Kaoru Seki

Functional localization of bilateral auditory cortices using an MRI-linked whole head magnetoencephalography (MEG) system

In 20 healthy subjects, auditory evoked magnetic fields were measured over the entire head, using a helmet-shaped 66-channel MEG system linked to MRI. When the left or right ear was stimulated by 60 msec 2 kHz tones, the prominent 100 msec response (N100m) appeared significantly earlier in the contralateral hemisphere than in the ipsilateral one. In 16 cases, the N100m dipolar field patterns were clear in both hemispheres, overlapping each other across the midline. The N100m sources were estimated using a 2-dipole model in a spherical conducting medium with the size and location of the sphere determined individually according to the MRI images. No differences were found between the contralateral and ipsilateral N100m dipole positions in one hemisphere. When superimposed on MRI, the N100m dipoles were located precisely on the upper surface of bilateral temporal lobes with a standard deviation of 2.2 mm in the superior-inferior direction. In 16 right handed males, the right hemispheric N100m dipoles were 6 mm anterior to the left hemispheric dipoles. The whole head MEG is suitable to see small but significant differences of bilateral cerebral function, with exceptionally high spatial resolution, confirmed by the MRI-linked system.

Neuromagnetic evidence that the P100 component of the pattern reversal visual evoked response originates in the bottom of the calcarine fissure

Visual evoked magnetic fields due to pattern reversal stimuli were measured in 5 normal subjects using a helmet-shaped 66 channel magnetoencephalography system linked to MRI. The magnetic topography of the prominent 100 ms response (P100m) evoked by fullfield visual showed a double-dipole pattern in the occipital areas of all subjects. Right or left half-field stimuli and upper or lower quadrant-field stimuli evoked a single-dipole pattern in the contralateral occipital area. The P100m sources were then localized using a current dipole model and superimposed on MRI images of each subject. The visual cortex was morphologically variable among the subjects, but the P100m dipoles were all localized at the lateral bottom of the calcarine fissure. Moreover, these P100m dipoles had similar orientations for both half-or quadrant-field stimuli. These results suggest that the P100m is located in a smaller part of the striate cortex than previously reported.

Sleep Spindles in Magnetoencephalography and Electroencephalography

Sleep spindles are one of the characteristic EEG transients of sleep. They are found at a frequency range of 12.5–16.0 Hz and primarily at the vertex. This important pattern, however, is poorly understood neurophysiologically and its origin remains controversial. A previous neuromagnetic study (Hughes et al., 1976) indicated there are few sleep spindles in MEG recordings, however, we observed many MEG spindles above the central vertex (Nakasato et al., 1988). We also indicated that sleep spindles were under three distinct conditions: (i) when they were detected simultaneously in MEG and EEG, (ii) when they were detected alone in MEG without comparable EEG activity, (iii) when they were detected alone in EEG without comparable MEG activity. In this study, we measured the MEG at larger number of point over the scalp to show the spatial distribution of sleep spindles in MEG.

Neuromagnetic Identification of the Somatosensory Cortex in Cases with Arteriovenous Malformation Adjacent to the Central Sulcus

It is critical to preserve eloquent cortices during surgery of cerebral arteriovenous malformations (AVMs). Although MRI scans may identify “anatomical” central suclus, it is controversial whether the “functional” central sulcus can be shifted due to AVMs. Cortical recording of somatosensory evoked potentials (SEPs) can be used to recognize the central sulcus during open surgery. However, the cortical SEPs are not available during surgery for large AVMs, intravascular surgery, or stereotaxic radiosurgery. In the present study, somatosensory evoked fields (SEFs) were measured to localize the “functional” central sulcus non-invasively in cases with AVM adjacent to the central sulcus.

Effect of Color on Visual Evoked Magnetic Fields with Pattern-Reversal Stimulation

Visual image is supposed to be processed by parallel pathways that analyze different properties of vision such as form, depth, motion, and color. Positron emission tomography (PET) studies of functional specializations of the human visual system have demonstrated that the fusiform gyrus is playing a role in color processing[1],[2]. Poor time resolution of PET, however, gave no information about the time course of the activity in the brain. Although visual evoked potentials (VEPs) studies with dipole source analysis [3],[4] supplemented PET results, no direct comparison with the anatomy of the visual cortex of subjects was made. This is because spatial resolution of source localization in VEPs is low[4]. We have been investigating visual evoked magnetic fields (VEFs), which in principle have much better spatial resolution than VEPs, and found that the most prominent response (P100) evoked with pattern-reversal stimulation was located in the primary visual cortex[5].

Visual Evoked Fields for Pattern Reversal Stimuli in Patients with Occipital Lobe Lesions

The “cruciform model” of the visual cortex suggests that the P100 generators of the visual evoked responses to pattern reversal (PR) stimuli are located throughout the entire striate cortex, including the interhemisphere surface and calcarine fissures. Our recent studies of visual evoked fields (VEFs) to PR stimuli [1–3] have localized the P100m sources near the lateral end of the calcarine fissures. We suggested that a smaller part of the striate cortex contributes to the P100m response than in the cruciform model. In the present study, we measured PR-VEFs in patients with homonymous hemianopsia due to unilateral occipital lobe lesions, to demonstrate the correlation between PI00m patterns and occipital lesions.

Visual Evoked Magnetic Fields: Bilateral Dipole Pattern for Full-Field Stimuli

Pattern reversal stimulus is most frequently used for clinical application of visual evoked potentials (VEPs). Half-field stimulus has been employed to separate the left and right hemispheric responses. In the partial-field stimulus, however, subjects have to fix their eyes on a given point during the entire procedure; if the visual fixation is not strict, bilateral occipital responses may interfere with each other. In a clinical applications, therefore, the partial-field stimulus may not be suitable for inexperienced patients.

Origin of Slow Wave Observed in Cerebrovascular Disease

Due to excellent spatio temporal resolution and noninvasive nature of the measurements, magnetoencephalography (MEG) becomes a powerful tool for clinical application. This is especially the case in evoked MEG applied for presurgical diagnosis of the patients who suffered brain diseases. However, in spontaneous activity such as epileptic discharges, applicability is not so simple. This is because, for spontaneous waves, several brain activities or noises were superimposed on the detected signal, and separation for target activity is generally difficult. Small array of sensing coils also restricted the spatial resolution and sometimes brought erroneous results simply because they could not cover the entire brain at once. The appearance of whole-head MEG directly linked to magnetic resonance imaging (MRI) reduces data acquisition time, patient fatigue and localization error. It also provides objective information about patients which is essential for correct diagnosis. Spontaneous MEG studies used to be concentrated to epileptic spike activity. However, slow wave activity, typically observed by electroencephalography (EEG), is quite common in stroke, tumor, epilepsy, cerebrovascular disease, or head injury. Earlier investigations of slow wave activity with MEG [1–3] were restricted to small (1 to 37) channel system. Here we present the results of whole head MEG and source analysis of a slow wave activity.

Normalized N100m Latency of the Auditory Evoked Fields After Surgical Removal of Temporal Lobe Gliomas

Auditory evoked potentials have been measured in patients with several temporal lobe diseases. The N100 responses disappear in patients with lesions on bilateral superior temporal gyrus[1]. However, separation of a unilateral abnormality in evoked potentials, is difficult due to the spearing effect by tissue layers with inhomogeneous electric conductivities. Activity on the normal hemisphere may interfere with abnormal activity on the diseased hemisphere. Magnetoencephalography (MEG) is known to be less influenced by the inhomogeneous head conductivity. We have found that the whole head MEG is especially suitable to identify differences in bilateral cerebral function [2–4].

Cerebral Glucose Metabolism in Epilepsy Patients Associated with Organic Brain Damage

Four patients with porencephaly (Cases 1 and 2 ) , Sylvian fissure hematoma with ruptured middle cerebral artery (MCA) aneurysm (Case 3) and postoperative large frontal meningioma (Case 4) were studied (Table 1). Major brain damage demonstrated by CT and MRI was in the occipitoparietal lobe (Case 1 ) , the frontal lobe at the motor cortex (Case 2) and the frontal pole (Case 4) and the fronto-temporal lobe facing the Sylvian fissure (Case 3 ) . The type of seizures was CPS (Case 1: psychomotor seizure), GTC and CPS (Case 2 : long history