the adaptations the brain makes in
response to that feedback.
The feedback can be in the form of:
-movement of robotic arm
-motion of wheelchair
-word processing
-motion of cursor
screen
-neurofeedback
19
20
CLASSIFICATION OF 5 MENTAL
TASKS
Tasks Classification
Movement Imagery 10000
Trivial
Multiplication
01000
Geometric Figure
Rotation
00100
Nontrivial
Multiplication
00010
Relax 00001
21
22
Path a TM/Forward
MI/Right
MI/Right
MI/Right
R/Stop
Path b TM/Forward
MI/Right
MI/Right
GFR/Left
R/Stop
start
goal
start goal
A B
23Random buttons flashing at periodic time intervals
National Brain Research Centre
Professor , Prof. Neeraj Jain has
been doing research on BCI.
A research project on developing
a brain controlled robot, funded
by DRDO has been taken up by
NBRC.[1]
BCI developed by DAIICT
Professor Mr. Suresh Ranjan has
helped an IIM-A alumnus Mr.
Suresh Karat who had been
paralyzed for 13 years.[2]
24
Prof. Neeraj
Jain
Mr. Suresh
Karat
25
Brainfingers
Brainfingers hardware and software allow you to control your computer totally hands-free.
Mindwave
It safely measures
brainwave signals and
monitors the attention
levels.
EPOC
It uses a set of
sensors to tune into
electric signals
produced by the brain.
26
•BCI has opened new avenues for scientific research.
•It has raised new hopes for patients with ALS,
Locked-in Syndrome and other neurodegenerative
diseases.
•In the past invasive techniques have helped patients to
regain their eyesight and have also helped them to
operate robotic arms .
•The current BCI research has shifted towards non-
invasive techniques.
•Various companies are developing neuroheadsets to give
a new channel of monitoring thoughts and making use of
them in new gaming experience.
•Currently BCI systems are still expensive and away from
reach of many patients.
•Continued research in this field will surely solve all such
problems and prove to be a boon for many.
27
[1]http://neurogadget.com/2011/03/23/indian-scientists-working-
on-brain- controlled-robot-to-help-disabled-patients/1438
[2] http://articles.timesofindia.com/2012-04-
07/news/31304725_1_device-iim-1- graduate-daiict
[3]http://www.jisce.org/attachments/File/Vol-1_Issue-
5/E0143081511.pdf
•http://www.aksioma.org/brainloop/bci_dependent.html
•http://computer.howstuffworks.com/brain-computer-interface2.htm
•http://www.nbrc.ac.in/
•http://www.emotiv.com/
•http://store.neurosky.com
•http://mybrainnotes.com/memory-language-brain.html
28
29