J. Biomedical Science and Engineering, 2011, 4, 788-796
doi:10.4236/jbise.2011.412097 Published Online December 2011 (http://www.SciRP.org/journal/jbise/ JBiSE ).
Published Online December 2011 in SciRes. http://www.scirp.org/journal/JBiSE
A review of developments of EEG-based automatic medical
support systems for epilepsy diagnosis and seizure detection
Yuedong Song
Computer Laboratory, University of Cambridge, Cambridge, United Kingdom.
Received 21 October 2011; revised 15 November 2011; accepted 5 December 2011.
ABSTRACT
Epilepsy is one of the most common neurological
disorders-approximately one in every 100 people
worldwide are suffering from it. The electroencepha-
logram (EEG) is the most common source of infor-
mation used to monitor, diagnose and manage neu-
rological disorders related to epilepsy. Large amounts
of data are produced by EEG monitoring devices,
and analysis by visual inspection of long recordings
of EEG in order to find traces of epilepsy is not rou-
tinely possible. Therefore, automated detection of
epilepsy has been a goal of many researchers for a
long time. Until now, reviews of epileptic seizure de-
tection have been published but none of them has
specifically reviewed developments of automatic
medical support systems utilized for EEG-based epi-
leptic seizure detection. This review aims at filling
this lack. The main objective of this review will be to
briefly discuss different methods used in this research
field and describe their critical properties.
Keywords: Electroencephalo gram; Epileptic Seizure; Au-
tomatic Diagnostic Systems; Feature Analysis; Recogni-
tion
1. INTRODUCTION
Epilepsy is a neurological disorder affecting around 1%
of the world’s population (about 50 million people) [1].
An epileptic seizure can be characterized by means of
paroxysmal occurrence of synchronous oscillations. This
kind of seizures can mainly be divided into two classes
in terms of the extent of connection of different brain
fields: partial seizures and generalized seizures. Partial
seizures begin from a circumscribed field of the brain,
usually called epileptic foci. Determined by their type,
they may or may not impair consciousness. Generalized
seizures involve most fields of the brain and may cause
loss of consciousness and muscle contractions or stiff-
ness. Electroencephalography (EEG) is an important
clinical tool, monitoring, diagnosing and managing neu-
rological disorders related to epilepsy. In comparison
with other approaches such as Magnetoencephalography
(MEG) and functional Magnetic Resonance Imaging
(fMRI), EEG is a clean, cost effective and safe technique
for monitoring brain activity.
In spite of available dietary, drug and surgical treat-
ment options, currently nearly one out of three epilepsy
patients cannot be treated. They are completely subject
to the sudden and unforeseen seizures which have a
great effect on their daily life, with temporary impair-
ments of perception, speech, motor control, memory
and/or consciousness. Many new therapies are being
investigated and among them the most promising are
implantable devices that deliver direct electrical stimula-
tion to affected areas of the brain. These treatments will
greatly depend on robust algorithms for seizure detection
to perform effectively. Because the onset of the seizures
cannot be predicted in a short period, a continuous re-
cording of the EEG is required to detect epilepsy. How-
ever, analysis by visual inspection of long recordings of
EEG, in order to find traces of epilepsy, is tedious,
time-consuming and high-cost. Therefore, automated
detection of epilepsy has been a goal of many research-
ers for a long time. Computers have long been suggested
for handling this problem and thus, automatic medical
support systems for identifying electroencephalographic
changes have been under study for many years. The
whole procedure can be divided into two modules: fea-
ture extraction and classification (shown in Figure 1).
The performance of automatic diagnosis systems de-
pends on both the feature extraction methods and the
classification algorithms applied. Until now, although
many methodologies have been developed for automatic
epileptic seizure detection, there is no literature specifi-
cally contributing to the review of development of
automatic medical support systems utilized for EEG-
based epileptic seizure detection. In this review, we
briefly investigate different approaches used in this re-
search field and describe their critical properties.

Y. D. Song / J. Biomedical Science and Engineering 4 (2011) 788-796 789
EEG d ata acq uisition
Feature extraction (SampEn)
Classification models
Diagnosis decision by the neurologist
Figure 1. Schematics of the proposed di-
agnostic expert system: the whole system
can be mainly divided into two modules,
namely developing feature extraction meth-
ods and developing classification models .
The review is organized as follows: Section 2 de-
scribes the EEG database launched b y [2] which is wide-
ly used in epileptic seizure detection . Section 3 discusses
the characteristics of differen t feature extraction and clas-
sification methods for automatic epileptic seizure diag-
nosis and detection. Section 4 presents some results of
studies on automatic epileptic seizure diagnosis and de-
tection using EEG databases except for EEG database
described in [2]. Section 5 discusses predictability of
epileptic seizure from human EEGs. Section 6 concludes
the paper.
2. EEG DATABASE
In The most popular and widely used database for the
study of EEG-based epileptic seizure detection was
launched by University of Bonn [2] which is described
as follows:
The whole EEG data is composed of five sets (de-
noted A-E), each containing 100 single-channel EEG
data of 23.6 s duration. Sets A and B were taken from
surface EEG recordings of five healthy volunteers with
eyes open and closed, respectively. Sets C, D and E
originated from the EEG archive of presurgical diagno-
sis. Signals in Set C were recorded from the hippocam-
pal formation of the opposite hemisphere of the brain,
and signals in Set D were recorded from within the epi-
leptogenic zone. While Sets C and D contain only brain
activity measured during seizure free intervals, Set E
contains only seizure activity. All EEG signals were
recorded with the same 128-channel amplifier. The data
were digitized at 173.6 samples per second at 12-bit
resolution. Band pass filter was set to 0.53 - 40 Hz. Fig-
ure 2 describes the electrode placement for recording of
EEG signals. Figure 3 describes examples of EEG sig-
nals of Set A, Set D and Set E, where the difference can
be seen in terms of the value of amplitudes and waveform.
A summary of the EEG data set is shown in Table 1.
3. DEVELOPMENT OF
METHODOLOGIES FOR
AUTOMATIC EPILEPTIC SEIZURE
DIAGNOSIS AND DETECTION
Development of EEG signal processing techniques is
closely related to its characteristics. EEG is a random
and unstable signal. Abnormal EEG recordings can be
divided into EEGs with non-paroxysmal abnormality
and EEGs with paroxysmal abnormality according to
their appearance form [3]. EEGs with paroxysmal ab-
normality are composed of spike wave, spike-and-slow-
wave and sharp wave. Spike wave is the basic form of
EEGs with paroxysmal abnormality and its time length
is 20 ms ~ 70 ms. Most spike wave appears with nega-
tive phase but sometimes it appears with positive phase,
diphasic waveform and triphasic waveform [3]. Spike-
and-slow-wave which has duration of 200 ms ~ 500 ms
appears after spike wave. Sharp wave is similar with
spike wave but its duration (generally 70 ms ~ 200 ms)
is longer than that of spike wave. The extraction of epi-
leptic characteristic wave is of great importance in epi-
leptic diagnosis, localization and epileptic seizure detec-
tion. In order to choose the most suitable methods for an
automatic epileptic seizure detection system, it is neces-
sary to understand what features are employed and their
corresponding properties. Next, several feature extrac-
tion methods widely used in epileptic seizure detection
are described.
Figure 2. Scheme of the locations of surface electrodes in
terms of the international 10 - 20 systems for recording EEG
patterns. Names of the electrode are derived from their ana-
omical locations [2]. t
C
opyright © 2011 SciRes. JBiSE

Y. D. Song / J. Biomedical Science and Engineering 4 (2011) 788-796 790
(a)
(b)
(c)
Figure 3. Sample EEG recordings. (a) Normal EEG; (b) Interictal EEG; (c) Ictal EEG.
JBiSE

Y. D. Song / J. Biomedical Science and Engineering 4 (2011) 788-796 791
Table 1. Summary of the clinical data: The whole EEG data is composed of five sets (denoted A-E), each containing 100 sin-
gle-channel EEG data of 23.6 s duration. Sets A and B were taken from surface EEG recordings of five healthy volunteers with eyes
open and closed, respectively. Sets C D, and E originated from the EEG archive of presurgical diagnosis. Signals in Set C were re-
corded from the hippocampal formation of the opposite hemisphere of the brain, and signals in Set D were recorded from within the
epileptogenic zone. While Sets C and D contain only brain activity measured during seizure free intervals, Set E contains only sei-
zure activity.
Data Set A Data Set B Data Set C Data Set D Data Set E
Subjects Five healthy subjects Five healthy subjectsFive epileptic patientsFive epileptic patients Five epileptic patients
Electrode type Surface Surface Intracranial Intracranial Intracranial
Electrode-placement International 10 - 20
system International 10 - 20 system Opposite to epileptogenic zone Within epileptogenic zone Within epileptogenic zone
Patient’s state Awake and eyes open
(Normal) Awake and eyes closed (Normal) Seizure-free (Interictal) Seizure-free (Interictal) Seizure activity (Ictal)
Number of epochs 100 100 100 100 100
Epoch duration (s) 23.6 23.6 23.6 23.6 23.6
3.1. Frequency Domain Analysis and
Time-Frequency Domain Analysis
Frequency domain analysis is based on Fourier trans-
form which decomposes EEG signals into different fre-
quency domains. Epileptic EEG recordings can be de-
tected in terms of the difference between epileptic EEG
data and normal EEG d ata in frequency domain [4,5]. In
most cases slow wave appears in epileptic patients’ EEG
recordings, hence epileptic abnormality which cannot be
detected in time domain is revealed by means of fre-
quency analysis. However the weakness of frequency
analysis is that by means of Fourier analysis, the ob-
tained signals is its total spectrum and it cannot be used
for local analysis. Furthermore, since methods based on
Fourier transform cannot provide important EEG dy-
namic information in time domain and frequency do-
main simultaneously, it is not suitab le for analyzing time
series signals like EEG signals which have characteris-
tics of instability and randomness. In recent years,
time-frequency domain analysis has been increasingly
used for feature extraction of epileptic EEG. The most
widely-used approach is Wavelet Transform [6-8].
Wavelet transform can be utilized for analyzing signals
in different sub-bands in a selective way, which is suit-
able for extracting epileptic characteristics and increases
detection performance of the system. Contrary to Fourier
transform, wavelet transform supplies a more flexible
approach of time-frequency representation of a signal by
means of using analysis windows with varied size. The
important characteristic of wavelet transform is that it
supplies precise time information at high frequencies
and precise frequency information at low frequencies.
This characteristic is of great importance, since signals
in biomedical applications usually include low frequency
information with long time duration and high frequency
information with short time duration. By means of wav-
elet transform, transient characteristics are accurately
captured and it is localized in both time and frequency
domain. In [9], wavelet transform was employed for
detecting and characterizing epileptiform discharges in
the form of 3-Hz spike and wave complex in patients
with absence seizure. [10] extracted features in time-
domain as well as frequency-domain of the EEG re-
cordings and fed them into a recurrent neural network.
[11] developed a system on the basis of deciding the
seizure probability of a set of EEG recordings; wavelet
decomposition and data segmentation were integrated
for calculating a priori probabilities required for the
Bayesian formulation applied in training and testing op-
eration. On the whole, for the feature analysis using
wavelet transform-based methods, the main problem lies
in the choice of mother wavelet. The general choice is
Daubechies wavelets which have similar waveform with
spike wave [12].
3.2. Complex Analysis
After EEG signals are analyzed in time-frequency do-
main, nonlinear measures such as largest Lyapunov ex-
ponent [13,14] and entro py [15-17] are utilized for quan-
tifying the degree of complexity within a time series.
When utilized with EEG, those measures help compre-
hending EEG dynamics and underlying chaos in the
brain. Lyapunov exponents are a quantitative measure
for differentiating among different kinds of orbits on the
basis of their sensitive dependence on the initial condi-
tions, and are employed for deciding the stability of any
steady-state behaviour. Entropy is a concept handling
predictability and randomness, with higher values of
entropy always related to less system order and more
randomness. In [15], different entropy-based features
C
opyright © 2011 SciRes. JBiSE

Y. D. Song / J. Biomedical Science and Engineering 4 (2011) 788-796 792
that are utilized to normal and epileptic electroencepha-
logram recordings were compared and then were tested
by applying the adaptive neuro-fuzzy inference systems.
The above-mentioned feature extraction methods used
for EEG signal analysis include an assumption that the
underlying signal dynamic mechanism is composed of a
linear superposition of complex exponentials. But the
intrinsic basis functions are usually presumed a priori
rather than extracted from the EEG recordings in an
adaptive way. The obtained power spectrum derived
from the analysis involves spu rious power readings if the
EEG time series signals we are interested in include
more than pure low frequency functions, and contain
energy that always stands for nonlinearities in the ana-
lyzed EEG recordings. Th e reason is that nonlinearity in
the EEG data will be stand for within the power spec-
trum as higher-order harmonics because the transform
itself employs an accumulation of trigonometric func-
tions. As long as the signal transform is formed, it is
hard to discriminate true power-frequency EEG signals
from spurious energy representation because of nonlin-
earities. Hence every time-varying frequen cy representa-
tion will be averaged out within the power spectrum.
Therefore some novel signal decomposition approaches
are required to obtain underlying oscillators originated
from a seizure signal without any assumptions of the
underlying waveform or specific time-scales of the os-
cillatiors, which is capable of presenting the dynamic of
EEG signals in an adaptive way.
3.3. Classification Models
After features in EEG signals are extracted utilizing the
above-mentioned signal processing methods, different
techniques based on pattern recognition are then devel-
oped for classifying these obtained feature vectors. In
order to select the most suitable classifier for a set of
features, the properties of the available classifiers have
to be understood. In recent years, several classification
models have been developed for handling EEG signals
classification for epileptic seizure detection, and among
these methods, Neural Network-based methods and
Support Vector Machine-based methods (SVM) are two
widely-used classification paradigms. Artificial Neural
Networks (ANN) has been widely used in pattern recog-
nition, signal prediction and feature extraction due to its
excellent self-learning capability, self-adaptive capabil-
ity and strong parallel processing mechanism. A variety
of algorithms on the basis of ANN have been employed
in EEG signal classification and epileptic seizure detec-
tion [18-22]. The learning mechanism of neural net-
works can be mainly divided into two categories, namely
supervised learning and unsupervised learning. Super-
vised learning needs prior knowledge of the analysed
data and the back-propagation methods are implemented
for the training of weights in neural networks. The un-
supervised learning paradigm, on the contrary, has fewer
requirements for the prior knowledge of data, and pat-
terns with similar characteristics are clustered together
by systems. Initial EEG data points and some extracted
features using other methods such as waveform charac-
teristic parameters detected by utilizing time domain
analysis, results of wavelet decomposition, etc., can be-
come inputs of neural networks. However the use of
neural network refers to many parameters and options
such as training parameters, network structures and ini-
tial weights and so on, which may have great impact on
the training procedure of neural networks. A large num-
ber of experiments are thus required to choose optimal
parameter sets and a large amount of data is also needed
for testing performance of neural networks. The conflict
between performance and computation complexity in
artificial neural networks is usually figured out by means
of trial and the problem regarding how to select optimal
number of hidden nodes in neural networks still remains
unsolved. In [23], a method based on iteration was de-
veloped to handle EEG signals piecewise, which reduces
the computation time and cost of neural networks.
The Support Vector Machine (SVM) is a supervised
machine learning paradigm capable of solving linear and
non-linear classification and regression problems [23].
SVM paradigm was first proposed in [24] based on the
ideas of statistical learning theory and structural risk
minimization. Due to its accuracy and capabilit y of han-
dling a great number of predictors, it has been widely
used in EEG signal classification and epileptic seizure
detection [25-29]. Most of classification models divide
categories utilizing hyperplanes which separate the
categories by means of a flat plane in the predictor space.
Support vector machines expand the concept of hyper-
plane separation to data which cannot be divided linearly,
through mapping the predictors into a higher-dimen-
sional space where data can be divided linearly. SVM
classification models have many advantages. A special
global optimum for its parameters, such as the degree d
of the kernel function and misclassification trade-off
factor c controling the trade-off between the maximum
margin and the minimum training error, can be found by
means of quadratic programming optimization. Nonlin-
ear boundaries are able to be utilized without much extra
computational effort. Furthermore the performance of
SVM is very competitive with other classification mod-
els. A weakness SVM has is that the problem complexity
is related to the order of the number of patterns rather
than the order of the dimension of the patterns. The gen-
eral quadratic programming algorithm will usually fail
and unique-purpose optimizers employing problem-
C
opyright © 2011 SciRes. JBiSE

Y. D. Song / J. Biomedical Science and Engineering 4 (2011) 788-796 793
specific speedups need to be utilized for resolving the
optim i z ation problems.
The above-mentioned methods for automatic epileptic
seizure detection have their own characteristic; the per-
formance of detecting epileptic seizure using these de-
veloped systems will be increased if we can integrate
these methods for enhancing their self- adaptive capabil-
ity. In order to obtain power spectra in patients with sei-
zures, multiple signal classification methods were de-
veloped in [30]. Methodologies on the basis of the com-
bination of statistical time series analysis, k-nearest
neighbour clustering and chaos theory were proposed in
[31]. Although many methods for EEG-based epileptic
seizure detection have been developed recently and have
shown good experimental results, there are still some
problems which need to be solved when applied in
clinical settings. In the study of EEG-based epileptic
seizure detection, due to the lack of publically available
EEG databases and the limitation of clinical data sam-
ples, most proposed methods were developed using only
EEG databases with small number of data samples and it
is very likely that they are not applicable in real situa-
tions, which makes it difficult to conduct an in-depth
investigation of adaptive methodologies for clinical ap-
plication. In addition, the EEG data compression is also
a problem in this research field. In clinical epileptic sei-
zure detection from human Electroencephalograms, the
systems used usually have 8, 16, 32 or more electrode
channels and the duration of EEG recordings are very
long. Huge number of data processing tasks will have
direct impact on the applicability of the developed algo-
rithms, making it difficult to detect epileptic seizures in a
real-time situation efficiently.
4. STUDIES ON EPILEPTIC SEIZURE
DIAGNOSIS AND DETECTION USING
OTHER EEG
Most studies about developing epileptic seizure diagno-
sis and detection systems that were mentioned above are
mainly based on the EEG database described in [2]. In
addition to this EEG database, some studies are also
conducted using other EEG resources. [32] developed a
fuzzy rule-based seizure detection system on the basis of
knowledge from experts’ reasoning. A total of 302.7
hours of intracranial EEG data recordings obtained from
21 patients with 78 seizures was employed for assessing
the system. Spectral, temporal and complexity features
were extracted from IEEG recordings and joined by
utilizing the fuzzy rule-based system in a spatio-tempo-
ral way for detecting epileptic seizures. The system
showed an excellent performance with a sensitivity of
98.7%, an average detection latency of 11 seconds and a
false detection rate of 0.27/h. [33] defined a generalized
nonlinear method for identifying seizure EEG segments
from non-seizure segments using nonlinear decision
functions with the flexibility in selecting any degree of
complexity and with any number of dimensions. A per-
formance assessment of the correlation sum according to
sensitivity, specificity and accuracy in its capability of
discriminating seizure signals from non-seizure signals
was supplied. A total of 126 EEG signals from 11 se-
quential patients were handled and the correlation sum
was calculated from non-overlapping scrolling windows
with 1 second duration. The experimental observations
showed a significant decrease in the amplitude of the
correlation sum prior to the onset of seizures. The ap-
proach with k-fold cross validation conducted with a
sensitivity of 92.31%, a specificity of 91.67% and an
accuracy of 91.84%, which shows its suitability for off-
line seizure detection. [34] tried to identify the seizure
onset patterns by using an evolutionary scheme which
searches for optimal kernel types and parameters for
support vector machine. They considered the fractal di-
mension, Lyapunov exponent and wavelet entropy for
feature extraction and the classification accuracy of this
method was evaluated using the CHB-MIT dataset. A
comparison of experimental results revealed that the
proposed approach outperformed that of general support
vector machine, and the accuracy rate achieved 96.29%
for sensitivity and 100% for specificity. In [35], a novel
algorithm based on wavelet analysis was proposed for
detecting epileptic seizures from scalp EEG signals.
They used wavelet packet transform to decompose the
EEG data from each channel. In terms of the obtained
wavelet coefficients, a patient-specific measure was de-
veloped for quantifying the separation between non-
seizure and seizure sign als within the frequency ra nge of
1 - 30 Hz. The measure was utilized for determining a
normalized index called combined seizure index which
is obtained for each EEG channel. Significant increase
during seizure on set is observed using combined seizure
index and channel alarms were then generated by one-
sided cumulative sum test on the basis of this normalized
index. The approach was evaluated on EEG recordings
originated from fourteen patients with sixty-three sei-
zures during 75.8 hours. The results showed a low false
detection rate of 0.51/h, a high sensitivity of 90.5% and
a median detection delay of seven seconds.
5. PREDICTABILITY OF EPILEPTIC
SEIZURES FROM HUMAN EEGS
The human brain is considered as a dynamic system,
because epileptic networks in human beings are compli-
cated nonlinear architectures and the interactions are
supposed to reveal nonlinear behaviour. These ap-
proaches support the point that quantification of changes
C
opyright © 2011 SciRes. JBiSE

Y. D. Song / J. Biomedical Science and Engineering 4 (2011) 788-796 794
in the human brain originating from EEG may predict
epileptic seizures, but conventional approaches are not
able to identify particular change before seizure happens.
[36] utilized nonlinear dynamics methods into clinical
epilepsy analysis and their point is that seizure can be
thought of as a change of the brain with epilepsy from
chaotic to a more regular circumstances. Hence the spa-
tial-temporal characteristics of the brain with epilepsy
are not the same for various clinical circumstances. They
conduct more investigations on the basis of temporary
evolution of a nonlinear dynamic analysis method called
the largest Lyapunov exponent for patients having tem-
porary lobe epilepsy [37] and concluded that the EEG
action is growingly less chaotic when the seizure moves
towards. Because of these pioneering researches, non-
linear approaches originated from dynamical system the-
ory have been used for quantifying the transitions of
human brain dynamics prior to the beginning of seizures.
[38] performed investigation on the increase of non linear
complexity from human neuronal networks before sei-
zure happens on the basis of the information from
changes in the neuron al complexity lo ss, wh ich ou tlin ing
the complicated content of the correlation dimension.
[39] noticed that the alterations in the correlation integral
can be utilized for pursuing precisely the beginning of
seizure for a patient with temporal lobe epilepsy, where-
as [40] showed that by means of changes of the fre-
quency and amplitude, those alterations in the correla-
tion integral can be fully explained. In [41], a sudden
decrease in the dynamical similarity during the period
before seizur e happens w as observed and that action was
getting more and more noticeable when the beg inning of
seizure moved forwards. [42] found that the energy in
human EEG signals raises before seizure happens, and in
their following studies the pro of of epileptic seizure pre-
dictability on the basis of the choice of diverse of nonlin-
ear and linear characteristics of the EEG was supplied
[43,44] made use of 4 different nonlinear quantification
approaches under the framework of the Lyapunov theory
and observed important preictal changes. Most of the
above-mentioned researches in epilepsy prediction are
conducted on the basis of intracranial EEG recordings.
However two problems need to be considered and solved
when it comes to the study of scalp EEG recordings. 1)
scalp EEG data are more subject to eye and muscle arte-
facts as well as environmental noise than the intracranial
EEG data; 2) the significant information in EEG signals
are weakened and mixed in the propagation by means of
soft bone and tissue. Conventional nonlinear analysis
approaches like sample entropy or the Lyapunov expo-
nents are influenced by the above-mentioned two prob-
lems and hence they cannot be used to discriminate be-
tween slightly different chaotic rules in the scalp EEG
[45]. One method for handling those problems is to de-
fine various nonlinear measures generating better results
in comparison with the conventional nonlinear analysis
methods for the scalp EEG recordings. [46] followed
this method for analyzing scalp EEGs and developed an
approach on the basis of the phase-space dissimilarity
measures to predict epileptic events from human scalp
EEG recordings. The method developed on the basis of
dynamical entrainment has revealed good results as well
on human scalp EEG recordings for epileptic seizure
predictability [47,48].
6. CONCLUSIONS
Diagnosing epilepsy needs acquisition of patients’ EEG
recording and collecting additional clinical information.
Large amounts of data are produced by EEG monitoring
devices and analysis by visual inspection of long re-
cordings of EEG in order to find traces of epilepsy is not
routinely possible. Research into automatic detection
systems for epilepsy has been increasingly popular dur-
ing these years. The problem of signal classification for
epileptic seizure detection is considered as a typical pat-
tern-recognition problem which includes feature extrac-
tion and classification. In this paper, we briefly reviewed
different methods developed for automatic epileptic sei-
zure detection and describe their critical properties.
Various feature extraction techniques on the basis of
frequency domain analysis, time-frequency domain ana-
lysis and complex analysis were discussed; respectively
and classification models employed for designing medi-
cal support systems of auto matic epileptic seizur e detec-
tion were also discussed. On the other hand, although
predictability of epileptic seizure originating from hu-
man intracranial and scalp EEGs has been approved,
more studies need to be conducted for increasing the
accuracy of prediction.
REFERENCES
[1] Iasemidis, L.D., Shiau, D.S., Chaovalitwongse, W., Sac-
kellares, J.C., Parda los, P.M., Principe, J.C., Ca rney, P. R.,
Prasad, A., Veeramani, B. and Tsakalis, K. (2003) Adap-
tive epileptic seizure prediction system. IEEE Transac-
tion on Biomedical Engineering, 50, 616-627.
doi:10.1109/TBME.2003.810689
[2] Andrzejak, R.G., Lehnertz, K., Mormann, F., Rieke, C.,
David, P. and Elger, C.E. (2001) Indications of nonlinear
deterministic and finite-dimensional structures in time se-
ries of brain elect rical activity: Dependence on recording
region and brain state. Physical Reviews E, 64, 061907.
doi:10.1103/PhysRevE.64.061907
[3] IFSECN (1974) A glossary of terms most commonly used
by clinical electroencephalographers. Electroencephalogra-
phy and Clinical Neurophysiology, 37, 538-548.
doi:10.1016/0013-4694(74)90099-6
[4] Gotman, J., Flanagan, D., Zhang, J. and Rosenblatt, B.
C
opyright © 2011 SciRes. JBiSE

Y. D. Song / J. Biomedical Science and Engineering 4 (2011) 788-796 795
(1997) Automatic seizure detection in the newborn: Me-
thods and initial evaluation. Electroencephalography and
Clinical Neurophysiology, 103, 356-36 2.
doi:10.1016/S0013-4694(97)00003-9
[5] Polat, K. and Günes, S. (2007) Classification of epilepti-
form EEG using a hybrid system based on decision tree
classifier and fast fourier transform. Applied Mathemat-
ics and Computation, 187, 1017-1026.
doi:10.1016/j.amc.2006.09.022
[6] Übeyli, E.D. (2009) Combined neural network model
employing wavelet coefficients for EEG signals classifi-
cation. Digital Signal Processing, 19, 297-308.
[7] Subasi, A. (2007) EEG signal classification using wave-
let feature extraction and a mixture of expert model. Ex-
pert Systems with Applications, 32, 1084-1093.
doi:10.1016/j.eswa.2006.02.005
[8] Zandi, A.S., Javidan, M., Dumont, G.A. and Tafreshi, R.
(2010) Automated real-time epileptic seizure detection in
scalp eeg recordings using an algorithm based on wavelet
packet transform. IEEE Transactions on Biomedical En-
gineering, 57, 1639-1651.
doi:10.1109/TBME.2010.2046417
[9] Adeli, H., Zhou, Z. and Dadmehr, N. (2003) Analysis of
EEG records in an epileptic patient using wavelet trans-
form. Journal of Neuroscience Methods, 123, 69-87.
doi:10.1016/S0165-0270(02)00340-0
[10] Srinivasan, V., Eswaran, C. and Sriraam, N. (2005) Arti-
ficial neural network based epileptic detection using time-
domain and frequency-domain features. Journal of Medi-
cal Systems, 29, 647-660.
doi:10.1007/s10916-005-6133-1
[11] Saab, M.E. and Gotman, J. (2005) A system to detect the
onset of epileptic seizures in scalp EEG. Clinical Neuro-
physiology, 116, 427-442.
doi:10.1016/j.clinph.2004.08.004
[12] Indiradevi, K.P., Elias, E., Sathidevi, P.S., Dinesh, S. and
Radhakrishnan, K. (2008) A multi-level wavelet approach
for automatic detection of epileptic spikes in the electro-
encephalogram. Computers in Biology and Medicine, 38,
805-816. doi:10.1016/j.compbiomed.2008.04.010
[13] Übeyli, E.D. (2009) Automatic detection of electroence-
phalographic changes using adaptive neuro-fuzzy inference
system employing Lyapunov exponents. Expert Systems
with Applic atio ns , 36, 9031-9038.
doi:10.1016/j.eswa.2008.12.019
[14] Ghosh-Dastidar, S., Adeli, H. and Dadmehr, N. (2007)
Mixed-band wavelet-chaos-neural network methodology
for epilepsy and epileptic seizure detection. IEEE Trans-
actions on Biomedical Engineering, 54, 1545-1551.
doi:10.1109/TBME.2007.891945
[15] Kannathal, N., Choo, M.L., Rajendra, A.U. and Sada-
sivan, P.K. (2005) Entropies for detection of epilepsy in
EEG. Computer Method and Programs in Biomedicine,
80, 187-194. doi:10.1016/j.cmpb.2005.06.012
[16] Ocak, H. (2008) Optimal classification of epileptic sei-
zures in EEG using wavelet analysis and genetic algo-
rithm. Signal Processing, 88, 1858-1867.
doi:10.1016/j.sigpro.2008.01.026
[17] Srinivasan, V., Eswaran, C. and Sriraam, N. (2007) Appro-
ximate entropy-based epileptic EEG detection using arti-
ficial neural networks. IEEE Transactions on Information
Technology in Biomedicine, 11, 288-295.
doi:10.1109/TITB.2006.884369
[18] Orhan, U., Hekim, M., Ozer, M. (2011) EEG signals clas-
sification using the K-means clustering and a multilayer
perceptron neural network model. Expert Systems with
Applications, 38, 13475-13481.
doi:10.1016/j.eswa.2011.04.149
[19] Kumar, S.P., Sriraam, N., Benakop, P.G. and Jinaga, B.C.
(2009) Entropies based detection of epileptic seizures
with artificial neural network classifiers. Expert Systems
with Applic atio ns , 37, 3284-3294.
doi:10.1016/j.eswa.2009.09.051
[20] Naghsh-Nilchi, A.R. and Aghashahi, M. (2010) Epilepsy
seizure detection using eigen-system spectral estimation
and multiple layer perceptron neural network. Biomedi-
cal Signal Processing and Control, 5, 147-157.
doi:10.1016/j.bspc.2010.01.004
[21] Kocyigit, Y., Alkan, A. and Erol, H. (2008) Classification
of EEG recordings by using fast independent component
analysis and artificial neural network. Journal of Medical
Systems, 32, 17-20. doi:10.1007/s10916-007-9102-z
[22] Guo, L., Rivero, D., Dorado, J., Rabunal, J.R. and Pazos,
A. (2010) Automatic epileptic seizure detection in EEGs
based on line length feature and artificial neural networks.
Journal of Neuroscience Methods, 191, 101-109.
doi:10.1016/j.jneumeth.2010.05.020
[23] Boser, B.E. (1992) A training algorithm for optimal mar-
gin classifiers. Proceedings of 5th Annual Workshop of
Computational Learning Theory, Pennsylvania, 144-152.
[24] Cortes, C. and Vapnik, V. (1995) Support vector n etworks.
Machine Learning, 20, 273-297.
doi:10.1007/BF00994018
[25] Guler, I. and Übeyli, E.D. (2007) Multiclass Support Vec-
tor Machines for EEG-Signals Classification. IEEE Tran-
sactions on Information Technology in Biomedicine, 11,
117-126. doi:10.1109/TITB.2006.879600
[26] Gardner, A.B., Krieger, A.M., Vachtsevanos, G. and Litt,
B. (2006) One-class novelty detection for seizure analy-
sis from intracranial EEG. Journal of Machine Learning
Research, 7, 1025-1044.
[27] Hsu, K.C. abd Yu, S.N. (2010) Detection of seizures in
EEG using subband nonlinear parameters and genetic
algorithm. Computers in Biology and Medicine, 40, 823-
830. doi:10.1016/j.compbiomed.2010.08.005
[28] Chandaka, S., Chatterjee, A. and Munshi, S. (20 09) Cross-
correlation aided support vector machine classifier for
classification of EEG signals. Expert Systems with Ap-
plications, 36, 1329-1336.
doi:10.1016/j.eswa.2007.11.017
[29] Übeyli, E.D. (2008) Analysis of EEG signals by com-
bining eigenvector methods and multiclass support vec-
tor machines. Computers in Biology and Medicine, 38,
14-22. doi:10.1016/j.compbiomed.2007.06.002
[30] Alkan, A., Koklukaya, E. and Subasi, A. (2005) Automa-
tic seizure detection in EEG using logistic regression and
artificial neural network. Journal of Neuroscience Meth-
ods, 148, 167-176. doi:10.1016/j.jneumeth.2005.04.009
[31] Chaovalitwongse, W.A., Fan, Y. and Sachdeo, E.C. (2007)
On the time series k-nearest neighbour classification of
abnormal brain activity. IEEE Transactions on Systems,
Man and Cybernetics-Part A: Systems and Humans, 37,
1005-1016. doi:10.1109/TSMCA.2007.897589
[32] Aarabi, A., Fazel-Rezai, R. and Aghakhani, Y. (2009) A
C
opyright © 2011 SciRes. JBiSE

Y. D. Song / J. Biomedical Science and Engineering 4 (2011) 788-796 796
JBiSE
fuzzy rule-based system for epileptic seizure detection in
intracranial EEG. Clinical Neurophysiology, 120, 1648-
1657. doi:10.1016/j.clinph.2009.07.002
[33] Tito, M., Cabrerizo, M., Ayala, M., Barreto, A., Miller, I,
Jayakar, P. and Adjouadi, M. (2009) Classification of elec-
troencephalographic seizure recordings into ictal and in-
terictal files using correlation sum. Computers in Biology
and Medicine, 39, 604-614.
doi:10.1016/j.compbiomed.2009.04.005
[34] Zavar, M., Rahati, S., Akbarzadeh-T, M.-R. and Ghase-
mifard, H. (2011) Evolutionary model selection in a wave-
let-based support vector machine for automated seizure
detection. Expert Systems with Applications, 38, 10751-
10758. doi:10.1016/j.eswa.2011.01.087
[35] Zandi, A.S., Javidan, M., Dumont, G.A. and Tafreshi, R.
(2010) Automated real-time epileptic seizure detection in
scalp EEG recordings using an algorithm based on wave-
let packet transform. IEEE Transactions on Biomedical
Engineering, 57, 1639-1651.
doi:10.1109/TBME.2010.2046417
[36] Iasemidis, L.D., Shiau, D.S., Sackellares, J.C., Pardalos,
P.M. and Prasad A. (2004) A dynamical resetting of the
human brain at epileptic seizures: Application of nonlin-
ear dynamics and global optimization techniques. IEEE
Transactions on Biomedical Engineering, 51, 493-506.
doi:10.1109/TBME.2003.821013
[37] Iasemidis, L.D., Sackellares, J.C., Zaveri, H.P. and Wil-
lians, W.J. (1990) Phase space topography and the Lya-
punov exponent of electrocorticograms in partial seizures.
Brain Topography, 2, 187-201. doi:10.1007/BF01140588
[38] Lehnertz, K. and Elger, C.E. (1995) Spatio-temporal dy-
namics of the primary epileptogenic area in temporal lobe
epilepsy characterized by neuronal complexity loss. Elec-
troencephalogram Clinical Neurophysiology, 95, 108-117.
doi:10.1016/0013-4694(95)00071-6
[39] Lerner, D.E. (1996) Monitoring changing dynamics with
correlation integrals: Case study of an epileptic seizure.
Physica D, 97, 563-576.
doi:10.1016/0167-2789(96)00085-1
[40] Osorio, I., Harrison, M.A.F., Lai, Y.C. and Frei, M.G.
(2001) Observations on the application of the correlation
dimension and correlation integral to the prediction of
seizures. Journal of Clinical Neurophysiology, 18, 269-
274. doi:10.1097/00004691-200105000-00006
[41] Van Quyen, M.L., Martinerie, J., Baulac, M. and Varela,
F.J. (1999) Anticipating epileptic seizures in real time by
a non-linear analysis of similarity between EEG record-
ings. NeuroReport, 10, 2149-2155.
doi:10.1097/00001756-199907130-00028
[42] Litt, B., Estellera, R., Echauz, J., D’Alessandro, M., Shor,
R., Henry, T., Pennell, P., Epstein, C., Ba kay, R., Dichter,
M. and Vachtsevanos, G. (2001) Epileptic seizures may
begin hours in advance of clinical onset: A report of five
patients. Neuron, 30, 51-64.
doi:10.1016/S0896-6273(01)00262-8
[43] D’Alessandro, M., Esteller, R., Vachtsevanos, G., Hinson,
A., Echauz J. and Litt B. (2003) Epileptic seizure predic-
tion using hybrid feature selection over multiple intrac-
ranial eeg electrode contacts: A report of four patients.
IEEE Transactions on Biomedical Engineering, 50, 603-
615. doi:10.1109/TBME.2003.810706
[44] M oser, H.R., We ber, B ., Wieser, H.G. and Meier, P.F. (1 999)
Electroencephalogram in epilepsy: Analysis and seizure
prediction within the fra mework of Lyapunov theory. Phy-
sica D, 130, 291-305.
doi:10.1016/S0167-2789(99)00043-3
[45] Hively, L.M., Protopopescu, V.A. and Gailey, P.C. (2000)
Timely detection of dynamical change in scalp EEG sig-
nals. Chaos, 10, 864-875. doi:10.1063/1.1312369
[46] Hively, L.M. and Protopopescu, V.A. (200 3) Channel-con-
sistent forewarning of epileptic events from scalp EEG.
IEEE Transactions on Biomedical Engineering, 50, 584-
593. doi:10.1109/TBME.2003.810693
[47] Sackellares, J., Iasemidis, L., Shiau, D., Gilmore, R. and Ro-
per, S. (1999) Detection of the preictal transition from
scalp EEG recordings. Epilepsia, 40, 176.
[48] Shiau, D., Iasemidis, L., Suharitdamrong, W., Dance, L.,
Chaovalitwongse, W., Pardalos, P., Carney, P. and Sac-
kellares, J. (2003) Detection of the preict al period by dy-
namical analysis of scalp EEG. Epilepsia, 44, 233-234.