Q1998 VOL. 29 NO. 2

CLINICAL ELECTROENCEPHALOGRAPHY

Mapping Epileptic Foci by the Dipole Tracing Method
in a Brain Tumor Patient with Olfactory Seizures:
Comparison with Intraoperative Electrocorticograms
Nobuyoshi Shibata, Fumio Kubota, Yukiteru Machlyama, Akio Takahashi and Keiichl Mlyamoto

Key Words
Brain Tumor
Dipole Tracing Method
Electrocorticography
Epileptic Focus
Olfactory Seizure
~~

frequency of the seizures. After about 2 months, he decided
to stop taking them. Although the seizures persisted, he did
not seek medical advice until an attack of unconsciousness
occurred at work. He was then examined in our department
in April 1995 on the advice of his superior. An EEG examination revealed frequent spikes, predominantly at FB, T4,
and T6. A MRI examination of the head revealed a homoge
neous high intensity mass on the T2-weighted image in the
right temporal lobe-hippocampus-amygdala region. The
patient was immediately referred to the Neurosurgery
department of our hospital. Anterior temporal lobectomy
was performed on July 20, 1995. The tumor was found in
the right uncus extending to the right amygdala. The size of
the tumor was about 3 x 3 x 2 cm. A 5-cm anterior temporal
lobe and a 2.5 -cm anterior hippocampus were resected.
Histologically the tumor was a benign ganglioglioma. There
have been no seizures since the operation, and no spikes
have been detected on EEG examinations to date.

~~

INTRODUCTION
The Dipole Tracing Method (DTM) was developed and
named by Homma et al.' DTM is an analysis to calculate
the location of the EEG source from recorded EEG signals
and approximates the source to one or more dipoles. DTM
can determine the location of the dipoles by iterative calculations using algorithms minimizing the squared difference between the potentials actually recorded from the
scalp and those calculated from the dipoles.
There are many brain mapping methods based on the
equivalent current dipole (ECD) which estimate the location of the electrical source.2 Several of these are based
on calculations derived from brain potentials and, as they
postulate a standardized model of the head, they are inaccurate and there have been no reports comparing the calculated positions with the epileptic foci.
In recent years four brain mapping methods, including
our methods, have been developed using the true shape
of the
All of them use a boundary element algorithm composed of three layers and accuracy has become
much higher. However, there have been few reports that
compare the calculated positions with the actual locations
of the epileptic foci. In this study we estimated the locations of epileptic foci using DTM',6in a patient with a brain
tumor and compared the results with electrocorticograms
(ECoGs) recorded during surgical resection in order to
show the high accuracy of DTM.

METHODS
An EEG was recorded from 21 electrodes arranged
according to the International 10/20 System. The reference electrode was placed on the left ear lobe. All of the
EEGs were recorded using a data recorder (XR-9000;
TEAC) for later analysis. Sampling rate was 1000Hz. Onedipole analysis was performed on the spike peaks using
an EEG analyzer (CDT-1000; Chuo Electronics Co. Ltd.,
Tokyo, Japan). Dipole estimations of the same spikes
were performed at 5 different consecutive time points with
an interval of 1 msec. Creating a three-layer (SSB)
model,' horizontal slices produced by X-ray CT at 5-mm
intervals were read by scanner, and the surface of the
scalp and the outer and inner contours of the skull were
delineated. The surface of the scalp to the outside of the
skull was designated the scalp layer, the outside of the
skull to the inside of the skull, the skull layer, and inside
the skull was the brain layer. The ratio of the electric con-

CASE REPORT
The patient was a 24-year-old male, the product of a
normal delivery. His family history for epilepsy was unremarkable. In 1990 (at 18 years of age) the patient began to
experience "bad odors" lasting 20-30 seconds 2-3 times a
day. At 20 years of age he began to experience attacks of
unconsciousness lasting about 1 minute, 1-2 times a
month, and was examined at Hospital T. The patient began
to take antiepileptic drugs, but there was no change in the

~

~~

~

-

Nobuyoshi Shibata, M D , Fumio Kubota, M D , and Yukiteru
Machiyama, M D , Ph D , are from the Department of Neuropsychiatty,and
Ako Takahashi. M D , is from the Department of Neurosurgery, Gunma
University S c h d of Medeine, Gunrna. Japan Keiichi Miyamolo IS wifh
Chuo ElecfronlcsCo Ltd ,Tokyo,Japan
Requests for repnnts should be addressed to Nobuyoshi Shibata. M D ,
Department of Neuropsychlatry. Gunrna Universty School of Medicine, 339-22 Showa-Cho, Maebashi,Gunma, 371-8511 Japan

91

Olggewx. 29 No.2

CLINICAL ELECTROENCEPHALOGRAPHY

Fpl

-

-=--

FPZ
F3 ---AF4
Cf

-

-

c4--hP3

P4
01

02

-

F7-

FE
T3
T4
TS

T6

--b---

FpZ

--

Fz

A

cz

A

PZ-

ozFigure 1.
The ECDs on a MR image, located in the

hippocampus.

Figurn 2
The ECDs on a MR image, located in the nght superior temporal gyrus.

92

OlW8 WX. 29 NO. 2

CLINICAL ELECTROENCEPHALOGRAPHY

1- 2

Figun 3.
An ECoG of the qht.-

The ECoG recorded spikes in two places, one in the
nght temporal bbe and the other in the nght hippocampus.
In the nght temporal lobe, the ECoG oneachcerebralgyfus
was recorded with bipolar leads using a strip electrode consisting of 10 electrodes in 2 rows. Later the ECoGs on the
nghthippocampuswere recordedwithtnpobleadsusinga
strip electrode consisting of 4 electrodes in a single row.
The schematic diagrams of the position of the electrodes
were made comparing them with the intraoperative video
recording showing the elecbodes on the brain.
RESULTS
Two of the 14 spikes analyzed were exduded from the
results because the dipdam was less than 98%. Twentyf i e ECDs were mapped to the right hippocampus from 5
of the 12 spikes adopted, and 35 ECDs to the right superior temporal gyfus using the remaining 7. Thus, they were
mapped to two places, the right hippocampus or the right
superior temporal gyrus, independently by DTM.
Fgure 1 shows the posrtions of the ECDs on the right
hippocampus on a MR image. The EEG on the lefl of the
image shows the typical spike analyzed. The ECDs are
located in the right hippocampus adjacent to the posterior
P O I h l Of the k&n. FQUm2 Shows the PoSikIlS Of these
ECDs on a MR image. The ECDs are located in the right
superior temporal gym.
Fgure 3 shows an ECoG of the right hippocampus. A

Figun4.
A recordingwith the strip electrode placedon the nght superior
tempordgynrs.

ductivity of the layers was calculated as l:l/W:l.' The
equivalent cunent dipole (ECD) was computed as vectors
on the coordinates of a Mimensional model of the head,
and the calculated position was marked on a MR image.
EEG recordingtime was about 1 hour, and 14 representative spikes chosen by inspection from the spikes that
appeared were used as the subject of the a n a m . As a
result of the analysis, only ECDs in which dipdarity was
98% or more were adopted to avoid aliasing.'

93

OlggevoL.29No.2

CLINICAL ELECTROENCEPHALOGRAPHY

&

A-B
C- D

E- F
G- H

K- L

.-

M- N
0- P

Q- R
S- T

u- v
w- x

pus and the other in the right superior temporal gyms.
Spikes were recorded in the right hippocampus and the
right superior temporal gyrus on the ECoGs of this patient.
These places matched the positions mapped by DTM,
proving that mapping by DTM is highly accurate.
Our patient had two foci, one near the tumor and the
other detached from it. He had no seizures caused by the
epileptic locus in the right superior temporal gym;therelore, this focus mlght be the secondary locus incubated by
the tumor. AIthough the reason why the focus appeared
was undear, the ECoG dearly showed that the epileptic
locus lay in the nght superior temporal gyrus. In the past,
the epileptic foci in patients with seizures caused by brain
lesions were claimed to lie in the vicinity of the lesion.
However, as a result of advances in brain mapping,
epileptic foci have been found not only near brain lesions,
but, in quite a few exceptional cases some distance away
horn them,or hno loci with one dose, but the other some
distance away horn them?1oWe thought our patient was
also an exceptional case.
Our subject was a brain tumor patient with seizures,
not an epileptic patient without a brain lesion. DTM postulates a uniform parenchyma 01 the brain, so the large
W i that is different from the m l brain parenchyma
in the electric conductivity makes the mapped points inaccurate. The results of the present study show that DTM is
highly accurate. panially because the brain lesion was not

schematic diagram of the position of the electrodes on the
hippocampus is at the lower left of the figure. The electrode numbers comqmd to the number of the leads on
the ECoG at the upper nght. We marked the approximate
positions of ECDs with an asterisk. On the E M , spikes
appeared in leads 1-2 and 2-3.and phase reversal is
seen,suggesting that the epileptic locus is located in the
vicinity of electrode 2. This almost matches the positions
01 the ECDs mapped in the hippocampus by DTM. Fgure
4 shows a recording with the strip electrode placed on the
right superior temporal gyms, and Figure 5 shows the
ECoG recorded in the vicinity 01 the right temporal lobe. A
schematic diagram of the electrodes placed on the cerebral cortex is shown at the left of the figure. ECoGs were
recorded at different times from the frontal lobe, superior
temporal gyrus,middle temporal gynrs, and inferior temporal gyrus, and arranged as shown in the figure. Spikes
were frequently recorded in leads I-J. Positive spikes are
seen in leads 0-P. The time that they were recorded is different but their polarity is reversed, suggesting that the
epileptic focus is in the vicinity 01 electrodes J and 0. This
site bears a close correlation with the ECDs mapped in
the right superior temporal gyms by DTM.
DISCUSSION
In the present study we used DTM to map the epileptic
loci of a patient with a brain lesion. Two independent locations were mapped as a result, one in the right hippocam-

94

0 1998 VOL. 29 NO. 2

CLINICAL ELECTROENCEPHALOGRAPHY

so large and the electrical conductivity level of the lesion
was close to that of the brain parenchyma.
Flink et all1 have reported that DTM is highly accurate
in epileptic patients, although their sample was small and
they have not examined the details. Roth et all2have illustrated the difference between the calculated position and
the epileptic foci using their method. They found a considerably large spread in the ECoGs. No mapping methods
derived from brain potentials have shown the difference
between the calculated position and the epileptic foci on a
diagram. There have been no reports that the difference
between them was calculated, because there is some
spread in both the calculated positions and the epileptic
foci and it is difficult to calculate this spread. In this study
our recording time of the ECoG was too short to examine
the brain extensively. We could, however, find the epileptic
foci clearly, as there was, fortunately, only a small spread
in the ECoGs of our patient. The results of the present
study show on a chart that the calculated positions and
the epileptic foci were very close. Therefore, we conclude
that DTM is highly accurate in patients with brain lesions
associated with epileptic seizures.
DTM is an analysis to calculate the location of the
EEG source from recorded EEG signals. Using DTM we
can find the location of the epileptic foci and get material

to judge surgical indication of epileptic patients. DTM is a
noninvasive method to examine all patients whose EEG
can be recorded. In addition, DTM only needs simple
equipment including a computer and some input units. In
this study we used a three-layer SSB model. We are now
developing a realistic four-layer (SSLB) model in consideration of liquor layer space: DTM will be developed and
become more accurate.
SUMMARY
We estimated the position of the epileptic foci in a
case of brain tumor with olfactory seizures using the
Dipole Tracing Method (DTM) and compared the results
with electrocorticograms (ECoGs) recorded during surgical resection. The case was a 24-year-old male.
Electroencephalograms (EEG) showed frequent focal
spikes in the right temporal area. Magnetic resonance
imaging revealed a tumor in the right hippocampus region.
We analyzed the spikes using DTM with a CDT-1000 EEG
analyzer. The locations of two independent foci were analyzed; one was thought to be in the right hippocampus and
the other in the right superior temporal gyrus. When the
ECoG was taken, the results were in very close correlation with those of DTM, demonstrating the accuracy of
DTM in the estimation of the location of epileptic foci in
epileptic seizures with brain lesions.

REFERENCES
1. Homma S, Musha T, Nakajima Y, et al. Location of electric
current sources in the human brain estimated by the dipole
tracing method of the scalp-skull-brain (SSB) head model.
Electroencephalogr Clin Neurophysiol 1994; 91: 374-382.

7. Homma S, Musha T, Nakajima Y, et al. Conductivity ratios of
the scalp-skull-brain head model in estimating eqivalent
dipole sources in human brain. Neurosci Research 1995;
22: 51-55.

2. Nakasato N, Levesque MF, Earth DS, Baumgartner C,
Rogers RL, Sutherling WW. Comparisons of MEG, EEG,
and ECoG source localization in neocortical partial epilepsy
in humans. ElectroencephalogrClin Neurophysiol 1994; 91:
171-178.

8. Awad IA, Rosenfeld J, Ah1 J, Hahn JF, Lirders H. Intractable
epilepsy and structural lesions of the brain: mapping, resection strategies, and seizure outcome. Epilepsia 1991; 32:
179-186.
9. Boon PA, Williamson PD, Fried I, et al. Intracranial, intraaxial, space-occupying lesions in patients with intractable partial seizures: an anatomoclinical, neuropsychological, and
surgical correlation. Epilepsia 1991; 32: 467-476.

3. Roth BJ, Balish M. Gorbach A, Sato S. How well does a
three-sphere mode( predict positions of dipoles in a realistically shaped head? Electroencephalogr Clin Neurophysiol
1993; 87: 175-184.
4. Cuffin BN. A method for localizing EEG sources in realistic
head models. IEEE Trans Biomed Eng 1995; 42: 68-71.

10. Kamada K, Isu T, Takahashi T, Tanaka T. Remote epileptogenic focus detected by electrocorticogram in a case of cavemous angioma.Acta Neurochir (Wien) 1994; 127: 236-239.

5. Yvert 8, Bertrand 0, Echallier JF, Pemier J. Improved dipole
localization using local mesh refinement of realistic head
geometries: an EEG stimulation study. Electroencephalogr
Clin Neurophysiol 1996; 99: 79-89.

11. Flink R, Sigfrid B, Hagbarth KE. The electric dipole method
in the preoperative evaluation of epilepsy surgery patients.
ElectroencephalogrClin Neurophysiol 1995; 97: 579-580.

6. Shibata N, Hattori S, Kifune A, Kubota F, Machiyama Y,
Takahashi A. Estimation of epileptic foci by dipole tracing
method in epileptic seizures with structural lesions of the
brain. Epilepsia 1997; 38 (suppl3): 210.

12. Roth BJ, KO D, von Albertini-Cadetti IR, Scaffidi D, Sato S.
Dipole localization in patients with epilepsy using the realistically shaped head model. Electroencephalogr Clin
Neurophysiol 1997; 102: 159-166.

95