- Email: [email protected]
Comparison of source localization of interictal epileptic spike potentials in patients estimated by the dipole tracing method with the focus directly recorded by the depth electrodes
Neuroscience Letters 304 (2001) 1±4
www.elsevier.com/locate/neulet
Comparison of source localization of interictal epileptic spike potentials in patients estimated by the dipole tracing method with the focus directly recorded by the depth electrodes Ikuo Homma a,*, Yuri Masaoka a, Kenichi Hirasawa b, Fumitaka Yamane b, Tomokatsu Hori b, Yoshiwo Okamoto c a
2nd Department of Physiology, Showa University School of Medicine, Hatanodai 1±5±8, Shinagawa-ku, Tokyo 142±8555, Japan b Department of Neurosurgery, Tokyo Women's Medical University, Kawadachou, Shinjuku-ku, Tokyo, Japan c Department of Electric Engineering, Chiba Institute of Technology, Tudanuma 2±17±1, Narashino 275±0016, Japan Received 20 January 2001; received in revised form 7 March 2001; accepted 8 March 2001
Abstract The purpose of the study was to investigate the accuracy of location of equivalent current dipoles estimated by the dipole tracing method (DT) utilizing a realistic 3-shell (scalp±skull±brain) head model (SSB±DT). Three patients with intractable complex partial seizures, diagnosed as having typical temporal seizures were investigated. We recorded the interictal spike potentials with surface electrodes (International 10/20 system) and with intracerebral depth electrodes simultaneously. We compared the location of dipoles of the spikes estimated by the SSB±DT with the focus of the spikes determined by the recording from the depth electrodes. We found that the location of the dipoles estimated by SSB±DT corresponded to the location of the depth electrodes, which could record the epileptic spikes. This ®nding proved that SSB±DT is reliable and valid for estimating neural activity in deep locations such as the limbic system. q 2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Dipole tracing method; Scalp±skull±brain head model; Epileptic spike; Temporal seizures; Electoencephalograms; Intracerebral depth electrodes; Amygdala
Electroencephalograms (EEG) are the time courses of electric potentials recorded by the surface electrodes on the scalp apart from their sources in the brain. The source distribution generated by neural excitations is approximated by equivalent current dipoles. In order to estimate these dipoles, the dipole tracing method has been introduced and under development [3,6,11,12]. Since the potential distributions on the scalp surface are smeared out by a poorly conductive skull, a multiple-shell head model is indispensable to take this smearing effect into account. Concentric sphere models have been used because of their simplicity [2,4]. However, in order to estimate the dipoles accurately, it is necessary to make use of a realistic multiple-shell head model. The dipole tracing method (DT) utilizing a realistic 3-shell (scalp±skull±brain) head model has been called SSB±DT [7]. Recently, Flink et al. [5] compared source locations of interictal spikes in patients * Corresponding author. Tel.: 181-3-3784-8112; fax: 181-337840200. E-mail address: [email protected] (I. Homma).
with temporal lobe epilepsy estimated from scalp surface potentials by means of SSB±DT with those estimated from subdural potentials by means of the single-shell brain model, and showed that dipoles estimated in both ways coincide well with each other. However, it could not be con®rmed whether these dipoles correspond to the same source because the scalp surface potentials and the subdural potentials were recorded separately. A method to determine the location of the epileptic focus in patients with epileptic seizures is important for preoperative evaluation. Intracerebral depth electrode placement is safer and more accurate to diagnose partial epilepsies in which the focus is presumed to originate in the limbic system [9]. In the present study, we recorded the interictal spike potentials with routine scalp EEG electrodes and with intracerebral depth electrodes simultaneously; in addition, we compared the location of the interictal epileptic spikes estimated by SSB±DT with the location of the intracerebral depth electrodes which could record the interictal epileptic spikes.
0304-3940/01/$ - see front matter q 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S03 04 - 394 0( 0 1) 01 74 6- 3
2
I. Homma et al. / Neuroscience Letters 304 (2001) 1±4
Fig. 1. Left panel: simultaneous recordings of the interictal spike potentials with 19 surface electrodes (FP1 to Cz) and with depth electrodes (left amygdala electrodes; 1±6, left hippocampus electrodes; 1±6). The number of depth electrodes counted from the center of the CT image from subject A is indicated in the right panel. Right panel: CT image with depth electrode from subject A (upper right). The results of SSB±DT with axial, coronal and sagittal views during the time between the dotted lines are indicated in the left panel (bottom). The dipoles were concentrated in the left amygdala, especially in the left amygdala depth electrodes 1 and 2. Upper left indicates an enlargement of the potentials having a polarity between the left amydala depth electrodes 1 and 2 (upper left).
Fig. 2. Upper panel: CT images with depth electrodes from three subjects. Lower panel: three-dimensional images of result of SSB±DT in each subject. The dipoles were concentrated in the left amygdala, corresponding with the location of the left amygdala depth electrodes, which could record interictal spikes.
I. Homma et al. / Neuroscience Letters 304 (2001) 1±4
The present study was approved by the Ethical Committee of Showa University School of Medicine and Tokyo Women's Medical University. Three patients (subject A: male, aged 19; subject B: female, aged 26; subject C: male, aged 18) with intractable complex partial seizures were investigated. All patients had experienced epileptic seizures for more than 10 years and had been diagnosed as having typical temporal seizures. There were no structural ®ndings in the amygdala on the MRI. X-ray computed tomography (thickness of 5 mm) was obtained from each patient with marks of reference points (nasion, inion, bilateral premearus points and vertex) several days before EEG recordings. The interictal spike potentials were simultaneously recorded by intracerebral depth electrodes and nineteen surface EEG electrodes placed on the scalp according to the international 10/20 system with reference electrodes on both earlobes. Locations of 19 surface electrodes and locations of reference points (nasion, inion, bilateral pre-meatus points and vertex) were measured with a three-dimensional digitizer (Freepoint 3D, Science Accessories). The potentials recorded from the electrodes were ampli®ed and ®ltered by an EEG recorder (bandpass: 0.53±120 HZ, EEG-1100, Nihon Kohden) and stored on an EEG analyzer (DAE-2100, Nihon Kohden). CT images in DICOM3 format were used for making each subject's own 3-layer head model for dipole-tracing. The EEG, location of reference points and electrodes, head shape of the scalp, skull and brain were used for SSB±DT by Brain Space Navigator (BS-navi, Japan Graphics). At any given moment, the distribution of the potential recorded from the scalp surface was generated from one or two equivalent current dipoles. Equivalent current dipoles were estimated every 2 ms. The accuracy of estimation was expressed by means of `dipolarity' [7]. In this study dipolarity larger than 98% was considered as signi®cant [8]. Two intracerebral depth electrodes were bilaterally placed in the brain through a craniotomy. Depth electrodes were composed of six contact points; four electrodes from the tip placed in the amygdala had a 5 mm interval and the other two had a 10 mm interval (Unique Medical Co., LTD.). Two other depth electrodes were also placed in the brain, and the tip was placed in the head of the hippocampus (hippocampus electrode). Interictal spike potentials were simultaneously recorded with surface electrodes and depth electrodes. A typical example of potential recordings taken from subject A is shown in the left panel of Fig. 1. Interictal spike and wave potentials were clearly seen in the left depth electrodes, in particular, in the left amygdala electrodes. The upper left in the right panel of Fig. 1 shows that the potentials observed in the left amygdala depth electrode 1 and 2 were reversed with polarity between the potentials. Electrode positions in the brain were assumed by the CT images as shown in the upper right panel of Fig. 1. Small but corresponding potentials with those in the depth electrodes were seen in
3
the surface electrodes. Dipole tracing was performed in time between the red vertical dotted lines which is shown in the left panel of potential recordings in Fig. 1. With regard to estimated dipoles, dipolarity was more than 98%, as shown in the right bottom panels in Fig. 1 from the axial, coronal and sagittal views. The other two patients showed interictal spikes in the left amygdala, and their depth electrode positions are indicated in CT images in the upper right panel of Fig. 2. The lower panel of Fig. 2 shows the estimated dipoles in each head model. Dipoles estimated from the interictal spikes in three subjects were located in the left amygdala. The purpose of the study was to investigate the accuracy of location of equivalent current dipoles estimated by SSB± DT. We compared the location of the dipoles of interictal spikes estimated by SSB±DT with the location of intracerebral depth electrodes, which could record the interictal epileptic spikes. All three epileptic patients showed the focus of the spikes in the left amygdala and SSB±DT estimated dipoles in the same area. Flink et al. [5] compared the location of dipoles estimated by SSB±DT from the surface electrodes with the location of dipoles by single-shell DT from the subdural electrodes in patients with temporal lobe epilepsy. Flink's study showed that the location of dipoles estimated in both ways were in the same area; however, it was impossible to estimate the same spike with simultaneous recordings by the surface electrodes and the subdural electrodes. The subdural electrodes could record only the potentials from the source generator close to the electrodes [1]. Flink et al. [5] reported that the dipole location in the deep area could not be estimated by the subdural electrodes. In this study, the spikes were simultaneously recorded with the surface electrodes and with the depth electrodes. We found that the location of the dipoles from the surface electrodes by SSB±DT corresponded to the location of the depth electrodes, which could record the epileptic spikes. This ®nding proved that SSB±DT is reliable and valid for estimating neural activity in deep locations such as the limbic system, and this non-invasive method could be used not only for the determination of epileptic foci but also for psychological [10] and clinical studies. This research was supported by the Arata Kanamaru Fund. We thank Suzanne Knowlton for preparation of the manuscript. [1] Blom, S., Flink, R., Hetta, J., Hilton-Brown, P., Hoffstedt, C., Osterman, P.O. and SpaÂnnare, B., Interictal and ictal activity recorded with subdual electrodes during preoperative evaluation for surgical treatment of epilepsy, J. Epilepsy, 2 (1989) 9±20. [2] Cohen, D., Cuf®n, B.N., Tunokuchi, K., Maniewski, R., Purcell, C., Cosgrove, G.R., Ives, J., Kennedy, J.G. and Schomer, D.L., MEG versus EEG localization test using implanted sources in the human brain, Ann. Neurol., 28 (1990) 811±817.
4
I. Homma et al. / Neuroscience Letters 304 (2001) 1±4 [3] Cuf®n, B.N., A comparison of moving dipole inverse solutions using EEGs and MEGs, IEEE Trans. Biomed. Eng., BME-32 (1985) 905±910. [4] Cuf®n, B.N., Cohen, D., Yunokuchi, K., Maniewski, R., Purcell, C., Cosgrove, G.R., Ives, J., Kennedy, J. and Schomer, D., Tests of EEG localization accuracy using implanted source in the human brain, Ann. Neurol., 29 (1991) 132±138. [5] Flink, R., Homma, S., Kanamaru, A., Miyamoto, K. and Okamoto, Y., Source localization of interictal epileptiform spike potentials estimated with a dipole tracing method using surface and subdual EEG recordings, Clin. Neurophysiol. Suppl., 53 (2000) 1±11. [6] He, B., Musha, T., Okamoto, Y., Homma, S., Nakajima, Y. and Sato, T., Electric dipole tracing in the brain by means of the boundary element method and its accuracy, IEEE Trans. Biomed. Eng., 6 (34) (1987) 406±414. [7] Homma, S., Musha, T., Nakajima, Y., Okamoto, Y., Blom, S., Flink, R., Hagbarth, K.-E. and MostroÈm, U., Location of electric current sources in the human brain estimated by the dipole tracing method of the scalp±skull±brain (SSB) head
[8]
[9]
[10] [11]
[12]
model, Electroenceph. Clin. Neurophysiol., 91 (1994) 374± 382. Kanamaru, A., Homma, I. and Hara, T., Movement related cortical sources for elbow ¯exion in patients with brachial plexus injury after intercostal-musculocutaneous nerve crossing, Neurosci. Lett., 274 (1999) 203±206. Levesque, M.F., Zhang, J.X., Wilson, C.L., Behnke, E.J., Harper, R.M., Lufkin, R.B., Engel Jr., J. and Crandall, P.H., Stereotactic investigation of limbic epilepsy using a multimodel image analysis system, J. Neurosurg., 73 (1990) 792± 797. Masaoka, Y. and Homma, I., The source generator of respiratory-related anxiety potentials in the human brain, Neurosci. Lett., 283 (2000) 21±24. Okamoto, Y., Teramachi, Y., Musha, T., Tsunakawa, H. and Harumi, K., Limitation of the inverse problem in body surface potential mapping, IEEE Trans. Biomed. Eng., BME-30 (1983) 754±794. Smith, D.B., Sidman, R.D., Flanigin, H., Henke, J. and Labiner, D., A reliable method for localizing deep intracranial sources of the EEG, Neurology, 35 (1985) 1702±1707.
www.elsevier.com/locate/neulet
Comparison of source localization of interictal epileptic spike potentials in patients estimated by the dipole tracing method with the focus directly recorded by the depth electrodes Ikuo Homma a,*, Yuri Masaoka a, Kenichi Hirasawa b, Fumitaka Yamane b, Tomokatsu Hori b, Yoshiwo Okamoto c a
2nd Department of Physiology, Showa University School of Medicine, Hatanodai 1±5±8, Shinagawa-ku, Tokyo 142±8555, Japan b Department of Neurosurgery, Tokyo Women's Medical University, Kawadachou, Shinjuku-ku, Tokyo, Japan c Department of Electric Engineering, Chiba Institute of Technology, Tudanuma 2±17±1, Narashino 275±0016, Japan Received 20 January 2001; received in revised form 7 March 2001; accepted 8 March 2001
Abstract The purpose of the study was to investigate the accuracy of location of equivalent current dipoles estimated by the dipole tracing method (DT) utilizing a realistic 3-shell (scalp±skull±brain) head model (SSB±DT). Three patients with intractable complex partial seizures, diagnosed as having typical temporal seizures were investigated. We recorded the interictal spike potentials with surface electrodes (International 10/20 system) and with intracerebral depth electrodes simultaneously. We compared the location of dipoles of the spikes estimated by the SSB±DT with the focus of the spikes determined by the recording from the depth electrodes. We found that the location of the dipoles estimated by SSB±DT corresponded to the location of the depth electrodes, which could record the epileptic spikes. This ®nding proved that SSB±DT is reliable and valid for estimating neural activity in deep locations such as the limbic system. q 2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Dipole tracing method; Scalp±skull±brain head model; Epileptic spike; Temporal seizures; Electoencephalograms; Intracerebral depth electrodes; Amygdala
Electroencephalograms (EEG) are the time courses of electric potentials recorded by the surface electrodes on the scalp apart from their sources in the brain. The source distribution generated by neural excitations is approximated by equivalent current dipoles. In order to estimate these dipoles, the dipole tracing method has been introduced and under development [3,6,11,12]. Since the potential distributions on the scalp surface are smeared out by a poorly conductive skull, a multiple-shell head model is indispensable to take this smearing effect into account. Concentric sphere models have been used because of their simplicity [2,4]. However, in order to estimate the dipoles accurately, it is necessary to make use of a realistic multiple-shell head model. The dipole tracing method (DT) utilizing a realistic 3-shell (scalp±skull±brain) head model has been called SSB±DT [7]. Recently, Flink et al. [5] compared source locations of interictal spikes in patients * Corresponding author. Tel.: 181-3-3784-8112; fax: 181-337840200. E-mail address: [email protected] (I. Homma).
with temporal lobe epilepsy estimated from scalp surface potentials by means of SSB±DT with those estimated from subdural potentials by means of the single-shell brain model, and showed that dipoles estimated in both ways coincide well with each other. However, it could not be con®rmed whether these dipoles correspond to the same source because the scalp surface potentials and the subdural potentials were recorded separately. A method to determine the location of the epileptic focus in patients with epileptic seizures is important for preoperative evaluation. Intracerebral depth electrode placement is safer and more accurate to diagnose partial epilepsies in which the focus is presumed to originate in the limbic system [9]. In the present study, we recorded the interictal spike potentials with routine scalp EEG electrodes and with intracerebral depth electrodes simultaneously; in addition, we compared the location of the interictal epileptic spikes estimated by SSB±DT with the location of the intracerebral depth electrodes which could record the interictal epileptic spikes.
0304-3940/01/$ - see front matter q 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S03 04 - 394 0( 0 1) 01 74 6- 3
2
I. Homma et al. / Neuroscience Letters 304 (2001) 1±4
Fig. 1. Left panel: simultaneous recordings of the interictal spike potentials with 19 surface electrodes (FP1 to Cz) and with depth electrodes (left amygdala electrodes; 1±6, left hippocampus electrodes; 1±6). The number of depth electrodes counted from the center of the CT image from subject A is indicated in the right panel. Right panel: CT image with depth electrode from subject A (upper right). The results of SSB±DT with axial, coronal and sagittal views during the time between the dotted lines are indicated in the left panel (bottom). The dipoles were concentrated in the left amygdala, especially in the left amygdala depth electrodes 1 and 2. Upper left indicates an enlargement of the potentials having a polarity between the left amydala depth electrodes 1 and 2 (upper left).
Fig. 2. Upper panel: CT images with depth electrodes from three subjects. Lower panel: three-dimensional images of result of SSB±DT in each subject. The dipoles were concentrated in the left amygdala, corresponding with the location of the left amygdala depth electrodes, which could record interictal spikes.
I. Homma et al. / Neuroscience Letters 304 (2001) 1±4
The present study was approved by the Ethical Committee of Showa University School of Medicine and Tokyo Women's Medical University. Three patients (subject A: male, aged 19; subject B: female, aged 26; subject C: male, aged 18) with intractable complex partial seizures were investigated. All patients had experienced epileptic seizures for more than 10 years and had been diagnosed as having typical temporal seizures. There were no structural ®ndings in the amygdala on the MRI. X-ray computed tomography (thickness of 5 mm) was obtained from each patient with marks of reference points (nasion, inion, bilateral premearus points and vertex) several days before EEG recordings. The interictal spike potentials were simultaneously recorded by intracerebral depth electrodes and nineteen surface EEG electrodes placed on the scalp according to the international 10/20 system with reference electrodes on both earlobes. Locations of 19 surface electrodes and locations of reference points (nasion, inion, bilateral pre-meatus points and vertex) were measured with a three-dimensional digitizer (Freepoint 3D, Science Accessories). The potentials recorded from the electrodes were ampli®ed and ®ltered by an EEG recorder (bandpass: 0.53±120 HZ, EEG-1100, Nihon Kohden) and stored on an EEG analyzer (DAE-2100, Nihon Kohden). CT images in DICOM3 format were used for making each subject's own 3-layer head model for dipole-tracing. The EEG, location of reference points and electrodes, head shape of the scalp, skull and brain were used for SSB±DT by Brain Space Navigator (BS-navi, Japan Graphics). At any given moment, the distribution of the potential recorded from the scalp surface was generated from one or two equivalent current dipoles. Equivalent current dipoles were estimated every 2 ms. The accuracy of estimation was expressed by means of `dipolarity' [7]. In this study dipolarity larger than 98% was considered as signi®cant [8]. Two intracerebral depth electrodes were bilaterally placed in the brain through a craniotomy. Depth electrodes were composed of six contact points; four electrodes from the tip placed in the amygdala had a 5 mm interval and the other two had a 10 mm interval (Unique Medical Co., LTD.). Two other depth electrodes were also placed in the brain, and the tip was placed in the head of the hippocampus (hippocampus electrode). Interictal spike potentials were simultaneously recorded with surface electrodes and depth electrodes. A typical example of potential recordings taken from subject A is shown in the left panel of Fig. 1. Interictal spike and wave potentials were clearly seen in the left depth electrodes, in particular, in the left amygdala electrodes. The upper left in the right panel of Fig. 1 shows that the potentials observed in the left amygdala depth electrode 1 and 2 were reversed with polarity between the potentials. Electrode positions in the brain were assumed by the CT images as shown in the upper right panel of Fig. 1. Small but corresponding potentials with those in the depth electrodes were seen in
3
the surface electrodes. Dipole tracing was performed in time between the red vertical dotted lines which is shown in the left panel of potential recordings in Fig. 1. With regard to estimated dipoles, dipolarity was more than 98%, as shown in the right bottom panels in Fig. 1 from the axial, coronal and sagittal views. The other two patients showed interictal spikes in the left amygdala, and their depth electrode positions are indicated in CT images in the upper right panel of Fig. 2. The lower panel of Fig. 2 shows the estimated dipoles in each head model. Dipoles estimated from the interictal spikes in three subjects were located in the left amygdala. The purpose of the study was to investigate the accuracy of location of equivalent current dipoles estimated by SSB± DT. We compared the location of the dipoles of interictal spikes estimated by SSB±DT with the location of intracerebral depth electrodes, which could record the interictal epileptic spikes. All three epileptic patients showed the focus of the spikes in the left amygdala and SSB±DT estimated dipoles in the same area. Flink et al. [5] compared the location of dipoles estimated by SSB±DT from the surface electrodes with the location of dipoles by single-shell DT from the subdural electrodes in patients with temporal lobe epilepsy. Flink's study showed that the location of dipoles estimated in both ways were in the same area; however, it was impossible to estimate the same spike with simultaneous recordings by the surface electrodes and the subdural electrodes. The subdural electrodes could record only the potentials from the source generator close to the electrodes [1]. Flink et al. [5] reported that the dipole location in the deep area could not be estimated by the subdural electrodes. In this study, the spikes were simultaneously recorded with the surface electrodes and with the depth electrodes. We found that the location of the dipoles from the surface electrodes by SSB±DT corresponded to the location of the depth electrodes, which could record the epileptic spikes. This ®nding proved that SSB±DT is reliable and valid for estimating neural activity in deep locations such as the limbic system, and this non-invasive method could be used not only for the determination of epileptic foci but also for psychological [10] and clinical studies. This research was supported by the Arata Kanamaru Fund. We thank Suzanne Knowlton for preparation of the manuscript. [1] Blom, S., Flink, R., Hetta, J., Hilton-Brown, P., Hoffstedt, C., Osterman, P.O. and SpaÂnnare, B., Interictal and ictal activity recorded with subdual electrodes during preoperative evaluation for surgical treatment of epilepsy, J. Epilepsy, 2 (1989) 9±20. [2] Cohen, D., Cuf®n, B.N., Tunokuchi, K., Maniewski, R., Purcell, C., Cosgrove, G.R., Ives, J., Kennedy, J.G. and Schomer, D.L., MEG versus EEG localization test using implanted sources in the human brain, Ann. Neurol., 28 (1990) 811±817.
4
I. Homma et al. / Neuroscience Letters 304 (2001) 1±4 [3] Cuf®n, B.N., A comparison of moving dipole inverse solutions using EEGs and MEGs, IEEE Trans. Biomed. Eng., BME-32 (1985) 905±910. [4] Cuf®n, B.N., Cohen, D., Yunokuchi, K., Maniewski, R., Purcell, C., Cosgrove, G.R., Ives, J., Kennedy, J. and Schomer, D., Tests of EEG localization accuracy using implanted source in the human brain, Ann. Neurol., 29 (1991) 132±138. [5] Flink, R., Homma, S., Kanamaru, A., Miyamoto, K. and Okamoto, Y., Source localization of interictal epileptiform spike potentials estimated with a dipole tracing method using surface and subdual EEG recordings, Clin. Neurophysiol. Suppl., 53 (2000) 1±11. [6] He, B., Musha, T., Okamoto, Y., Homma, S., Nakajima, Y. and Sato, T., Electric dipole tracing in the brain by means of the boundary element method and its accuracy, IEEE Trans. Biomed. Eng., 6 (34) (1987) 406±414. [7] Homma, S., Musha, T., Nakajima, Y., Okamoto, Y., Blom, S., Flink, R., Hagbarth, K.-E. and MostroÈm, U., Location of electric current sources in the human brain estimated by the dipole tracing method of the scalp±skull±brain (SSB) head
[8]
[9]
[10] [11]
[12]
model, Electroenceph. Clin. Neurophysiol., 91 (1994) 374± 382. Kanamaru, A., Homma, I. and Hara, T., Movement related cortical sources for elbow ¯exion in patients with brachial plexus injury after intercostal-musculocutaneous nerve crossing, Neurosci. Lett., 274 (1999) 203±206. Levesque, M.F., Zhang, J.X., Wilson, C.L., Behnke, E.J., Harper, R.M., Lufkin, R.B., Engel Jr., J. and Crandall, P.H., Stereotactic investigation of limbic epilepsy using a multimodel image analysis system, J. Neurosurg., 73 (1990) 792± 797. Masaoka, Y. and Homma, I., The source generator of respiratory-related anxiety potentials in the human brain, Neurosci. Lett., 283 (2000) 21±24. Okamoto, Y., Teramachi, Y., Musha, T., Tsunakawa, H. and Harumi, K., Limitation of the inverse problem in body surface potential mapping, IEEE Trans. Biomed. Eng., BME-30 (1983) 754±794. Smith, D.B., Sidman, R.D., Flanigin, H., Henke, J. and Labiner, D., A reliable method for localizing deep intracranial sources of the EEG, Neurology, 35 (1985) 1702±1707.