Hidden Markov modelling of spike propagation from interictal MEG data.

For patients with partial epilepsy, automatic spike detection techniques applied to interictal MEG data often discover several potentially epileptogenic brain regions. An important determination in treatment planning is which of these detected regions are most likely to be the primary sources of epileptogenic activity. Analysis of the patterns of propagation activity between the detected regions may allow for detection of these primary epileptic foci. We describe the use of hidden Markov models (HMM) for estimation of the propagation patterns between several spiking regions from interictal MEG data. Analysis of the estimated transition probability matrix allows us to make inferences regarding the propagation pattern of the abnormal activity and determine the most likely region of its origin. The proposed HMM paradigm allows for a simple incorporation of the spike detector specificity and sensitivity characteristics. We develop bounds on performance for the case of perfect detection. We also apply the technique to simulated data sets in order to study the robustness of the method to the non-ideal specificity-sensitivity characteristics of the event detectors and compare results with the lower bounds. Our study demonstrates robustness of the proposed technique to event detection errors. We conclude with an example of the application of this method to a single patient.

Language-induced seizure. A case of cortical origin.

After anastomosis of the superficial temporal to the middle cerebral artery, partial motor seizures referable to the left inferior motor strip and Broca's area developed in a 48-year-old man. These seizures were readily induced by attempts to speak, to read silently or aloud, and to write. The seizures were not induced by simple movements of the ocular, facial, jaw, throat, and extremity muscles, primary sensory stimuli, or higher intellectual function such as calculation. The patient had nonfluent dysphasia, but he had no comprehension disorder. A clear focal onset of electrographic seizures was recorded maximally at the left temporal region on reading aloud and at the left central area on writing. The cerebral evoked potentials were mildly abnormal on the affected hemisphere. This case shows cortical onset of language-induced seizure.

Cortical sensory representation of the human hand: size of finger regions and nonoverlapping digit somatotopy.

Findings differ on cortical representation of fingers between human and animal studies, and on digit somatotopy among human studies. To resolve these differences, we mapped cortical sensory representation of each of the five digits and of median and ulnar nerves in three patients, using focal peripheral electrical shock stimuli. We compared locations and sizes of cortical regions among digits and nerves, using the model of a current dipole in a sphere applied to electrocorticography from subdural grids. Cortical representation was larger for the index finger than for the little finger and for the middle finger than for the ring finger, which are similar to findings in the monkey but different from Penfield's classic sensory homunculus. The thumb was larger than the middle finger, as in the homunculus. There was nonoverlapping somatotopy of all digits in each patient. These findings demonstrate a previously unrecognized similarity of cortical sensory organization of the fingers between humans and other primates.

Somatosensory evoked fields and potentials following tibial nerve stimulation.

We studied evoked magnetic fields and electrical potentials following stimulation of the tibial nerve in a group of 24 normal subjects. Both magnetic and electrical recordings demonstrated a series of oscillatory patterns consisting of four peaks (two positive and two negative) occurring between 40 and 100 msec. Magnetic field source localization of all four peaks using a dipole-in-a-sphere model indicated that all four peaks emanated from the same cortical surface located within the longitudinal fissure, an area typically associated with somatosensory function.

Seizures induced by playing music.

A 67-year-old organist and minister with diabetes mellitus had stereotyped focal seizures of the left lower face, jaw, and neck. Attacks occurred spontaneously or were induced when he played a specific hymn on the organ. The seizures were not induced by reading, singing, hearing, or playing the hymn silently. The patient had interictal weakness of the left lower face and left side of the tongue. Focal seizures were recorded on an electroencephalogram (EEG) at the right temporofrontal area. This patient illustrates partial seizures induced by playing music.

The magnetic field of epileptic spikes agrees with intracranial localizations in complex partial epilepsy.

The magnetoencephalogram (MEG) and electroencephalogram (EEG) were measured during interictal epileptic spikes in nine patients with complex partial seizures. The MEG localization estimates were compared with localizations by intraoperative cortical electrodes, subdural electrodes, stereotaxic depth electrodes, anatomic imaging, postoperative pathologic analysis, and postoperative follow-up. In all patients, MEG localization estimates were in the same lobe as the epileptic focus determined by invasive methods and EEG. In two patients, it was possible to quantify precisely the accuracy of MEG localization by mapping a spike focus that was visually indistinguishable on MEG and cortical recordings. In both patients, MEG localization was approximately 12 mm from the center of the cortical spike focus on intracranial recordings. In eight patients, MEG showed tangential dipolar field patterns on the spontaneous record, but EEG did not. In one patient, a cortical epileptic discharge was detected only on MEG for some discharges and only on EEG for other discharges. The MEG did not detect deep spikes with present levels of environmental noise.

The magnetic and electric fields agree with intracranial localizations of somatosensory cortex.

We measured the magnetoencephalogram (MEG), electroencephalogram (EEG), and electrocorticogram (ECoG) after stimulation of contralateral median nerve in four patients with partial epilepsy evaluated for surgery. Quantitative localization estimates from equivalent source modeling were compared with locations of central fissure in hand sensorimotor area determined by cortical stimulations, intraoperative photographs, and examination after excision in frontal lobe. We also measured MEG and EEG in nine control subjects. MEG and EEG localizations were within 2.5 cm of the estimated location of central fissure in all 13 subjects. In the three patients who had complete mapping of all three fields, the average distance of localizations from central fissure was approximately 4 mm in both MEG and EEG, 3 mm in ECoG, and 3 mm in combined MEG and EEG. MEG was simpler than EEG, which was simpler than ECoG. MEG resolved ambiguities in both EEG and ECoG. The combination of the three fields added information about the spatiotemporal activity of somatosensory cortex. Localization of central fissure was essential to surgical treatment.

Focal functional anatomy of dorsolateral frontocentral seizures.

We compared 6 patients with dorsolateral frontocentral seizures to 7 patients with temporal lobe seizures. We determined general seizure location by structural lesions in 7 patients, bilateral depth electrodes in 4, and EEG and semiology in 2. We then mapped seizure cortex and essential cortex using chronic ECoG arrays. Two ECoG patterns were similar in frontal and temporal seizures. Focal patterns were near lesions and resections. Regional patterns were distant from lesions but not associated with worse surgical outcome. "Dipolar" seizure patterns occurred in one-half of frontal patients with maps like somatosensory evoked responses, consistent with focal seizure anatomy and involvement of sensorimotor cortex. Dipole location estimates were near centers of seizure cortex determined by lesions, semiology, and outcome. Six temporal patients had focal excisions that gave significant seizure reduction in all. All frontocentral patients had focal excisions that significantly reduced seizures except in 1 patient with progressive disease. We conclude that dorsolateral frontocentral seizures have focal functional anatomy that can be predicted by ictal ECoG.

Presurgical evaluation for partial epilepsy: relative contributions of chronic depth-electrode recordings versus FDG-PET and scalp-sphenoidal ictal EEG.

One hundred fifty-three patients with medically refractory partial epilepsy underwent chronic stereotactic depth-electrode EEG (SEEG) evaluations after being studied by positron emission tomography (PET) with 18F-fluorodeoxyglucose (FDG) and scalp-sphenoidal EEG telemetry. We carried out retrospective standardized reviews of local cerebral metabolism and scalp-sphenoidal ictal onsets to determine when SEEG recordings revealed additional useful information. FDG-PET localization was misleading in only 3 patients with temporal lobe SEEG ictal onsets for whom extratemporal or contralateral hypometabolism could be attributed to obvious nonepileptic structural defects. Two patients with predominantly temporal hypometabolism may have had frontal epileptogenic regions, but ultimate localization remains uncertain. Scalp-sphenoidal ictal onsets were misleading in 5 patients. For 37 patients with congruent focal scalp-sphenoidal ictal onsets and temporal hypometabolic zones, SEEG recordings never demonstrated extratemporal or contralateral epileptogenic regions; however, 3 of these patients had nondiagnostic SEEG evaluations. The results of subsequent subdural grid recordings indicated that at least 1 of these patients may have been denied beneficial surgery as a result of an equivocal SEEG evaluation. Weighing risks and benefits, it is concluded that anterior temporal lobectomy is justified without chronic intracranial recording when specific criteria for focal scalp-sphenoidal ictal EEG onsets are met, localized hypometabolism predominantly involves the same temporal lobe, and no other conflicting information has been obtained from additional tests of focal functional deficit, structural imaging, or seizure semiology.

Investigation of multiple simultaneously active brain sources in the electroencephalogram.

We present a method of investigating multiple simultaneously active brain sources that overlap both in space and time in the scalp electroencephalogram (EEG). In order to identify the contributions of the individual brain sources to measured potentials, we applied principal component analysis and various methods of rotating the principal components including a newly developed rotation procedure using frequency criteria. We related the results of these multivariate statistical techniques to a new physical model using multiple current dipoles with fixed anatomical locations and time-varying activities. We thus are able to study 3-dimensional location, time activity and interaction of multiple simultaneously active brain sources in the scalp EEG.

Multiple source localization using genetic algorithms.

We present a new procedure for localizing simultaneously active multiple brain sources that overlap in both space and time on EEG recordings. The source localization technique was based on a spatio-temporal model and a genetic algorithm search routine. The method was successfully applied to the localization of two dipole sources from several sets of simulated potentials with various signal-to-noise ratios (SNR). The different SNR values resembled evoked responses and epileptic spikes as commonly seen in the laboratory. Results of the simulation studies yielded localization accuracy ranging from 0.01 to 0.07 cm with an SNR of 10; from 0.02 to 0.26 cm with an SNR of 5; and from 0.06 to 0.73 cm when the SNR was equal to 2. Additionally, two sets of simulations were based on the dipole arrangements and time activities of data obtained during electrical stimulation of the median nerve in human subjects. These studies yielded localization accuracy within 0.1 cm. We also studied the localization accuracy of the algorithm using a physical model incorporating potential measurements of two current dipoles embedded in a sphere. In this situation the algorithm was successful in localizing the two simultaneously active sources to within 0.07-0.15 cm.

Identifying and referring patients with epilepsy for surgery.

Surgery appears to be useful for a number of medically refractory epileptic patients. Temporal lobectomy, the most common surgical procedure, achieves significant reductions in seizures for most treated patients. With recent advances in technology, successful epilepsy localization can be performed in many patients without the need for invasive intracranial recordings.

Human somatosensory cortical finger representation as studied by combined neuromagnetic and neuroelectric measurements.

We studied somatotopy of human hand somatosensory cortex using evoked responses recorded on magnetoencephalogram (MEG) and scalp-electroencephalogram (EEG) in conjunction with dipole modeling. We found a somatotopic arrangement of cortical digit representations with a sensory sequence from lateral inferior to medial superior in the anatomical order thumb, index finger, middle finger, ring finger, and little finger. MEG alone was able to reproduce this sensory sequence more accurately than scalp-EEG alone. However, the combined information provided by both techniques improved localization accuracy even further. As MEG and scalp-EEG are complementary and confirmatory techniques, this combined approach was useful to get more complete information on the functional organization of human hand somatosensory cortex.

A multivariable evoked potential measure applied to human epilepsy.

Based upon a three-dimensional plot (response plane) representing a condensation of human sensory evoked potential data obtained from a paired stimulus protocol, a study was undertaken to investigate electrophysiological changes following sodium valproate administration in subjects with photosensitive epilepsy. Initial results in two patients and two normal subjects suggest that the measure is reproducible and sensitive to electrophysiological changes accompanying anticonvulsant administration.

Interictal cerebral metabolism in partial epilepsies of neocortical origin.

We performed interictal [18F]fluorodeoxyglucose positron emission tomography (FDG PET) in 24 patients with partial epilepsy of neocortical origin. Two-thirds of patients had regions of hypometabolism. The zone of intracranially recorded electrographic ictal onset was always located in a region of hypometabolism, in those with hypometabolism. Hypometabolic regions in partial epilepsies of neocortical origin were usually associated with structural imaging abnormalities. Regional hypometabolism occasionally occurred without localizing ictal scalp EEG and cerebral magnetic resonance imaging findings, however. FDG PET may be useful in directing placement of intracranial electrodes for presurgical evaluation of refractory neocortical seizures.

Surgical treatment of partial epilepsy arising from the insular cortex. Report of two cases.

Despite its documented connections with many limbic structures, the role of the insula in the etiology of partial seizures is poorly understood. Two patients are described in whom lesions of the insula were associated with intractable partial seizures. In the first patient, the seizures involved visceral sensory hallucinations followed by motor automatism. Seizures in the second patient began with somatic sensory hallucinations and then produced visceral motor effects. Both patients were found to have low-grade astrocytomas of the insula. In both instances, resection of the lesion and adjacent insular cortex resulted in a cure of the seizures. These cases are placed within the context of the existing literature on the subject.

Influence of magnetic source imaging for planning intracranial EEG in epilepsy.

BACKGROUND: Magnetic source imaging (MSI) is used routinely in epilepsy presurgical evaluation and in mapping eloquent cortex for surgery. Despite increasing use, the diagnostic yield of MSI is uncertain, with reports varying from 5% to 35%. To add benefit, a diagnostic technique should influence decisions made from other tests, and that influence should yield better outcomes. We report preliminary results of an ongoing, long-term clinical study in epilepsy, where MSI changed surgical decisions. METHODS: We determined whether MSI changed the surgical decision in a prospective, blinded, crossover-controlled, single-treatment, observational case series. Sixty-nine sequential patients diagnosed with partial epilepsy of suspected neocortical origin had video-EEG and imaging. All met criteria for intracranial EEG (ICEEG). At a surgical conference, a decision was made before and after presentation of MSI. Cases where MSI altered the decision were noted. RESULTS: MSI gave nonredundant information in 23 patients (33%). MSI added ICEEG electrodes in 9 (13%) and changed the surgical decision in another 14 (20%). Based on MSI, 16 patients (23%) were scheduled for different ICEEG coverage. Twenty-eight have gone to ICEEG, 29 to resection, and 14 to vagal nerve stimulation, including 17 where MSI changed the decision. Additional electrodes in 4 patients covered the correct: hemisphere in 3, lobe in 3, and sublobar ictal onset zone in 1. MSI avoided contralateral electrodes in 2, who both localized on ICEEG. MSI added information to ICEEG in 1. CONCLUSION: Magnetic source imaging (MSI) provided nonredundant information in 33% of patients. In those who have undergone surgery to date, MSI added useful information that changed treatment in 6 (9%), without increasing complications. MSI has benefited 21% who have gone to surgery.

Magnetic source imaging localizes epileptogenic zone in children with tuberous sclerosis complex.

The authors assessed whether magnetoencephalography/magnetic source imaging (MEG/MSI) identified epileptogenic zones in patients with tuberous sclerosis complex (TSC). In six TSC children with focal seizures, ictal video-EEG predicted the region of resection with 56% sensitivity, 80% specificity, and 77% accuracy (p = 0.02), whereas interictal MEG/MSI fared better (100%, 94%, and 95%, respectively; p < 0.0001). Interictal MEG/MSI seems to identify epileptogenic zones more accurately in children with TSC and focal intractable epilepsy.