IEEE IPAS' 14: INTERNATIONAL IMAGE PROCESSING APPLICATIONS AND SYSTEMS CONFERENCE 2014

Detection of High Frequency Oscillations

(HFOs) in the 80-500 Hz Range in Epilepsy
Recordings Using Decision Tree Analysis
Sahbi Chaibi*, Tarek Lajnef *, Mounir Samet*, Karim Jerbi\ and Abdennaceur Kachouri§*
* National Engineering School of Sfax, LET! Laboratory, ENIS BPW3038-Sfax, Tunisia
Email: sahbi.chaibi@yahoo.fr
tarek.lajnef@gmail.com
mounir.samet@enis.rnu.tn
.;. Psychology Department, University of Montreal, QC, Canada
Email: karim.jerbi@inserm.fr
§ ISSIG: Higher Institute of Industrial Systems, Gabes CP 6011, Tunisia
abdennaceur.kachouri@enis.rnu.tn

Abstract-Discrete High Frequency Oscillations

classified into two sub-categories: Ripples (80 -250
(HFOs) in the range of 80-500 Hz have recently
Hz) and Fast Ripples (250-500 Hz). HFOs have
received attention as a promising reliable biomarkers
been discovered in recent years in both rats and
for epileptic activity, both in scalp EEG as well as in
human models of epilepsy [I, 2, 3 ]. In fact, there is
intracranial recordings. HFOs are often characterized
by variable durations (10-100 ms) and rates of a fair amount of clinical evidence suggesting that
occurrence (17.5 ± 9.5 / min). The total duration of HFOs might be specific surrogate markers of the
HFOs is extremely small compared to the entire seizure onset zone (SOZ). In addition, elucidating
length of the EEG signals to be analyzed which, in the the role of HFOs can have a profound impact on the
case of intracerebral recordings, are generally understanding of the fundamental neural
acquired over several days and sometimes up to
mechanisms underlying epileptic phenomena. To
weeks. As a result, visual marking of HFOs events
date, during interictal periods, higher durations,
associated with large amounts of EEG data is
higher rates, and higher power in HFOs were
extremely tedious, inevitably subjective and requires a
observed within the seizure onset zone (SOZ) than
great deal of mental concentration. Therefore,
automatic detection of HFOs can be very useful to outside of it [1,4,5,6]. Furthermore, some studies
propel the clinical use of HFOs as biomarkers of suggest that HFO bursts mark epileptogenicity
epileptogenic tissue and is crucial when rather than lesion type [7]. Moreover, some studies
conducting large-scale investigations of HFO activity. on human epilepsy have also proved that there is a
As a first step towards robust and reliable automatic good correlation between post-surgical outcome
detection, we propose in this paper a new method for
and resection of the brain region containing
HFOs detection based on Decision Tree analysis. The
channels with high HFO rates [8,9]. A recent body
performance and added value of the proposed method
of research suggests that HFOs could be useful in
are evaluated by comparing it with five other
predicting the timing of the onset of epileptic
previously proposed methods. The HFO detection
performances were tested in terms of sensitivity, False seizures [\0, 11,12]. More importantly, higher
Discovery Rate (FDR) and Area Under the ROC HFOs rates are significantly correlated with higher
Curve (AUC). Our results demonstrate that the seizure frequency in epileptic patients [\3 ]. The
decision-tree approach yields low false detection ranking of channels according to HFO rate
(FDR=8.62 %) but that, in its current indicated that HFOs remained confined to the same
implementation, it is not highly sensitive to HFO
region during ictal and interictal periods and seem
events (sensitivity=66.96 %). Nevertheless some
to be a more reliable indicator of the seizure onset,
advantages of the method are discussed and paths for
while spikes and sharp waves presented a wider
further improvements are outlined.
spread during seizures than interictal periods
Keywords Epilepsy, lntracerbral EEG,
- High [14,15]. All these results have generated a strong
Frequency Oscillations (HFOs) and Decision Tree. interest in the study of HFOs and their mechanistic
role related to epileptogenisis over the past few
I. INTRODUCTION years. Generally, intracerebral EEG is recorded
High frequency oscillations (HFOs) are during one to three weeks in patients with drug­
oscillatory brain waves that reflect the short-term resistant intractable epilepsy. During this period,
synchronization of neural activity. HFOs are often clinicians seek to identify, among other clinical
markers, the presence and properties of HFOs and
978-1-4799-3866-7114/$31.00 2014 IEEE subsequently extract the maximum amount of
IEEE IPAS' 14: INTERNATIONAL IMAGE PROCESSING APPLICATIONS AND SYSTEMS CONFERENCE 2014

information that allows for the delineation of the presence of interictal HFOs upon initial review.
epileptogenic zone with the highest possible Second, a group of three consecutive patients with
precision. intractable epilepsy was considered in the current
The duration of HFOs is 10-100 ms, with an study. Finally, channels containing different
average occurrence rate of 12 ± 17 Imin for Ripples background power levels were taken into
and 7 ± 18 I min for Fast Ripples [16]. Therefore, consideration.

the duration of HFOs is extremely small compared

III. DESCION TREE HFO DETECTION METHOD
to the entire length of the entire intracerebral EEG
recordings. The visual marking of HFOs is thus Decision trees are mathematical classification

associated with large amounts of EEG data is an rules where the decision is based on successive tests

extremely tedious process and requires a great deal associated with features, values and classes

of mental concentration and focus. Fatigue or organized in a tree structure based on information

distraction during visual marking can easily lead to theory. In our case, the EEG data is composed of

errors (detection of false positives and false two types or classes (background activity and HFO

negatives) [16,17]. Visual scoring of HFOs is activity). The implementation of the decision tree is

highly time-consuming: processing 10 channels of done using the following six features (Fl to F6)

10-min recordings could take up to 10 h of hard computed from the filtered signal (x) in the 80-500

work even for an experienced reviewer. Besides, Hz band and the time-frequency map (C) in the 80-
visual methods apply subjective interpretations. By 500 Hz band.
contrast, the automatic detection of HFOs relies on • F1 is the mean + standard deviation defined as
the application of a set of mathematical rules follows:
implemented in a software environment that aimes
Fl = x + G (x) (1)
to detect and identify these events of interest not
• F2 is the difference between maximum and
only faster but also more objectively. Broadly
minimum coefficients of time frequency map
speaking, automatic detection of HFOs may have
computed using the CMOR2-1.114 wavelet
several important applications and implications,
family.
such as: i) Propelling the clinical use of HFOs as a
reliable indicator of epileptogenic zone [18, 19] ii) F2 = max I C(t, f) - min I C (t, f) (2)
t,f t,f
The development of new clinical diagnostic
procedures and novel therapeutic interventions for
• F3 is the mean + standard deviation of TEO
patients with pharmaco-resistant epilepsy, and iii) (Teager Energy Operator) :
Providing a robust tool with which neurologists and
researchers can systematically analyze large
F3 = TEO + G (TEO) (3)

amounts of EEG data much faster than via visual

scoring.
Where TEO (n) = x n ( )2 - x n ( + 1). x ( n - 1).
The TEO operator is defined in [20]. It is one
In the present paper, we propose a new method
of HFOs detection, based on Decision Tree of the most popular methods available for
analysis. The performance and added value of the estimating the instantaneous frequency and
proposed method are evaluated by comparing it amplitude.
with five other existing methods. The HFO
• F4 is the rate of zero crossings defined as
detection performances were tested in terms of
follows:

I abS(diff C + s�gn(x» ) /N
sensitivity, False Discovery Rate (FOR) and Area
(4)
Under the Curve (AUC). The organization of the F4 =

paper is as follows: After this introduction, Section

II describes the clinical EEG database used in this Where N is the length of signal x.
study. In Section III, a detailed description of the • F5 is the maximum coefficient of auto-
proposed method is provided. In Section IV, we correlation:
describe the visual marking procedure and the FS = max(Rxx (t» (5)
performance metrics. Section V describes the
results which are then followed by discussion and • F6 is the maximum of the Power Spectrum
concluding remarks. Density defined as follows :

II. CLINICAL EEG DATABASE F6 = max (PSD(x» (6)

The clinical database used in this study was
recorded using the Stellate Harmonie Routine EEG
Initially, we selected 200 events divided into
System (Natus Medical Inc.) at the Montreal
1 00 events of background activity and 100 events
Neurological Institute and Hospital (MNI), Canada.
of HFO activity. Then, the 6 different features
At acquisition, the data was low-pass filtered at
associated with these events are calculated. The
500Hz and sampled at 2000 Hz. The sampled data
matrix containing the values of features and the
was then quantized using a 16 bit analog-to-digital
associated matrix classes of these events are used in
converter. The used EEG channels were selected
the learning phase to produce the optimized tree
based on the following criteria: First, the clear
IEEE IPAS' 14: INTERNATIONAL IMAGE PROCESSING APPLICATIONS AND SYSTEMS CONFERENCE 2014

(Tmax) which we will use in automatic recognition. Orectified = p. () ( 9)

The sequencing of the features and their
optimization in Tmax are measured based on the (J is the standard deviation of filtered and rectified

Entropy Gain [21] which is defined as follows: baselines segments selected manually. � is the input
training parameter to be optimized. Learning and
automatic recognition flowcharts of HFOs by the
" IXai=)
- HeX)
. ( 7)
Gam ex,aj ) -
method based on the decision tree is shown in
1 Ix - (Xai=J
valeurs (ai) --
- L.. v E
H
Figurel .

Where Xaj=v is the set of examples, on which a
considered feature aj takes the value
notation, IXI indicates the cardinality of the set X.
Given a set X of examples of which a portion p+ is
positive and portion p_ is negative. (with p+ + p_
IV. VISUAL MARKING OF HFO EVENTS AND
PERFORMANCE METRICS
v and the
Visual marking of HFOs was performed by two
reviewers trained in electrophysiology and HFO
=
analysis. Each event was marked as a relevant

1). The entropy HeX) is defined as follows:
Where 0 <;:: HeX) <;::1.
Once the learning phase is completed, the automatic
recognition of HFOs using the optimized decision
tree (Tmax) can be done.
Based on the six features FI-F6 and Tmax, each
associated window positioned at every time sample
is tested, to verify whether it matches the HFO class
or the background class. First, a sliding window of
length 50 ms with predefined overlap of 1 sample
scans all the samples of the original EEG signal.
The length of window (50 ms) is approximately
equal to the average duration of all HFOs events
used in the learning phase. Once scanning the entire
signal is completed, the samples corresponding to
the HFO class are set to 1 and the other segments
are set to O. Next, the segments with value 1 are
delimited in time. Each portion of the filtered and
rectified signal with level 1 above the threshold
ei\ectified) is marked as a probable HFO if it has at
least six peaks (equivalent to
threshold C\ectified is defined as follows:
HFO; if it is seen as an oscillation with at least
three consecutive cycles in the frequency band 80-
500 Hz, and could be clearly distinguished from the
average of the background in the filtered signal (80-
500Hz). In addition, to be confirmed, each event is
visually verified also in the unfiltered EEG signal.
Every event detected by the two reviewers was
considered as a relevant HFO burst. In contrast, the
remaining segments were considered "background
activity". Note that although these segments may
contain various other oscillatory or transient events,
we use the term "background activity" here as an
indicator for the absence of an HFO event.
The performance measures we used are sensitivity,
False Discovery Rate (FOR) and Area Under a
Curve (AUC), which are respectively defined as:
0p
. . . (10)
SenSltivlty = 100 -
pos
FP (II)
FOR= 100 --­
TP + FP
(12)
3 cycles). The

EEG signal EEG signal

Temporal filtering and time frequency filtering in 80-500 Hz Temnoral filterin!:! and time freauencv filterin!:! in 80-500 Hz

*
100 HFO events,100 Background events Sliding window (SOms) with I sample overlap

*
Feature Calculation Feature Calculation

Optimization of decision tree (Tmax)

Associated classes ( HFO. Background) Detection of putative segments of HFOs

t
Thresholding by minimal cycle number (3 cycles)
Learning

Detected HFOs events

t
Automatic recognition

Fig. I. Learning and automatic recognition of HFO using the proposed descion tree framework.
IEEE IPAS' 14: INTERNATIONAL IMAGE PROCESSING APPLICATIONS AND SYSTEMS CONFERENCE 2014 4

The HFO detection algorithm produces a list of previously implemented, namely: RMS [3 ], CMOR
probable HFOs with their location and duration. [22], Bumps [23], Matching Pursuit [24] and HHT
Probable HFOs may correspond to positives or [25].
negatives. Positives: are the HFOs segments The RMS detector proposed by Staba [3 ] is
visually identified. Negatives: are different based on linear finite impulse response (FIR) filter
background segments that do not contain HFOs and the moving average of the root mean square
(We use Pas to indicate the number of positives, (RMS) feature. The description regarding the
and Dpas (Detected Positives) as the number of implementation and the varying input parameters
positives which overlap with at least one probable for this method are provided in [3 ]. The second
algorithm for HFOs detection is based on the
HFO. TP (True Positives): is the number of
complex Morlet wavelet (CMOR). The description
probable HFOs which overlap with at least one
of this method is provided in [22]. The bumps
positive event. FP (False Positives): is the number
detector was presented by Doshi (2011). This
of probable HFOs which do not overlap with any
method is based on bumps modeling technique. The
positive event. AUC is the area under the ROC
description of the implementation and various input
curve.
parameters for this method are provided in [23 ].
The method based on matching pursuit technique
V. RESULTS & DISCUSSION
for HFOs detection is based on classical matching
th
pursuit as described in [24]. Finally, the 5 method
The calculation of the performance (sensitivity to which we compared our approach is the HHT
and FOR) of the proposed algorithm is done by detector, was presented by Chaibi et al.(2013 ) and
changing the input parameter (P). Figure 2 shows that is based on Hilbert Huang Transform (HHT)
the ROC curve which characterizes the variation the and RMS features (see [25] for a formal description
sensitivity as a function of the FOR. The optimal and input parameter details). The results of
threshold is indicated by the blue arrow that performance comparison between different methods
corresponds to P = 2.66. Sensitivity and FDR are shown in Figure 3 and summarized in Table 1.
corresponding to this point are 66.96% and 8.62%
As shown in Figure 3 and Table 1, the most
respectively. The AUC of the ROC curve is 0.8208.
robust method -with the data used here- in terms of
best compromise between sensitivity and FOR
appears to be the HHT-based method yielding the
largest difference between the sensitivity and the
FOR. On the other hand, the method based on the
complex Morlet wavelet (CMOR) is the one that
provides the highest AUC. It is also important to
note that the bumps method can achieve a
sensitivity of 98% with an FOR that is
approximately equal to 23 % (Figure 3 ). One
advantage of the proposed decision-tree method

100
compared to other techniques, is that it is the
FDR method that provides the lowest false detections

Fig. 2. ROC curve of the algorithm based on the descion tree
obtained by varying the input parameter (�).
The comparison of performances between
several automated detectors of HFOs is not an easy
task. There are several challenges that account for
this difficulty: There are different techniques of
recording of EEG signals. Different types and
locations of electrodes have been used in the
recordings of EEG signals. The defmitions of HFOs
(e.g. frequency bands) vary across the literature. In
addition, there is also a lack of a "Gold Standard"
and a common database for researchers working on
the detection of HFOs. So, caution should be taken
into consideration when comparing the performance
among distinct detection algorithms.
Therefore, it is crucial to test different methods
using the same database and the same "Gold
Standard". For this reason, we propose in this paper
a comparative study between our proposed
algorithm and five other methods that have been
(FDR = 8.62%) when considering the ROC point of
best compromise (see arrows on Fig.3). However,
the main limitation of the current implementation of
the decision-tree method is the lack of sensitivity
(66.96%). This sensitivity is not particularly poor
but it is lower than what we report with the other
methods. Indeed, the choice of the method to be
used for detecting HFOs depends on the purpose,
the intended user and the minimal required
sensitivity, FOR and AUC performances (the latter
may depend on whether the method is used in a
fully or semi-automatic framework). In a semi­
automatic approach, an excessively sensItlve
method can be later corrected by visual inspection
of the results of the automatic procedure. In a fully
automatic method (or a pipeline that automatically
uses the output of the detector for further analysis
and decisions), a minimal false detection rate might
be a priority, even if it comes at the cost of missing
some events. As shown in Table 1, the FOR that
was obtained with the other 5 reference methods is
IEEE IPAS' 14: INTERNATIONAL IMAGE PROCESSING APPLICATIONS AND SYSTEMS CONFERENCE 2014

probably too high for practical and clinical positives that arise from spurious high-frequency
implementations. It is important to keep in mind bursts that are not true physiological HFOs and
here that, particularly in the quest for identifying which can be merely caused by the filtering of
HFOs, several phenomena can lead to false spikes and sharp waves.

100

90

80

70

60
Sensitivity

Bumps
50 MP
CMOR
40
Decision Tree
30 HHT
RMS
20

10

0
0 10 20 30 40 50 60 70 80 90 100
FDR

Fig. 3. Different ROC curves associated with different tested methods of HFOs detection.

TABLE!. OVERVIEW OF PERFORMANCE FOR ALL TESTED METHODS.

Method Author and Year Sensitivity (%) FDR(%) AliC

RMS Staba,2002 83.14 23.13 0.8723

CMOR Khalilov,2005 87.00 14.12 0.9406

Bumps Doshi,2011 95.86 20.96 0.9ll6

MP Chaibi,2013 88.93 16.62 0.8496

HHT Chaibi,2013 92.56 10.57 0.9220

Descion Tree Chaibi,2014 66.96 08.62 0.8208

VI. CONCLUSION among sensitivity, FOR is generally sought.

Keeping in mind the limitations of each method and
HFOs can be a powerful, reliable and accurate
the possible occurrence of false HFOs bursts, it is
functional indicator of the location of epileptogenic
likely that the best strategy will be a combination of
tissue in patients evaluated for surgical treatment.
different methods. Additionally, the field would
In addition, they may provide useful information
benefit from using standardized benchmarking data
and a strong impact on our understanding the
sets. Finally, as in all evaluations of detection
fundamental mechanisms underlying epileptogenic
methods that use human expert identifications as
zones and the relationship between epileptogenicity
ground truth, it is important to keep in mind that
and seizures. These advantages have produced an
human error could be a further source of variability
increasing interest in studying HFOs in recent years
when evaluating the performance of automatic HFO
and, in parallel, an increase in methodological
detectors and that one of the objectives of automatic
papers that have focused on the automated detection
detection is to reduce this variability by enforcing
of HFOs. Highly variable performances have been
objective criteria into the procedure.
reported across the various studies. The optimal
method must have (i) a high sensitivity (in order to
detect correctly all or most of the true HFO events),
(ii) as few false detections as possible and (ii)
should not be sensitive to spikes and transients
activities which can generate spurious HFO-like
events. These requirements are difficult to fulfill
simultaneously, and a compromise (trade-oft)
IEEE [PAS'14: INTERNATIONAL IMAGE PROCESSING APPLICATIONS AND SYSTEMS CONFERENCE 2014 6

Acknowledgment [16] Naeini RY. [Doctoral thesis] Automatic detection of high frequency
oscillations ofneural signals in epileptic patients [Doctoral thesis].
Authors wish to thank the Montreal Neurological Institute Montreal,Quebec,Canada:Concordia University; 2012.

and Hospital (MNI, Canada) for kindly providing the database [17] Gardner AB, Worrel GA, Marsh E, Dlugos D, Litt B. Human and
automated detec-tion of high-frequency oscillations in clinical
which we used to test and compare the performance of various
intracranial EEG recordings. ClinNeurophysiol 2007; 118:1134-43.
methods. We would like to thank Dr. Mohamed Dogui
[18] Zelmann R, Mari F, Jacobs J, Zijilmans M, Dubeau F, Gotman J. A
(Service of Functional Exploration of the Nervous System, comparisonbetween detectors of high frequency oscillations. Clin
CHU Sahloul of Sousse, Tunisia) for helping us doing visual NeurophysioI 2012;123:106-16.
identification process of HFOs in intracerbral EEG signals. [19] G.A. Worrell, K. Jerbi, K. Kobayashi, J.M. Lina, R. Zelmann, M. Le
Van Quyen. Recording and analysis techniques for high-frequency
References oscillations. Progress in lITeurobiology 2012- 98: 265-278.
[20] E. Kvedalen, Signal processing using the Teager Energy Operator and
othernonlinear operators, University of Oslo, 2003 (Cand. Scient
[I] Anatol Bragin, Jerome Engel, Charles L. Wilson, Ttzhak Fried , Gary Thesis).
W. Mathern. Hippocampal and entorhinal cortex high-frequency [21] Lachiche N., Apprentissage automatique : Arbres de decision. 2008.
oscillations (100-500 Hz) in human epileptic brain and in kainic
[22] Khalilov, I., Le Van Quyen, M., Gozlan, H., Ben-Ari, Y., 2005.
acid-treated rats with chronic seizures. Epilepsia 1999; 40(2): 127-
Epileptogenic actions of GABA and fast oscillations in the
137. developing hippocampus. lITeuron 2005; 48:787-796.
[2] Anatol Bragin, Charles L. Wilson, Richard J. Staba, Mark Reddick,
[23] Chiran Dilip Doshi. Methods for detecting high frequency oscillations
Itzhak Fried, Jerome Engel. Interictal High-Frequency Oscillations in ongoing brain signals: Applications to the determination of epileptic
(80-500Hz) in the Human Epileptic Brain: Entorhinal Cortex. Ann seizure zones,Marquette University,20II.
lITeuroI 2002; 52: 407-415.
[24] Sahbi Chaibi, Tarek Lajnef, Zied Sakka, Mounir Samet, Abdennaceur
[3] Staba, RJ.,Wilson,C.L.,Bragin,A., Fried, I., Engel, J.J. Quantitative Kachouri. A Comparison of Methods for Detection of High Frequency
analysis of high-frequency oscillations (80-500 Hz) recorded in
Oscillations (HFOs) in Human Intacerberal EEG Recordings.
human epileptic hippocampus and entorhinal cortex. J. lITeurophysiol American Journal of Signal Processing 2013,3(2): 25-34.
2002 ; 88 :1743-2152.
[25] Sahbi Chaibi, Zied Sakka, Tarek Lajnef, Mounir Samet, Abdennaceur
[4] MacReady N. Radiotherapy and localization of seizures cited as
Kachouri. Automated detection and classification of high frequency
promising therapies. Epilepsy Neurol Today 2008;8:12-4. oscillations (HFOs) in human intracereberal EEG. Biomedical Signal
[5] Elena Urrestarazu, Rahul Chander, Francois Dubeau, Jean Gotman. Processing and Control 8 (2013) 927- 934.
Interictal high-frequency oscillations (100-500Hz) in the intracerebral
EEG of epileptic patients. Brain 2007; 130: 2354-2366.
[6] Julia Jacobs, Pierre LeVan. Rahul Chander, Jeffery Hall, Francois
Dubeau, Jean Gotman. Interictal high-frequency oscillations (80-500
Hz) are an indicator of seizure onset areas independent of spikes in the
human epileptic brain. Epilepsia 2008; 49(11) :1893-1907.
[7] Jacobs J, LeVan P, Chatillon CE, Olivier A, Dubeau F, Gotman J.
High frequency oscil-Iations in intracranial EEGs mark
epileptogenicity rather than lesion type. Brain2009a;132:I022-37.
[8] Jacobs J, Zijlmans M, Zelmann R, Chatillon CE,Hall J, Olivier A, et
al. High-frequency electroencephalographic oscillations correlate with
outcome of epilepsy surgery. Ann lITeurol2010; 67:209-20.
[9] Wu JY, Sankar R, Lerner JT, Matsumoto JH, Vinters HV, Mathern
GW. Removing interictal fast ripples on electrocorticography linked
with seizure freedom in children. lITeurology 2010;75:1686-1694.
[10] Cuevas LA, Castillo-Toledo B, Medina-Ceja L, Ventura-Mejia C,
PardoPen K. An algo-rithm for on-line detection of high frequency
oscillations related to epilepsy. Comput Methods Prog Biomed
2013;110:354-60.
[II] Khosravani H,Mehrotra N,Rigby M,J. Hader W,Pinnegar CR,Pillay
N, et al.Spatial localization and time-dependant changes of
electrographic high fre-quency oscillations in human temporal lobe
epilepsy. Epilepsia 2009;50:605-16.
[12] Kalitzin S, Zijlmans M, Petkov G, Velis D, Claus S, Visser G, et al.
Quantification ofspontaneous and evoked HFO's in SEEG recordings
and prospective for pre-surgical diagnostics. Case study. In: 34th
Annual International Conference ofthe IEEE EMBS; 2012.
[13] Zijlmans M, Jacobs J, Zelmann R, Dubeau F, Gotman J. High
frequency oscillations and seizure frequency in patients with focal
epilepsy. Epilepsy Res 2009;85:287-92.
[14] Salami P, Levesque M, Gotman J, Avoli M. A comparison between
automated detec-tion methods of high-frequency oscillations (80-500
Hz) during seizures. JNeurosci Methods 2012;211:265-71.
[15] Gotman J. The epileptic focus: a zone or a network? The interictal
focus: spikes andHFOs. In: 30th International Epilepsy Congress
Montreal. Montreal Neurologicallnstitute McGill University; 2013.