Cortical Activation Mapping of Epileptiform Activity Derived From Interictal ECoG Spikes

8/3/2019 Cortical Activation Mapping of Epileptiform Activity Derived From Interictal ECoG Spikes

1/10

Epilepsia, 48(2):305314, 2007Blackwell Publishing, Inc.C 2007 International League Against Epilepsy

Cortical Activation Mapping of Epileptiform Activity Derived

from Interictal ECoG Spikes

Yuan Lai, Wim van Drongelen, Kurt Hecox, David Frim, Michael Kohrman, and Bin He

Department of Biomedical Engineering, University of Minnesota, Minneapolis, Minnesota; and Department of Pediatricsand Surgery, University of Chicago, Chicago, Illinois, U.S.A.

Summary: Purpose: To develop and evaluate a new corticalactivation mapping (CAM) method to obtain the neuronal acti-vation sequences from the cortical potential distributions.

Methods: Interictal electrocorticogram (ECoG) recordingswere analyzed for eight pediatric epilepsy patients to find thecortical activation maps, which were compared with thepatientsseizure-onset zones identified from ictal ECoG recordings. Vari-ousrelations betweenthe local activation time andcorticalpoten-tial were assumed. The most effective relation was determinedby accessing their capability to predict the seizure-onset zone.Computer simulations using a moving dipole source model werealso conducted to test the present approach in imaging the prop-agated cortical activity.

Results: In both clinical data analysis and computer simula-tions, the maximal amplitude proved to be the most effectivecriterion with which to determine the local cortical activation

time. The present method successfully predicted the seizure-onset zone in seven of eight patients by the CAM analysis ofECoG-recorded interictal spikes (IISs). For patients with mul-tiple seizure foci, each focus can be revealed by analyzing IISs

with different spatial patterns.Conclusions: The time difference between spike peaks of

the interictal events in the leading channel and other channelscan be effectively defined as the local cortical activationtime. The cortical activation mapping method based onthis time latency can be used to predict the seizure-onsetzones, suggesting that the present CAM method is usefulto assist the presurgical evaluation for the epilepsy patients.Key Words: Cortical activation mappingInterictal spikeElectroencephalographyECoGNeuronal propagationActivation timeSeizure-onset zoneCortical potentialdistributionEpilepsySource localization.

The interictal spike (IIS) is often used to locate epilep-

tiform activity and thus is clinically important. It is usu-

ally assumed that regions displaying the largest amplitude

of spikes are within the epileptogenic zone, and conse-

quently, such regions should be resected if possible to

render the patients seizure free (Tsai et al., 1993a, 1993b;

Kanazawa et al., 1996). However, the IISs may propa-

gate from their initial sites by uncertain neural pathways

(Alarcon et al., 1994, 1997; Ebersole, 2000; Ulbert et al.,

2004). Therefore one cannot conclude that the potential

amplitude distribution at the spike peak best represents

the character of the spike source unless the distributiondoes not change with temporal evolution of the interictal

discharge.

The distinction between primary and propagated spikes

was introduced as early as 1950s (Penfield and Jasper,

1954). Recent studies suggest rapid propagation of IISs

Accepted September 13, 2006.

Address correspondence and reprint requests to Dr. B. He at Uni-versity of Minnesota, Department of Biomedical Engineering, 7-105Hasselmo Hall, 312 Church Street SE, Minneapolis, MN 55455, U.S.A.

E-mail: [email protected]: 10.1111/j.1528-1167.2006.00936.x

during the discharges (Sutherling and Barth, 1989; Alar-

con et al., 1994, 1997; Emerson et al., 1995; Merlet et al.,

1996; Merlet and Gotman, 1999; Leal et al., 2002; Lantz

et al., 2003; Asano et al., 2003). Thus it is important to

discern where the IIS starts and how it propagates to lo-

cate its generators accurately. It has been shown that the

identification of brain regions initiating the epileptiform

activities can be potentially used to reduce the necessary

resection area (Alarcon et al., 1997).

Both human and animal experiments have been de-

signedto study theinitiation andpropagationof theepilep-

tiform activity (Alarcon et al., 1994; Tsau et al., 1999;Ulbert et al., 2004). Alarcon et al. (1994) used both depth

and surface electrodes to record the interictal discharges,

where the results regarding latency and spatial distribu-

tion suggest that relatively large areas of neocortex and

archicortex can be simultaneously or consecutively acti-

vated through fast association fibers or propagation along

the cortex during interictal activities. Tsau et al. (1999)

found that spontaneous epileptiform activity could be ini-

tiated in various cortical layers on neocortical slices har-

vested from rats. The activations were found to start from

a small area (less than 0.04 mm3) and spread smoothly

305


2/10

306 Y. LAI ET AL.

TABLE 1. Summary of patients in the present study

Patient Age (yr) Gender Res. Operation Outcome

1 8 F 2 RT, RP SSR2 6 M 2 LF, LP SF3 7 M 2 LF, LP SSR4 16 M 1 LF, LT SF

5 12 M 2 RF, RP SF6 12 F 2 LF, LO SSR7 10 F 2 LP SSR8 11 M 1 LO SF

L, left; R, right; T, temporal; P, parietal; F, frontal; O, occipital; SSR,

substantial seizure reduction; SF, seizure free; Res., resection.

from the initiation site to adjacent cortical areas, suggest-

ing that the initiation site is very confined to one of the

cortical layers. Ulbert et al. (2004) concluded from their

laminar analysis of human interictal spikes that the corti-

cal layer where the initial depolarization occurs may differaccording to whether the IIS is locally generated or propa-

gated from a distant location. All of these studies showthe

clinical importance of studying the initiation and propaga-

tion of the IIS and its significance in assisting presurgical

assessment.

In the present study, we propose a new activation-

mapping method to image the neuronal propagations from

subdural ECoG recorded during interictal spikes. The per-

formance of the proposed method was evaluated by com-

paring the results with the patients seizure-onset zone

identified from ECoG ictal recordings and surgical out-

comes. It should be mentioned that our special interest

in pediatric patients makes the subdural ECoG record-

ing appropriate to be used because children usually have

superficial neocortical epileptogenic foci close to the sub-

dural recording surface, generating much less deviation

than those foci located in deep structures such as the hip-

pocampus or amygdala.

MATERIALS AND METHODS

Patients

Eight pediatric patients (three girls, five boys; 6 to

16 years old; Table 1) with medically intractable par-

tial seizures were studied by using a protocol approvedby the Institutional Review Boards of the University of

Minnesota and the University of Chicago. Preoperative

MRI/CT scans and interictal and ictal long-term video-

EEG monitoring were conducted in the Pediatric Epilepsy

Center at the University of Chicago Childrens Hospital.

Among these patients, postoperative diagnosis showed

that seven of them had two epileptogenic foci, and one

of them had single focus. All patients had extratempo-

ral seizure foci (including frontal, parietal, and occipital

lobes) no matter how many brain areas were involved in

the seizure generation.

Patient 1 underwent a right anterior temporal lobectomy

including resection of the parahippocampal gyrus, and a

separate right parietal topectomy. Seizure frequency was

reduced from 10 seizures per week to 2.75 per week, after

surgery. Patient 2 underwent a left frontal lobectomy and

a left parietal topectomy and was seizure free at 2 year

follow-up. Patient 3 underwent left frontal and left pari-etaltopectomies, experiencing 85% seizure reduction after

surgery. Patient 4 underwentleft frontal lobectomy and left

temporal lobe disconnection and hippocampectomy. The

patient was seizure free at the 2 year postoperative check.

Patient 5 underwent a right subtotal frontal lobectomy,

and a right parietal topectomy, and initially experiencing

a 90% reduction in seizures. Seizures recurred prompting

a second surgery, for a right functional hemispherectomy

and leading to seizure free at 6 month follow-up. Patient 6

underwent left frontal and left occipital topectomies and

had seizure reduction from seven to eight per week to

two per week. Patient 7 underwent a left partial occipital

lobectomy, and seizures were reduced from multiple daily

seizures to five seizures per week. Patient 8 underwent a

left occipital lobectomy and was seizure free at 1 year

follow-up.

For eachpatient, long-term ECoG recordingswere care-

fully inspectedfor theoccurrence of interictal spikes. Only

those patients whose interictal ECoG spikes showed time

delays >10 ms at various recording sites were recruited

in the present study, so that the latency difference could

be determined without ambiguity under the 400-Hz ECoG

sampling rate.

Data acquisitionDuring the surgicalmonitoring for the epilepsy patients,

ECoGs were recorded by using multiple rectangular elec-

trode grids (8 8, 8 6, 8 4, etc.) and strips (8 1,

6 1, 4 1) with interelectrode distances of 810 mm

placed directly on the cortical surface. The ECoG record-

ings were referenced to the contralateral mastoid, sampled

at 400 Hz, and band-pass filtered from 1 to 100 Hz (BMSI

6000; Nicolet Biomedical Inc., Madison, WI, U.S.A.). For

each patient, multiple interictal spikes were visually iden-

tified according to the IFSECN criteria (Chatrian et al.,

1974). A time window of 200 ms centered at the peak of

the global field power (GFP) was used to select the spike

epochs for further analysis. In cases in which >50-mslatency differences were observed, only multichannel pat-

terns, repeatedly recorded, were analyzed to exclude the

possibility that the large latency differences did not repre-

sent recordings from independent asynchronous interictal

foci (Alarcon et al., 1994).

Cortical activation mapping

We hypothesized that the latency differences among

different ECoG recording sites were caused by neuronal

propagation, that is, not due to volume conduction. Unlike

cortical current density imaging (Dale and Sereno, 1993;

Epilepsia, Vol. 48, No. 2, 2007


3/10

CORTICAL ACTIVATION MAPPING DURING INTERICTAL SPIKES 307

Babiloni et al., 2005) or cortical potential imaging (He et

al., 1999, 2001, 2002; Zhang et al, 2003), which represent

electrical current density or potential field distributions

over the cortical surface at each instant, cortical activation

mapping (CAM) attempts to obtain the sequence of propa-

gation of neuronal activation over the surface of the cortex

from the spatiotemporal cortical potential distributions. Inthe present study, the cortical activation refers to the neu-

ronal activity due to the propagation as observed over the

cortical surface. It does not refer to cortical electrical or

pharmacologic stimulation. In the present study, we as-

sumed that a relation exists between the local neuronal

activation time and the cortical surface potentials. By ana-

lyzing the subdurally recorded interictal spikes, we aimed

to find a relation in terms of the degree of consistency with

respect to the seizure onset and propagation.

The activation time is defined as the time instant when

the local tissue is excited. Four criteria have been consid-

ered by using the peak amplitude, peak first derivative,

peak second derivative, and peak laplacian as indicators

of cortical neuronal activation. In other words, the cortical

activation time is determined from the time instant, when

the absolute value of (a) the amplitude of ECoG, (b) the

first temporal derivative of ECoG, (c) the second temporal

derivative of ECoG, and (d) the surface laplacian of ECoG

reaches their maximum. For these four relations, we tested

their abilities for predicting the initiation and propagation

of the epileptiform activities by analyzing the subdurally

recorded interictal spikes found in eight pediatric epilepsy

patients. We assumed that the cortical sites that show the

shortest latency may representthe initiation zone and those

cortical sites with later activations are to be found on thepropagation pathways of the epileptiform activities.

All ECoG channels were classified into activated or in-

activated channels according to their peak activities. A

channel is determined to be activated if its maximal am-

plitude exceeds 150% of the background activity. For the

activated channels, the four aforementioned criteria were

used to determine the local activation times as an estima-

tion of the cortical neuronal activation sequence during

the propagation of interictal spikes.

The estimated cortical activation sequences were then

compared with the ictal subdural recordings. The subdural

ictal ECoG recordings were visually inspected to deter-mine the onset zone and the pathway by which the activi-

ties spread to neighboring cortical areas. The relation that

led to the most consistent estimation of propagation pat-

tern was considered the optimal relation, suggesting that it

can indicate the initiation and propagation of epileptiform

activities.

Computer simulations

Besides the clinical data analysis, computer simula-

tions were conducted to demonstrate the abilities of differ-

ent criteria in estimating the cortical activation sequence.

The physiologic mechanism for the generation of human

interictal spikes is still poorly understood and therefore

complicated to model. Previous studies showed that sin-

gle moving dipoles can be used to model the genera-

tors of interictal epileptiform discharges (Krings et al.,

1998; Lemieux et al., 2001). Here we assumed a simpli-

fied source model to generate the IISs, in which a single

moving dipole of constant strength traveled along a lineinside the brain. The three-concentric-sphere head model

was used as the volume conductor model to calculate an-

alytically the dipole-generated cortical potential distribu-

tions,from which thelocalactivation times were estimated

by using the four given relations. These estimated activa-

tion sequences have been compared with the traces of the

moving dipoles to evaluate if and how CAM analysis can

reflect the source activities. Different source configura-

tions have been tested in the computer simulations with

various dipole eccentricities and orientations. In a certain

group of simulations, the dipole was assumed to move

along the diagonal on the cortical electrode grid with a

constant speed. The reference local activation time was

then determined by the time when the dipole was closest

to the specified cortical sites. Because the local activation

times at these cortical sites were also calculated by using

the four different criteria, the correlation coefficients were

computed between the reference and calculated activation

times. The relation that gives the most consistent results is

considered to represent the optimal relation between the

potential distribution and local cortical activation time.

RESULTS

The CAM analysis was performed to analyze ECoGrecordings during interictal spikes for eight pediatric pa-

tients with intractable seizures who either were seizure

free or had substantial seizure reduction after surgical re-

section. For each patient, only the interictal ECoG spikes

with 10-ms latency difference (four time points at 400-

Hz sampling rate) between the occurrences of peaks at

various recording sites were analyzed.

Cortical activation mapping in patients

The CAM analysis was performed in all eight patients

by using the four criteria. Typical results from two patients

(patients 3 and 5) are shown in Figs. 1 and 2. According

to the results from all of the patients with the exception ofpatient 6, the peak amplitude criterion returned the most

consistent estimate of ictal-onset zone (IOZ). The peak

first derivative criterion also revealed the initiation zones

that surround the IOZ, although it is not as accurate as the

peak amplitude criterion. In most patients, the peak sec-

ond derivative and peak laplacian criteria did not provide

reliable estimates of the IOZ.

Fig. 1a shows one example of an ECoG-recorded

interictal spike for patient 5. The time latencies can be ob-

served from the occurrence of the spike peak in different

channels. Clinical diagnosis found this patient to have

had a right frontal seizure focus. Figs. 1be are the



4/10

308 Y. LAI ET AL.

FIG. 1. Cortical activation mapping (CAM) analysis for patient 5. a: ECoG waveforms during an interictal spike recorded by 6 x 8 subduralelectrodes. b: Cortical activation time (CAT) determined by the time of occurrence of peak amplitude of the ECoG recordings from differentchannels. c: CAT obtained by the peak first derivative criteria. d: CAT obtained by peak second derivative criteria. e: CAT obtained bypeak laplacian criteria. f: ECoG recording channels and ictal-onset zone (IOZ). Black circles, Subdural electrodes. Pink circles, Corticalarea where seizures started.

FIG. 2. CAM analysis for patient 3. For details, see the caption of Fig. 1.



5/10


FIG. 3. An example in which CAM analysis does not correspond to IOZ determined from ictal ECoG recordings. a: Waveforms of interictalECoG recordings from patient 6. b: CAT map obtained from CAM analysis. c: Illustration of IOZ in patient 6 determined from ictal ECoGrecordings. Black circles, Subdural grid electrodes. Pink circles, IOZ determined from ictal ECoG recordings.

activation time mappings determined by criteria

(a) through (d), respectively. From Figs. 1b and c, time

latencies of 50 ms among various channels were

revealed, suggesting that the interictal activity initiated

from the lower left corner of the electrode plate and

propagated to the opposite corner. As comparison, it ishard to tell where the interictal event started from Figs.

1d and e. Figure 1f displays the configuration of subdural

electrodes, and those electrodes marked as pink are the

seizure onset identified by examining the ictal ECoG

recordings in the same patient.

The results of CAM analysis for patient 3 are displayed

in Fig. 2. This patient had a left parietal seizure focus as

shown in Fig. 2f, where the seizures were identified to ini-

tiate from the pink channels numbered 16, 24, and 32. In

comparing Figs. 2b to e with Fig. 2f, the activation map-

ping determined by criterion (a) is most consistent with

the IOZ identified from the ictal ECoG recording. Figures

2d and e show disordered activation patterns, where the

initiation of the interictal activity canhardlybe recognized

from these patterns.

For patient 6, the cortical area initiating the interictal

spikes revealed by CAM analysis was localized in thearea adjacent to the IOZ, as determined from ictal ECoG

recordings. Figure 3a shows one typical interictal ECoG

waveform in patient 6. Obvious time delays can be ob-

served among the different recording channels from the

waveform. Figure 3b displays the results by CAM anal-

ysis for this interictal spike. However, the revealed ini-

tiation cortical area of the IIS does not overlap with the

IOZ, which is represented by the pink electrodes on the

enlarged display of the intracranial electrode grid shown

in Fig. 3c, but rather is in anareawith2 cm distance from

the IOZ. The CAM analysis of other IISs in this patient



6/10

310 Y. LAI ET AL.

FIG. 4. Illustration of CAM analysis for patients with multiple epileptogenic foci. a: Waveforms of IISs with pattern 1 in patient 7. b: CAManalysis of the IIS shown in (a). c: Waveforms of IISs with pattern 2. d: CAM analysis of IISs shown in (c). e: Two epileptogenic zoneshave been indicated with blue circles. Focus 1 included channels 3, 4, 11, 12, and 13. Focus 2 included channels 33 and 34.

rendered similar results as that in Fig. 3b. Among the eight

total patients in the present study, this is the only patient

whose CAManalysiscannot successfully predict the IOZ.

In the present study, patients 17 had two epileptogenic

foci located in either the same or different lobes. Inspec-tion of their interictal spikes showed that different spa-

tiotemporal patterns existed among the IISs in the same

patient. Clustering of theIISs accordingto their spatiotem-

poral patterns is thus needed before the CAM analysis for

each type of IIS.

Patient 7 had two seizure foci located in the left parietal

lobe, as shown in Fig. 4e. Focus 1 was more anterior and

superior than focus 2. Two different patterns of interictal

ECoG spikes have been identified for this patient. Pat-

tern 1 always involved more anterior and superior parietal

activity, as shown in Fig. 4a. The results from the CAM

analysis for this type of IISs in Fig. 4b revealed the lo-cation of seizure generator 1. Figure 4c shows the typical

waveform of pattern 2, which mainly contains the more

posterior and inferior parietal activity. The CAM analy-

sis for the IISs with pattern 2 successfully indicated the

location of seizure focus 2, as shown in Fig. 4d. These

results suggest that IISs with different patterns should be

classified according to their spatial distributions for the

patients with multiple seizure foci (six of eight patients in

the present study) and then analyzed separately to locate

the individual epileptogenic zone.

Cortical activation mapping by computer simulations

Figures 5 and 6 show the results of the CAM analysis

from the computer simulations by using the protocol de-

scribed in the Methods section. A single moving dipole

oriented either radially or tangentially with different ec-centricities (0.60 to 0.85) was used to generate the corti-

cal potential distributions, from which the cortical activa-

tion sequences were estimated by using criteria (a)(b).

Although only the results from eccentricity of 0.80 are

included in Figs. 5 and 6, similar results were obtained

from simulations with other eccentricities. Table 2 sum-

marizes the results from these computer simulations by

providing the correlation coefficient (CC) between corti-

cal activation mapping and dipole moving patterns. It can

also be seen that the maximum amplitude criterion gives

the most consistent estimation of the source movement.

Fig. 5a shows the volume conductor, the locations of thecortical electrodes, and the trace of the moving dipole. A

radial dipole was assumed to move from the lower left

corner to the upper right corner under the grid electrodes.

Fig. 5b displays the simulated cortical potential distribu-

tions generated by the moving dipole. Figs. 5cf are the

activation mapping results on the 8 8 grid pad by using

four different criteria: maximal amplitude (C1), maximal

first derivative (C2), maximal second derivative (C3), and

maximal surface laplacian (C4), respectively. Consider-

ing that the dipole was moving from corner to corner at



7/10


FIG. 5. Computer simulation of CAM analysis by using a single moving radial dipole. a: The sphere represents the outermost surface ofthe three-sphere head volume conductor. Black circles, Cortical electrodes. Pink circles, The trace of dipole moving from lower left cornerto upper right corner. The eccentricity of the trace is 0.80. b: Cortical potential waveforms generated by the single moving radial dipole.cf: Cortical activation mapping from potential distributions using different criteria of maximal amplitude (C1), maximal first derivative (C2),maximal second derivative (C3), and maximal laplacian (C4), respectively.

a constant speed, C1 gave the most reasonable estima-

tion of the activation sequence, whereas C2, C3, and C4

generated less accurate or smeared results. Fig. 6a shows

the cortical potential waveforms generated by a tangential

dipole moving along the same route as shown in Fig. 5a.

Figs. 6be display the activation maps with C1, C2, C3,

and C4, respectively. Although the initiation location isroughly similar in these results, higher spatial resolution

FIG. 6. Computer simulation of CAM analysis using single tangential dipole moving from lower left corner to upper right corner with aneccentricity of 0.80. ae: See the caption of Fig. 5.

can beobservedin Fig.6b thanin the other results shown in

Fig. 6ce. Further examination also reveals that C1 gives a

more consistent estimation of the dipole source activities

than the other criteria.

DISCUSSION

The present study represents, to our knowledge, thefirst effort to image the cortical neuronal activation



8/10

312 Y. LAI ET AL.

TABLE 2. Correlation coefficient between dipole moving pattern and cortical activation mappings determined by different criteria

Correlation coefficient

Orientation Eccentricity Max amplitude Max 1st derivative Max 2nd derivative Max laplacian

Radial 0.60 0.9750 0.9327 0.5754 0.45730.70 0.9750 0.9387 0.6837 0.70950.80 0.9747 0.9246 0.8309 0.9159

0.85 0.9333 0.8855 0.8858 0.8448Tangential 0.60 0.9441 0.9451 0.9125 0.83540.70 0.9377 0.9474 0.9111 0.85860.80 0.9248 0.9477 0.9040 0.87710.85 0.9103 0.8047 0.7091 0.8114

sequence quantitatively during the epileptiform interictal

discharges from multichannel intracranial ECoG record-

ings. Although the approach itself is theoretically and

mathematically simple, the present study suggests that it

can be effectively used to identify the initiation and prop-

agation of epileptiform activity. By comparing the results

obtained from the CAM analysis with ECoG-recorded

seizure activities in eight pediatric patients, it was demon-

strated that the CAM analysis has successfully predicted

the IOZ in the majority of cases (seven of eight patients).

These promising results indicate that this method can po-

tentially be applied to assist in the presurgical evalua-

tion and surgical planning for patients with intractable

epilepsy.

Latency differences of paroxysmal interictal discharges

at different recording sites are usually


9/10


(EPSPs) (Ayala et al., 1973; Steriade 1998a, 1998b;

Castro-Alamncos2000). Conversely, different neurophys-

iologic synchronizing mechanisms have also been re-

ported in the literature (Taylor et al., 1984; Korn et al.,

1987; Amzica and Steriade, 2000; Kohling et al., 2001;

Traub et al., 2001). Due to the uncertainty of the mecha-

nisms of generation of both interictal and ictal discharges,thephysiologicrelationbetween IIS andseizure is not well

understood. It remains an open question whether and how

the occurrence of IISs can reflect the epileptogenic zone.

In this study, the CAM analysis successfully predicted the

seizure-onset zone in seven of eight patients but failed in

patient 6, as shown in Fig. 3. The reason for this failure is

unknown and requires additional exploration in additional

patients. It might be explained by the discrepancies in the

origin of IISs and seizure, but also could be generated by

the vertical propagation of IISs in the human brain.

The present method uses the two-dimensional cortical

potential distribution to perform the CAM analysis. This

works well with the pediatricpatients with superficial neo-

cortical epileptogenic foci, but it may generate inaccu-

rate estimation of initiation of IISs in cases in which the

IIS originates from deep cortical structures. Ulbert et al.

(2004) concluded from their experiments that the human

IISs start from cortical layer IV with powerful depolariza-

tion in the regions of the main route of ictal propagation,

which spread transversely to both supra- and infragranular

laminae. Because of this vertical propagation, the earliest

activation detected by the subdural electrodes may actu-

ally represent neuronal activity that has propagated away

from the initiation site. This misrepresentation is difficult

to avoid when using the two-dimensional potential distri-bution but might be corrected by three-dimensional map-

ping if depth-potential recording or estimationis available.

Further investigation is needed to fully address this issue.

To further evaluate the performance of different criteria

in imaging the activation sequence, computer simulations

were conducted by using simplified modelsto estimate the

cortical activation times from the potential distributions

(Table 2). The results shown in Figs. 5 and 6, together with

Table 2, reveal that the maximal-amplitude criterion gives

the mostconsistent estimationof the source-activationpat-

tern, which is also consistent with the results from clin-

ical data analysis. Note that the single moving dipole isjust a simplified implementation of generation of interictal

epileptiform activity, whose starting location is suggestive

of the initiation of IISs and apparently should be resected

to generate favorable surgical outcome. To fully evaluate

the performance by different criteria, simulations should

be performed by using more-realistic source models both

physiologically and pathologically.

Note that the quantitative volumes of the performed re-

sections are unclear from the present study because of

the lack of postoperative MRIs. Such postoperative MRIs

would provide a quantitative means of correlating the

origin of interictal epileptiform activity with the present

findings from the CAM analysis. Such quantitative anal-

ysis would be desirable in future investigations more pre-

cisely to localize the epileptogenic zones, as indicated

from the successful surgical treatments.

For some patients, it appears that the IOZs were located

at the edge of the grids (Figs. 1, 2, and 4). In this case,the possibility that the recorded epileptiform activities

originate from remote locations and propagate to cortex

underlying the grid electrodes should be excluded. Clini-

callythis can be accomplished by validating ECoG activity

against simultaneous scalp EEG recording, which would

indicate activity significantly beyond the putative focus.

The present study has shown the ability of the CAM

analysis to generate the topography of the activation se-

quence of neuronal populations. The patterns of initiation

and propagation of the IISs obtained from the CAM anal-

ysis have proved to be effective in predicting the IOZ in

the pediatric epilepsy patients, suggesting that the present

CAM method may be potentially used as an alternative to

define the epileptogenic zone.

Acknowledgment: We thank Christopher Wilke for usefuldiscussions and proofreading the manuscript. This work wassupported in part by NIH RO1 EB00178, NSF BES-0411898,and partly supported by the Biomedical Engineering Institute ofthe University of Minnesota.

REFERENCES

Alarcon G, Guy CN, Binnie CD, Walker SR, Elwes RDC, PolkeyCE. (1994) Intracerebral propagation of interictal activity in partial

epilepsy: implications for source localisation. Journal of Neurology,Neurosurgery and Psychiatry 57:43549.

Alarcon G, Garcia Seoane JJ, Binnie CD, Martin Mignel MC, Julen J,Polkey CE, Elwes RD, Ortiz Blasco JM. (1997) Origin and propaga-tion of interictal discharges in the acute electrocorticogram: impli-cations for pathophysiology and surgical treatment of temporal lobeepilepsy. Brain 120:22592282.

Amzica F, Steriade M. (2000) Neuronal and glial membrane potentialsduring sleep and paroxysmal oscillations in the neocortex. Journalof Neuroscience 20:66486665.

Asano E, Muzik O, Shah A, Juhasz C, Chugani DC, Sood S, Janisse J,Ergun EL, Ahn-Ewing J, Shen C, Gotman J, Chugani HT. (2003)Quantitative interictal subdural EEG analyses in children with neo-

cortical epilepsy. Epilepsia 44:425434.AsanoE, Muzik O,ShahA, JuhaszC, ChuganiDC, KagawaK, Benedek

K, Sood S, Gotman J, Chugani HT.(2004)Quantitative visualizationof ictal subdural EEG changes in children with neocortical focal

seizures. Clinical Neurophysiology 115:27182727.Ayala GF, Dichter M, Gumnit RJ, Matsumoto H, Spencer WA. (1973)

Genesis of epileptic interictal spikes: new knowledge of corticalfeedback systems suggests a neurophysiological explanation of briefparoxysms. Brain Research 52:117.

Babiloni F, Babiloni C, Carducci F, Cincotti F, Astolfi L, Basilisco A,Rossini PM, Ding L, Ni Y, Cheng J, Christine K, Sweeney J, He B.(2005) Assessing time-varying cortical functional connectivity withthe multimodal integration of high resolution EEG and fMRI data

by directed transfer function. NeuroImage 24:118131.Castro-Alamancos MA. (2000) Origin of synchronized oscillations in-

duced by neocortical disinhibition in vivo. Journal of Neuroscience20:91959206.

Chatrian GE, Bergamini L, Dondey M, Klass DW, Lennox-Buchtal M,PetersenI. (1974) A glossary of term most commonly used by clinical



10/10

314 Y. LAI ET AL.

electroencephalographers. Electroencephalography and ClinicalNeurophysiology 37:538548.

Dale AM,SerenoM. (1993) Improved localization of cortical activitybycombining EEGand MEG with MRI cortical surface reconstruction:a linear approach. Journal of Cognitive Neuroscience 5:162176.

Ebersole JS. (2000) Noninvasive localization of epileptogenic foci byEEG source modeling. Epilepsia 41:2433.

Emerson RG, Turner CA, Pedley TA, Walczak TS, Forgione M. (1995)

Propagation of patterns of temporal spikes. Electroencephalographyand Clinical Neurophysiology 94:338348.

He B, Wang Y, Wu D. (1999) Estimating cortical potentials from scalpEEGs in a realistically shaped inhomogeneous head model by meansof the boundary element method. IEEE Transactions on BiomedicalEngineering 46:12641268.

HeB, Lian J, Spencer KM,DienJ, DonchinE. (2001) A corticalpotentialimaging analysis of the P300 and novelty P3 components. HumanBrain Mapping 12:120130.

He B, Zhang X, Lian J, Sasaki H, Wu D, Towle VL. (2002) Boundaryelement method-based cortical potential imaging of somatosensoryevoked potentials using subjects magnetic resonance images. Neu-roImage 16:564576.

Holmes MD, Kutsy RL, Ojemann GA, Wilensky AJ, Ojemann LM.(2000)Interictal, unifocal spikes in refractory extratemporal epilepsypredict ictal origin and postsurgical outcome. Clinical Neurophysi-ology 111:18021808.

Hufnagel A, Dumpelmann M, Zentner J, Schijns O, Elger CE. (2000)Clinical relevance of quantified intracranial interictal spike ac-tivity in presurgical evaluation of epilepsy. Epilepsia 41:467478.

Kanazawa O, Blume WT, Girvin JP. (1996) Significance of spikes attemporal lobe electrocorticography. Epilepsia 37:5055.

Kohling R, Gladwell SJ, Bracci E, Vreugdenhil M, Jefferys JG. (2001)Prolonged epileptiform bursting induced by 0-Mg(2+) in rat hip-

pocampal slices depends on gap junctional coupling. Neuroscience105:579587.

Korn SJ, Giacchino JL, Chamberlin NL, Dingledine R. (1987) Epilepti-form burst activity induced by potassium in the hippocampus and itsregulation by GABA-mediated inhibition. Journal of Neurophysiol-ogy 57:325340.

Krings T, Chiappa KH, Cuffin BN, Buchbinder BR, Cosgrove GR.(1998) Accuracy of electroencephalographic dipole localization of

epileptiform activities associated with focal brain lesions. Annals ofNeurology 44:7686.

Lantz G, Spinelli L, Seeck M, de Peralta Menendez RG, SottasCC, Michel CM. (2003) Propagation of interictal epileptiform ac-tivity can lead to erroneous source localizations: a 128-channel

EEG mapping study. Journal of Clinical Neurophysiology 20:311319.

Leal AJ,Passao V, Calado E, Vieira JP,SilvaCunha JP.(2002)Interictalspike EEG source analysis in hypothalamic hamartoma epilepsy.Clinical Neurophysiology 113:19611969.

Lee SA, Spencer DD, Spencer SS. (2000)Intracranial EEG seizure-onsetpatterns in neocortical epilepsy. Epilepsia 41:297307.

Lemieux L, Krakow K, Fish DR. (2001) Comparison of spike-triggeredfunctionalMRI BOLDactivationand EEG dipolemodel localization.Neuroimage 14:10971104.

Merlet I, Garcia-Larrea L, Gregoire MC, Lavenne F, MauguiereF. (1996) Source propagation of interictal spikes in temporallobe epilepsy: correlations between spike dipole modelling and[18F]fluorodeoxyglucose PET data. Brain 119:377392.

Merlet I, Gotman J. (1999) Reliability of dipole models of epileptic

spikes. Clinical Neurophysiology 110:10131028.Penfield W, Jasper H. (1954)Epilepsy and the functionalanatomy of the

human brain. J & A. Churchill, London.Shamoto H, Nakajima T, Nakasato N, Iwasaki M, Shirane R, Itoh

M, Yoshimoto T. (2002) Mesial temporal lobe epilepsy with lat-eral temporal lobe abnormalities in magnetoencephalography andglucose metabolism. Journal of Clinical Neuroscience 9:192194.

Steriade M and Contreras D. (1998) Spike-wave complexes and fastcomponents of cortically generated seizures, I: role of neocortex andthalamus. Journal of Neurophysiology 80:14391455.

Steriade M, Amzica F, Neckelmann D, Timofeev I. (1998) Spike-wavecomplexes and fast components of cortically generated seizures,

II: extra- and intracellular patterns. Journal of Neurophysiology80:14561479.

Sutherling WW, Barth DS. (1989) Neocortical propagation in temporallobe spike foci on magnetoencephalography and electroencephalog-raphy. Annals of Neurology 25:373381.

Taylor C, Dudek F. (1984) Excitation of hippocampal pyramidal cellsby an electrical field effect. Journal of Neurophysiology 52:126142.

Traub RD, Whittington MA, Buhl EH, LeBeau FE, Bibbig A, Boyd S,Cross H, Baldeweg T. (2001) A possible role for gap junctions ingeneration of very fast EEG oscillations preceding the onset of, andperhaps initiating, seizures. Epilepsia 42:153170.

Tsau Y, Guan L, Wu JY. (1999) Epileptiform activity can be initiatedin various neocortical layers: an optical imaging study. Journal ofNeurophysiology 82:19651973.

Tsai ML, Chatrian GE, Pauri F, Temkin NR, Holubkov AL, Shaw CM,Ojemann GA. (1993) Electrocorticography in patients with medi-cally intractable temporal lobe seizures, I: quantification of epilepti-form discharges prior to resective surgery. Electroencephalographyand Clinical Neurophysiology 87:1024.

Tsai ML, Chatrian GE, Holubkov AL, Temkin NR, Shaw CM, Oje-mann GA. (1993) Electrocorticography in patients with medicallyintractable temporal lobe seizures, II: quantification of epileptiformdischarges following successive stages of resective surgery. Elec-troencephalography and Clinical Neurophysiology 87:2537.

Ulbert I, Heit G, Madsen J, Karmos G, Halgren E. (2004) Laminar anal-ysis of human neocortical interictal spike generation and propaga-tion: current source density and multiunit analysis in vivo. Epilepsia45(suppl 4):4856.

Zhang X, van Drongelen W, Hecox K, Towle VL, Frim DM, McGeeA, He B. (2003) High resolution EEG: cortical potential imaging ofinterictal spikes. Clinical Neurophysiology 114:19631973.


Documents

Cortical Activation Mapping of Epileptiform Activity Derived From Interictal ECoG Spikes