Chapter 2

Basics of Brain Computer Interface

Rabie A. Ramadan, S. Refat, Marwa A. Elshahed and Rasha A. Ali

Abstract 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. It is a collaboration in which a brain accepts and controls a mechanical
device as a natural part of its representation of the body. The BCI can lead to many
applications especially for disabled persons. Most of these applications are related
to disable persons in which they can help them in living as normal people.
Wheelchair control is one of the famous applications in this eld. In addition, the
BCI research aims to emulate the human brain. This would be benecial in many
elds including the Articial Intelligence and Computational Intelligence.
Throughout this chapter, an introduction to the main concepts behind the BCI is
given, the concepts of the brain anatomy is explained, and the BCI different signals
are stated. In addition, the used hardware and software for the BCI are elaborated.

Keywords Brain computer interface Systems of BCI BCI monitoring hard ware


and software BCI trends

R.A. Ramadan (&)

Computer Engineering Department, Cairo University, Cairo, Egypt
e-mail: [email protected]
S. Refat
M.A. Elshahed
R.A. Ali
Physics Department, Faculty of Women for Arts, Sciences and Education,
Ain Shams University, Cairo, Egypt
e-mail: [email protected]
M.A. Elshahed
e-mail: [email protected]
R.A. Ali
e-mail: [email protected]

Springer International Publishing Switzerland 2015 31

A.E. Hassanien and A.T. Azar (eds.), Brain-Computer Interfaces,
Intelligent Systems Reference Library 74, DOI 10.1007/978-3-319-10978-7_2
32 R.A. Ramadan et al.

2.1 Introduction

Brain Computer Interface (BCI) is a direct connection between computer(s) and

human brain. It is the most recent development of Human Computer Interface
(HCI). Unlike the traditional input devices (keyboard, mouse, pen etc.), the BCI
reads the waves produced from the brain at different locations in the human head,
translates these signals into actions, and commands that can control the computer
(s). The BCI can lead to many applications especially for disabled persons such as
[1]: (1) new ways for gamers to play games using their heads, (2) social interac-
tions; enabling social applications to capture feelings and emotions, (3) helping
partially or fully-disabled people to interact with different computational devices,
and (4) helping understanding more about brain activities and human neural net-
works. These applications depend on the basic understanding of how the brain
works. BCI applications utilize the brain and its nervous system functions where the
humans central nervous system consists of the spinal cord and the brain. One of its
tasks is to process and integrate incoming sensory stimuli received via peripheral
nerves and to give impulses back to actuators, e.g. to muscles or glands which cause
automatic or voluntary action. Furthermore the central nervous system, particularly
the brain, is responsible for higher integrative abilities such as thinking, learning,
production, and understanding of speech, memory, emotion etc. Finally vegetative
functions such as respiration and the cardio-vascular system are controlled by the
central nervous system.
The brain computer interface was not studied only for human but also for
animals. A Monkey in 2008 [2] was able to move a screen cursor as well as
controlling a robot arm. The benet of such study is to know how animals can think
and discover their brains as well. In addition, BCI is used with different human
patients capturing their brain signals. The BCI science goes beyond a communi-
cation tool for people are not able to communicate. It is gaining more attention from
healthy people for other purposes such as rehabilitation or hands-free gaming.
However, BCI tools still limited and need expert to deal with them which is one of
the BCI research challenges.
However, there are many challenges that faces the BCI when used in real world
tasks as follows:
(1) Low BCI signal strength: it has been noticed that extracting signals from the
brain is not an easy task since the signal strength in most of the cases are low.
In most of the cases, signal amplication is required. Many of the used toolkits
include such ampliers where some others do not include good ampliers.
(2) Data transfer rate (bandwidth): the best data transfer rate from a subject was 3
characters. Certainly, this is very low data transfer that makes the BCI
applications suffer from fast response as well as accurate control.
(3) High error rate: it is obvious that due to the low data transfer rate and the low
signal strength, the error percentage became high. In addition, the brain signal
is very high variability. Therefore, the expected error rate is high.
2 Basics of Brain Computer Interface 33

(4) Inaccurate signal classication: brain have some centers that signals can be
captured from them using electrodes. Classifying the captured signals suffer
from high interference and inaccurate classication. There are many signal
classication techniques are utilized including the computational intelligence
techniques that are recently proposed by authors in [3].
The goal of this chapter is to provide the reader with basic concepts of the BCI as
a science and from it came from. The anatomy of the brain is elaborated for better
understanding to the brain signals. The chapter goes on by providing the different
signals that were able to be captured recently. These signals are classied and their
characteristics are explained. Finally, the chapter explains the current hardware and
software components of the BCI including the commercial devices and their
properties.

2.2 Brain Anatomy

Amazingly, nothing in the world can be compared with the human brain. The three-
pound organ controls all body functions including receiving and interpreting
information from the outside world, and expressing the essence of the mind and
soul. Intelligence, creativity, emotion, and memories are a few of the many things
governed by the brain. The brain receives information through different sensors
such as sight, smell, touch, taste, and hearing. The brain constructs the received data
from the different sensors and form a meaningful message. The brain controls our
body movement of the arms and legs, thoughts, memory and speech. It also
determines how a human respond to different situations such as stress by regulating
our heart and breathing rate.
As it is known, the nervous system is another essential system in the human
body. The nervous system divided into central and peripheral systems. The central
nervous system is composed of two main parts which are the brain and spinal cord.
The peripheral nervous system is composed of spinal nerves that branch from the
spinal cord and cranial nerves that branch from the brain. The peripheral nervous
system includes the autonomic nervous system, which controls vital functions such
as breathing, digestion, heart rate, and secretion of hormones.
The brain skull represents the shield of the brain from injury. It is formed from 8
bones. These bones include the frontal, two parietal, two temporal, sphenoid,
occipital and ethmoid. The face is formed from 14 paired bones including the maxilla,
zygoma, nasal, palatine, lacrimal, inferior nasal conchae, mandible, and vomer [4].
Anatomically ve basic parts of the brain can be distinguished including
Cerebrum, Diencephalon, Cerebellum, Mesencephalon, and Medulla oblongata as
shown in Fig. 2.1. The cerebrum, located directly under the skull surface, is the
largest part of the brain. Its main functions are: (1) the initiation of complex
movement, (2) speech and language understanding and production, (3) memory,
and (4) reasoning. Brain monitoring techniques which make use of sensors placed
34 R.A. Ramadan et al.

Fig. 2.1 Brain anatomy [8]

on the scalp mainly record activities from the outermost part of the cerebrum; the
cortex. More inside the cerebrum the basal ganglions can be found which consists
of a number of nuclei controlling the direction of slow movements [5]. Also the
thalamus is located here which directs sensory information to appropriate parts of
the cortex. The second part of the brain is the Diencephalon. One important
function of the diencephalon is the forwarding of sensory information to other brain
areas. Besides that, it contains the hypothalamus which controls the body tem-
perature, the water balance and the ingestion to assure the state of homeostasis for
the body, i.e. good working conditions for all body cells. The coordination of all
kinds of movements is done in the third part which is the cerebellum. Therefore, it
cooperates closely with structures from the cerebrum (e.g. the basal ganglions).
Cerebellum and Cerebrum are connected via the Pons. However, the largest part of
the reticular system is located in the Mesencephalon where it controls vigilance and
the sleep-wake rhythm.
The Medulla Oblongata connects the brain with the spinal cord. Respiration and
the cardiovascular system are controlled by that part of the central nervous system.
Furthermore, a huge number of peripheral nerves pass through the medulla
oblongata. Compared to the brains of other mammals, the human brain has the
largest and best developed cortex. Neural processes related to abilities like complex
reasoning, speech and language etc. which distinguish humans from other mammals
take place in that part of the brain [6, 7].
Moreover, the cortex consists of two hemispheres which are connected via a
beam called corpus callosum. Each hemisphere is dominant for specic abilities.
For right handed persons, the right hemisphere is activated more during the rec-
ognition of geometric patterns, spatial orientation, the use nonverbal memory and
the recognition of non-verbal noises [8]. More activity in the left hemisphere can be
observed during the recognition of letters and words, the use verbal memory and
auditory perception of words and language. Each hemisphere is partitioned into ve
anatomically well-dened regions, the so called lobes as given in Fig. 2.2.
2 Basics of Brain Computer Interface 35

Fig. 2.2 Hemisphere

partitions [8]

2.3 Brain Computer Interface Types

Brain computer interface can be classied into three main groups as shown in
Fig. 2.3. Classication depends on the way that the electrical signal is obtained
from neuron cells in the human brain.

2.3.1 Invasive BCI Acquisition Techniques

In invasive BCI techniques, special devices have to be used to capture the brain
signals. Such devices are called Invasive BCI devices; devices that are based on
detecting from single area of brain cells is called single unit while the detection
from multiple areas is called multi-units [9]. Invasive BCI devices are inserted
directly into the human brain by a critical surgery as can be seen in Fig. 2.4. The
electro-corticogram (ECoG) are the obtained signals from these inserted electrodes
[10]. These devices have the highest quality of human brain signals but have the
risk of forming scar tissue.

Fig. 2.3 Brain computer interface types

36 R.A. Ramadan et al.

Fig. 2.4 Invasive BCI

electrodes [10]

Fig. 2.5 Partially invasive

BCI electrodes

2.3.2 Partially Invasive BCI Acquisition Techniques

Other devices that can capture the signal from the brain are the partially invasive
BCI devices. Devices are inserted in the skull on the top of human brain as depicted
in Fig. 2.5. These devices have bit weaker quality of human brain signals than
invasive BCIs and have less risk of forming scar tissue [11, 12].

2.3.3 Non Invasive BCI Acquisition Techniques

Non Invasive BCI devices are considered the safest type and low cost type of
devices. However, these devices have weaker human brain signals than other BCI
devices due to the skull. The detection of signals is done by some electrodes placed
on the scalp as given in Fig. 2.6. At the same time, placing such electrodes is easy
as well as portable. Most noninvasive techniques are constructed by recording
ElectroEncephaloGraphs (EEG) from the scalp. Recent EEG Non Invasive BCI
2 Basics of Brain Computer Interface 37

Fig. 2.6 A wireless non-

invasive signal capturing
device

devices have better temporal resolution due to use up to 256 electrodes on the
whole area of the human scalp. While others, Non Invasive BCI devices, use
functional Magneto-Resonance Imaging (fMRI), Positron Electron Tomography
(PET), MagnetoEncephaloGraphy (MEG) and Single Photon Emission Computed
Tomography (SPECT) [13, 14].

2.4 Types of BCI Signals

The brain generates an amount of neural activity. There are a plethora of signals,
which can be used for BCI. These signals are divide into two classes: spikes and
eld potentials [11]. Spikes reect the action potentials of individual neurons and
acquired through microelectrodes implanted by invasive techniques. Field poten-
tials are measure of combined synaptic, neuronal, and axonal activity of groups of
neurons and can be measured by EEG or implanted electrodes. The following is the
classication of EEG signals based on their frequencies/bands [15, 10].
Delta Signal. It is captured within the frequency range of 0.53.5 Hz. It tends to
be the highest in amplitude and the slowest waves. It is seen normally in adults
in slow wave sleep as well as in babies. A sample from the Delta signals is
shown in Fig. 2.7.
Theta. The frequency of this signals ranges from 3.5 to 7.5 Hz. Theta is linked to
inefciency and daydreaming. In fact, the very lowest waves of theta represent
the ne line between being awake or in a sleep. However, as shown in Fig. 2.8,
high levels of theta are considered abnormal in adults.

Fig. 2.7 Delta wave sample

38 R.A. Ramadan et al.

Fig. 2.8 Theta wave sample

Fig. 2.9 Alpha wave sample

Fig. 2.10 Beta wave

Fig. 2.11 Gamma wave

Alpha. As shown in Fig. 2.9, this signal frequency ranges from 7.5 to 12 Hz.
Hans Berger [12] named the rst rhythmic EEG activity he saw, the alpha
wave. Range seen in the posterior regions of the head on both sides, being
higher in amplitude on the dominant side. It is brought out by closing the eyes
and by relaxation. Several studies have found a rise in alpha power after
smoking marijuana.
Beta. Beta is another brain signal in which its frequency ranges from 12 Hz to
about 30 Hz. It is seen usually on both sides in a symmetrical distribution and it
is most evident frontally. Beta waves are often divided into 1 and 2 to get
more specic range. The waves are small and fast when resisting or suppressing
movement, or solving a math task. It has been noticed in these cases that there is
an increase of beta activity. The shape of such signal is shown in Fig. 2.10.
Gamma. It is a signal with frequency range of 31 Hz and up. It reects the
mechanism of consciousness. Figure 2.11 shows the shape of the Gamma signal.

2.5 Components of Interest

Components of particular interest to BCI can be divided into four categories which
are oscillatory EEG activity, event-related potentials (ERP), slow cortical potentials
(SCP), and neuronal potentials.
2 Basics of Brain Computer Interface 39

2.5.1 Oscillatory EEG Activity

Oscillatory EEG activity is caused by complex network of neurons that create

feedback loops. The synchronized ring of the neurons in these feedback loops
generates observable oscillations. There are two distinct oscillations of interest
which are: (1) the Rolandic mu-rhythm, in the range 1012 Hz, and (2) the central
beta rhythm, in the range 1418 Hz. This activity represents idling or rest state
[10].

2.5.2 Event-Related Potentials

Event-Related Potentials (ERPs) are time-locked responses by the brain that occur
at a xed time after a particular external or internal event. These potentials occur
when subjected to sensory, mental event, or the omission of a constantly occurring
stimulus. Exogenous ERP components occur due to processing of the external event
but independent of the role of the stimuli in the processing of information. On the
other hand, Endogenous ERP components occur when an internal event is pro-
cessed. It dependents on the role of the stimulus in the task and the relationship
between the stimulus and the context in which it occurred [10]. The ERP events can
be classied as follows:
Event-Related Synchronization/(De) synchronization
A particular type of ERP is characterized by the occurrence of an event-related
desynchronization (ERD) and an event-related synchronization (ERS). A decrease
in the synchronization of neurons causes decrease of power in specic frequency
bands. This phenomenon is dened as an ERD and can be identied by a decrease
in signal amplitude. ERS is characterized by an increase of power in specic
frequency bands that is generated by an increase in the synchronization of neurons
and/or in signal amplitude.
Visual-Evoked Potentials
Another type of ERF commonly used in BCI is the visual-evoked potential
(VEP), an EEG component that occurs in response to a visual stimulus. VEPs are
dependent on the users control of their gaze and thus require coherent muscular
control [16]. P300 is ERP component elicited in the process of decision making.
The P300 is thought to reect processes involved in stimulus evaluation or cate-
gorization. It is usually elicited using the oddball paradigm, in which low-proba-
bility target items are mixed with high-probability non-target item [17]. The user is
presented with a task that cannot be accomplished without categorization into both
categories. When an event from the rare category is displayed, it elicits a P300
component, which is a large positive wave that occurs approximately 300 ms after
event onset [10].
40 R.A. Ramadan et al.

Slow Cortical Potential

It is the slow cortical potential, which is caused by shifts in the depolarization
levels of certain dendrites. Negative SCP indicates the sum of synchronized
potentials, but positive SCP indicates the reduction of synchronized potentials from
the dendrites.
Neuronal Potential
Neuronal potential is a voltage spike from individual neurons. This potential can
be measured for a particular neuron or a group of neurons. The signal is a measure
of the average rate, correlation, and temporal pattern of the neuronal ring.
Learning can be measured through changes in the average ring rate of neurons
located in the cortical areas associated with the task [8].

2.6 Monitoring Brain Activity Using EEG

Several techniques have been used to monitor brain activities such as (1) Electro-
encephalography (EEG), (2) Magnetoencephalography (MEG), (3) Functional
Magnetic Resonance Imaging (fMRI), (4) Functional Near-Infrared Spectroscopy
(fNIRS), (5) Single Photon Emission Tomography (SPECT), and (6) Proton
Emission Tomography (PET). Each method has its own characteristics as well as
pros and cons. However, for several reasons the potential differences which can be
measured between two points of the scalp are very different from those could be
measured when electrodes were implanted directly in the brain. For instance, the
activity of the potential generators could be measured directly by:
1. A superposition of potentials generated in different areas of the cortex is mea-
sured using scalp electrodes since brain tissue and the liquor are conductive.
2. The amplitude of the originally generated potential differences is attenuated
because of the resistive properties of the tissue between the potential generators
and the electrode (e.g. liquor, skin, bone of the skull).
3. Capacities caused by cell membranes and other inhomogeneities (e.g. liquor-
skull, skull-skin) between potential generators and electrodes inuence the
amplitude of the EEG signals as a function of their frequency.
Therefore, the positions for EEG electrodes should be chosen in a way, which all
cortex regions are covered. For most applications, this is usually the whole cortex.
An internationally accepted standard for electrode placements is the 1020 system
(electrodes are placed at distances of 10 or 20 % of the length of several connec-
tions between some reference points) introduced in 1957 by the International EEG
Federation [18]. Electrodes were placed according to the 1020 system. Three
anatomical reference points must be determined before the 1020 system electrode
positions which are:
2 Basics of Brain Computer Interface 41

1. Nasion: the onset of the nose on the skull, below the forehead.
2. Inion: the bony protuberance which marks the transition between skull and
neck.
3. Pre-auricular reference point: located before the cartilaginous protrusion of the
acoustic meatus (the auditory canal).
Figure 2.12 shows the electrode positions of the 1020 system in their projection
on the cortex. The name for a particular electrode position reects the anatomical
region of the cortex above which it is located. Fp stands for frontopolar, F stands for
frontal, T stands for temporal, C stands for central, P stands for parietal, O stands
for occipital and A stands for auricular while G denotes the ground electrode. Even
numbers denote the right part of the head, odd numbers refer the left part.
Generally, there are two categories of artifacts can be distinguished in EEG
measurements which are biological and technical. The biological artifacts are
caused by the recorded subject and technical artifacts are caused by the EEG
recording device. The sources of many biological artifacts are dipoles originating
for example from muscular activities which are much stronger than the EEG related

Fig. 2.12 Electrode position. a Brain electrods. b top view of the brain and electrods
positions [12]
42 R.A. Ramadan et al.

dipoles. A superposition of both types of dipoles causes artifacts in the signal which
are often characterized by large peaks or uctuations of a particular morphology.
Sometimes, however, they can hardly be distinguished from the actual EEG. Other
biological artifacts inuence the contact between skin and electrode or the electrical
properties of the medium between potential generators and electrodes.

2.7 BCI System

Forming a BCI system requires following three main steps as shown in Fig. 2.13:
Step 1 is the signal acquistion, Step 2 is the signal processing, and Step 3 is the
data manipulation.
Step 3: Using these obtained signals to control in external devices or computer
depending on the application.
Step 1: Signal Acquisition
Signal acquisition process is required to capture the brain electric signals. The
electric signals could be recorded from the scalp, the surface of the brain, or from
the neural activity. Since the capture signals strength are usually low, they need to
be amplied. Then, to be used by computer applications, they need to be digitized.
Step 2: Signal Processing
In this step, obtained signals in step 1 are analyze to get the control signals.
Signal processing could be done through some other sub operations as follows:
Preprocessing
The rst part of signal processing is preparing the recording electric signal for
processing like enhancement to make the features clear for detection. Some ltering
techniques could be used in the preprocessing operation.
Feature extraction
Simply, feature extraction means extracting specic signal features. EEG
recordings not only contain electrical signals from the brain, but also several

Fig. 2.13 BCI signal processing

2 Basics of Brain Computer Interface 43

unwanted signals. Those unwanted signals may bias the analysis of the EEG and
may lead to wrong conclusions. Therefore, the digitized signals are subjected to
feature extraction procedures.
Signal Classication: translation algorithm
The next stage, the translation algorithm, in which it translates the extracted
signal features into device commands orders that carry out the users intent. The
signals are classied on both frequency and on their shape; the classication
algorithm might use linear methods or nonlinear methods.
Step 3: Data Manipulation
Once the signals are classied, the output is manipulated to suite the output
devices (e.g. computer screen).

2.8 BCI Monitoring Hardware and Software

BCI signals are very week signals that need special treatments to be handled
correctly. The strength of the measured signals is usually between 1 V and
100 mV along with the scalp impedance and other noises. In order to receive such
signals and display them on digital formats, suitable ampliers should be used.
Therefore, the BCI hardware can be divided into three classes; the rst class is the
electrodes while the second class is the signal ampliers. The third class is the real
time signal handling. Throughout this section, a brief description to each class is
provided.
The EEG measurement electors are usually made of gold or Ag/AgCl. The gold
electrodes are effective in measuring EEG, EMG or ECG signals as well. However,
the Ag/AgCl electrodes proved to be more effective when the EEG frequencies
below 0.1 Hz. In addition, there are two types of electrodes which are active and
passive electrodes. The active electrodes contain an amplier with gain 110 inside
it in which it reduces the noise and cable interferences. On the other hand, passive
electrodes do not include any ampliers in electrodes service. Such electrodes are
usually distributed on the scalp from 10 to 20 electrodes in most of the cases
(Fig. 2.14).

Fig. 2.14 Gold electrodes

[13]
44 R.A. Ramadan et al.

Fig. 2.15 Electrodes signal lters

Fig. 2.16 Electrodes

amplier

The other part of the BCI hardware is the biological signal amplier. It is one of
the important parts of physiological recording and analysis in which the brain
signals are very weak and it is used to amplify them. Figure 2.15 [19] is a sample of
BCI amplier. As can be seen in the Figure, signals captured by electrodes are
amplied through handling by the input stage component to remove the possible
noise produced from electrode-skin interfaces. In fact, it is a pre-amplied circuit
that could be simply based on simple op-amp OPA2277 devices as shown in
Fig. 2.16. The signal is also passed through two lters which are Low-pass lter
and Notch lter. After all, the signal is post amplied.
Real time signal recording and analysis is managed on different Operating
Systems including windows and Linux as well as Mac OS. C++ is one of the most
used language for analysis over C++ LabVIEW (National Instruments Corp.,
Austin, TX, USA) and MATLAB (The MathWorks Inc., Natick, USA) are mostly
used as programming languages. Different signals are utilized to control many
applications. There is some commercial software and hardware kits are already used
in some of the BCI applications. One of the software kits is the neurobci [8] in
which it allows users to develop their own Brain Computer Interface (BCI), bio- or
neurofeedback application, as created in Html/Jscript, C++, or Matlab. FieldTrip
2 Basics of Brain Computer Interface 45

Fig. 2.17 Emotive headset

[20] is a MATLAB toolbox that is used for analysis. It utilizes TCP connections for
multiple clients as at the same time.
DataSuite [21] is another software tool for data acquisition. The DataSuite
consists of two parts including the DataRiver and MatRiver. DataRiver is a data
management and synchronization real time engine while MatRiver is a MATLAB
client toolbox for DataRiver. The following Figure shows the data ow of Da-
taSuite. For more details about other BCI tools, the reader is encouraged to read the
survey by Arnaud et al. [22] Fig. 2.17: DataSuite data ow. Two computers each
running an instance of DataRiver are represented. One acquires data (left); the other
(right) uses MatRiver to perform data classication and feedback visualization.
Dashed lines indicate control signals.
Emotiv EEG neuroheadset [23] is a wireless BCI set; this set is a neuro-signal
acquisition and processing wireless neuroheadset. The set can be wirelessly con-
nected to a computer. One advantage of the set is it has 14 saline sensors offer
optimal positioning for accurate spatial resolution.
ModularEEG [24] is another EEG hardware created by the OpenEEG hardware
developers. The modularEEG has two or more EEG ampliers, and a 6-channel
signal capture board that connects to a PC via a standard serial cable. The modu-
larEEG has two types of electrodes which are active and passive electrodes. Some
skin preparation is required while there is no preparation is required when active
electrodes are used. Figure 2.18 shows the modularEEG board and the active
electrodes.

2.9 Brain Computer Interface Applications

BCI is interesting area to researchers because it can solve many problems which
seem to be impossible. The essential target of BCI applications is to convert the
users intent or thoughts to an action in external device or computer and control to
46 R.A. Ramadan et al.

Fig. 2.18 ModularEEG

device

these devices. Many applications of BCI concerned on patients suffer from disor-
ders of consciousness (DOC).These patients unable to make communication with
their around world [25].
By using BCI, these patients can control some devices to perform basic and
important jobs they need without helping like moving with wheelchair, getting
something for eating or drinking by using robotic legs or arms controlled by brain.
BCI technologies are used to restore the vision to blinds by connecting an external
camera with brain [26]. Applications on device control not include patients only,
but also healthy users like whose needs to perform many jobs at the same time like
divers, astronauts and drivers where they keep their hands on swimming, operate
equipment and the steering wheel [27].
Rabie et al. [28] developed a BCI based system that can help disabled persons to
use the web through their brains only. The authors developed a technique that
captures the eye signals through the brain to select the appropriate letters as well as
words to be written on the web browser. Another application that has been
developed is the wheelchair simulator that is controlled also by the BCI signals.
BCI used also on User-state monitoring which make alert to sleepy drivers or
students. Also, it extended physically to measure the heart beats for users. Many
applications focused in entertainment and playing games especially after using 3D
monitors, certain glasses and an EEG headset where the control on the game by
thoughts. EEG combined sometimes with eye movement on some applications for
security and safety where the system can monitor suspicious objects, deviant
behavior or arousal state. A common BCI application is neurofeedback training to
2 Basics of Brain Computer Interface 47

improve working, attention, executive functions and memory. Neuroergonomics is

an evaluation application used to estimate how well human abilities match a
technology. BCI used also in education and training techniques [25]. By sing BCI
based on EEG, patient can control or move the cursor by mental thoughts where the
patient can select words or letters [26].

2.10 BCI Trends

The BCI technology has achieved many goals in different working areas of medical
and nonmedical during the last ten decades. Researchers in this led are looking in
the future for more development in trends and applications. Since the current trends
are focused on utilizing of motor system which is related to: (1) the electroen-
cephalogram (EEG) signals for neurorehabilitation, (2) controlling of robot and
exoskeleton based on EEG signal, (3) implementation of BCI component on eld
programmable and recongurable computing system, and (4) solving the interop-
erability and standardization issues of BCI software platforms. Researchers looking
for developing a common le format for BCI data exchange and introducing an
accurate and robust pre-processing and feature extraction techniques of BCI signals.
One of the most future trends is introducing a new kind of sensory modality that
is more accurate and safe [29]. The revolution in nanotechnology will contribute in
the progress of BCI by producing a smaller and far superior chips that can implant
safely in the brain to yield high quality signals. The target is to increase the BCI
reliability and accuracy to be clinically useful. Moreover, a wireless brain implant is
an important technology today that lets people with mobility problems control a
computer or wheelchair with their thoughts. The wireless brain sensor can record
the activity of dozens of neurons in freely moving subjects [30]. Even though
wireless BCI systems may provide a number of advantages. There are still many
issues that need to be resolved including improving signal quality, more compact
and stylish system designs, and implementation of useful applications [31].
BCIs driven by auditory stimuli are a relatively new phenomenon. With some
key publications over the last 5 years, the auditory BCI approach has gained and
continues to gain momentum. It is underpinned by the BCI communitys efforts to
nd alternatives to the traditional BCI paradigms to meet the needs of end users
who require a non-visual communication system. The trends now are seeking to
study the effect of BCI auditory not only in communication eld but also in
attention monitoring and neurofeedback training to improve performance. In
addition, how BCI auditory can contribute in diagnosis and treatment of disorders
that have an auditory component is another issue to be handled.
Another trend now is Tactile and Bone-Conduction based BCI Paradigms. It has
been proposed to offer alternative ways to deliver sensory stimulation inputs which
could be crucial for patients suffering from weak or lost eye-sight or hearing.
Already several preliminary techniques have been developed to connect the BCI to
a traditional haptic interface or to utilize those interfaces as stimulation sources.
48 R.A. Ramadan et al.

Vibrotactile stimulation brings also a possibility to create bone-conduction sensory

effect in case of the head area exciters application. This point is still very pre-
liminary yet relative to the existing applications. It brings a very interesting pos-
sibility to deliver multimodal stimuli (somatosensory and auditory combined) to
TLS/ALS subjects with a very fast information transfer rate.
Currently, the eld of BCI requires deeper insights on how to capture the right
signals and then process them suitably. Efforts are being made to recognize the objects
as they are seen by the brain. These efforts will bring in newer dimensions in the
understanding of brain functioning, damage and repair. It is possible to recognize the
thoughts of the human brain by capturing the right signals from the brain in future [8].
The P300 is an event related potential, a measurable electrical charge that is
directly related with impulse. Therefore, by capturing the P300, a BCI can directly
translate a persons intent into electrical commands that control articial devices.
A P300 speller is based on this principle, where the detection of P300 waves allows
the user to write characters [26]. The trend now is solving the P300 speller clas-
sication problems such as, the detection of the presence of a P300 in the elec-
troencephalogram (EEG) and the combination of different P300 responses for
determining the right character to spell.
Over the past 20 years, Brain-Computer Interfaces (BCI) have been shown to be
very promising for numerous applications, such as rehabilitation or entertainment,
among many others. Despite this potential, most BCI applications remain proto-
types that are not used in practice, outside laboratories. The main reason is the
widely acknowledged low reliability of current BCI systems that are based on the
translation of the spontaneous non-invasive electroencephalogram (EEG); mental
tasks performed by the user are being too often incorrectly recognized by the BCI.
Poor recognition performances are due in part to imperfect signal processing
algorithms used to analyze EEG signals. However, another component in the BCI
loop may also be decient such as the signal generator, i.e., the user him/herself
who may not be able to reliably produce EEG patterns. Indeed, it is widely
acknowledged that BCI use is a skill, which means the user must be properly
trained to achieve successful BCI control. So the main trend now is improving
reliability of BCI by teaching and training the users to the BCI skills.

2.11 Conclusion

The BCI reads the waves produced from the brain at different locations in the
human head, translates these signals into actions, and commands that can control
the computer(s). Brain computer interface can be classied into three main groups
which depend on the way that the electrical signal is obtained from neuron cells in
the human brain. The brain generates an amount of neural activity. There are a
plethora of signals, which can be used for BCI. These signals divide into two
classes: spikes and eld potentials, Components of particular interest to BCI can be
2 Basics of Brain Computer Interface 49

divided into four categories which are oscillatory EEG activity, event-related
potentials (ERP), slow cortical potentials (SCP), and neuronal potentials. Several
techniques have been used to monitor brain activities; each technique has its own
characteristics as well as pros and cons. BCI is interesting area to researchers
because it can solve many problems which seem to be impossible, Many appli-
cations focused in entertainment and playing games especially after using 3D
monitors, certain glasses and an EEG headset where the control on the game by
thoughts. Researchers in this led are looking in the future for more development in
trends and applications. Since the current trends are focused on utilizing of motor
system.

References

1. Melody, M.: Real-world applications for brain-computer interface technology. IEEE Trans.
Neural Syst. Rehabil. Eng. 11(2), 162165 (2003)
2. Pei, X., Barbour, D., Leuthardt, E., Schalk, G.: Decoding vowels and consonants in spoken
and imagined words using electrocorticographic signals in humans. J. Neural. Eng. 8(4), 111
(2011)
3. Azar, A.T., Balas, V.E., Olariu, T.: Classication of EEG-based brain-computer interfaces.
advanced intelligent computational technologies and decision support systems. Stud. Comput.
Intell. 486, 97106 (2014)
4. Buckner, R., Andrews-Hanna, J., Schacter, D.: The brains default network: anatomy,
function, and relevance to disease. Ann. N. Y. Acad. Sci. 1124, 138 (2008)
5. Minshew, N., Keller, T.: The nature of brain dysfunction in autism: functional brain imaging
studies. Curr. Opin. Neurol. 23, 124130 (2010)
6. Gilbert, S., Dumontheil, I., Simons, J., Frith, C., Burgess, P.: Wandering minds: the default
network and stimulus-independent thought. Sci. Mag. 315(5810), 393395 (2007)
7. Looney, D., Kidmose, P., Mandic, P.: Ear-EEG: user-centered and wearable BCI.
Brain-Comput. Interface Res. Biosyst. Biorobot. 6, 4150 (2014)
8. Kameswara, T., Rajyalakshmi, M., Prasad, T.: An exploration on brain computer interface and
its recent trends. Int. J. Adv. Res. Artif. Intell. (IJARI) 1(8), 1722 (2013)
9. Wolpaw, J., Birbaumer, N., McFarland, D., Pfurtscheller, G., Vaughan, T.: Braincomputer
interfaces for communication and control. Clin. Neurophysiol. 113, 767791 (2002)
10. He, B.: Neural engineering. Springer publisher, New York, ISBN:978-1-4614-5226-3 (2013)
11. Wolpaw, J.: Brain-Computer interfaces: signals, methods, and goals. Proceeding of 1st
international IEEE EMBS conference on neural engineering, vol. 1, pp. 584585 (2003)
12. Lebedev, M., Nicolelis, M.: Brainmachine interfaces: past, present and future. Trends
Neurosci. 29(9), 536546 (2006)
13. Pfurtscheller G. Neuper. C.: Motor imagery and direct brain computer communication. Proc.
IEEE 89(7), 1123,113 (2001)
14. Schalk, G., Mellinger, J.: A practical guide to brain-computer interfacing with BCI. Springer,
ISBN:978-1849960915 (2010)
15. Yang, Z., Wang, Y., Ouyang, G.: Adaptive neuro-fuzzy inference system for classication of
background EEG signals from ESES patients and controls. Sci. World J. 2014, 140863 (2014)
16. Middendorf, M., Mcmillan, G., Calhoun, G., Jones, K.S.: Brain computer interfaces based on
steady state visual evoked response. IEEE Trans. Rehabil. Eng. 8(2), 211214 (2000)
17. Tamara, B., Howard, J.: Privacy by design in brain-computer interfaces. University of
Washington, UWEE, technical report number UWEETR-2013-0001 (2013)
50 R.A. Ramadan et al.

18. Ebner, A., Sciarretta, G., Epstein, C., Nuwer, M.: EEG instrumentation. The international
federation of clinical neurophysiology. Electroencephalogr. Clin. Neurophysiol. Suppl. 52,
710 (1999)
19. Xian-Jie, P., Tie-Jun, L., De-Zhong, Y.: Design of an EEG preamplier for brain-computer
interface. J. Electr. Sci Technol China 7(1), 5660 (2009)
20. Robert, O., Pascal, F., Eric, M., Jan-Mathijs, S.: FieldTrip: open source software for advanced
analysis of MEG, EEG, and invasive electrophysiological data. J. Comput. Intell. Neurosci.
2011, 156869 (2011)
21. Lingaratnam, S., Murray, D., Carle, A., Kirsa, S., Paterson, R., Rischin, D.: Developing a
performance data suite to facilitate lean improvement in a chemotherapy day unit. J. Oncol.
Pract. 9(4), 115121 (2013)
22. Arnaud, D., Christian K., Andrey V., Nima B., Robert O., Thorsten Z., Scott M.: MATLAB-
based tools for BCI research. In: Desney, S.T., Anton, N. (eds.) (B + H)CI: the human in
brain-computer interfaces and the brain in human-computer interaction book. ISBN 978-1-
84996-271-1, Springer, Chap. 14, pp. 241259 (2010)
23. Arindam, D., Robins, K., Suruchi, M., Sanjay, C., Alakananda, B., Anirban, D.: A low-cost
point-of-care testing system for psychomotor symptoms of depression affecting standing
balance: a preliminary study in india. Depression Res. Treat. 2013, 640861 (2013)
24. Setijadi, A., Novanda, O., Mengko, T.: Development of an experimental portable
electroencephalograph (case study: alpha wave detector). International conference on
electrical engineering and informatics (ICEEI), vol. 1, no. 6, pp. 1719 (2011)
25. Chatelle, C., Chennu, S., Noirhomme, Q., Cruse, D., Owen, A.M., Laureys, S.: Brain-
computer interfacing in disorders of consciousness. Brain Inj. 26(12), 15101522 (2012)
26. Anupama, H., Cauvery, N., Lingaraju, G.: Brain computer interface and its types-a study. Int.
J. Adv. Eng. Technol (IJAET) 3(2), 739745 (2012)
27. Jan, B., Van, E., Fabien, L., Michael, T.: Brain computer interfaces: beyond medical
applications. IEEE Comput. Soc. 45(4), 2634 (2012)
28. Rabie, A., Ahmed, E., Mohamed, A.: JustThink: smart BCI applications. Proceedings of
seventh international conference on bio-inspired computing: theories and applications.
advances in intelligent systems and computing, vol. 201, pp. 461472 (2012)
29. Alwasiti, H., Aris, I., Jantan, A.A.: Brain computer interface design and applications:
challenges and future. World Appl. Sci. J. 11(7), 819825 (2010)
30. Wang, U., Wang, Y., Jung, T.: A cell-phone based brain-computer interface for
communication in daily life. Proceedings of the 2010 international conference on articial
intelligence and computational intelligence (AICI10), pp. 233240 (2010)
31. Reza Fazel-Rezai (ed.): Brain-Computer Interface Systems - Recent Progress and Future
Prospects, ISBN: 978-953-51-1134-4, InTech. Chap. 11, pp. 3739, (2013)
http://www.springer.com/978-3-319-10977-0