Magnetoencephalography (MEG) measures magnetic fields that are produced by small electrical currents that arise from neuronal activity in the brain. Through analysis of the spatial distribution of the magnetic fields, MEG enables physicians to localize epilepsy-induced abnormal electrical activity within the brain. This information is then overlaid on a magnetic resonance (MR) image, which provides anatomical detail. Both functional and structural information about the brain is visible in the combined image.
MEG has a number of advantages over other imaging modalities, beginning with its noninvasive nature. “We don't have to inject anything into the patient,” unlike some nuclear imaging techniques, said Eduardo M. Castillo, Ph.D., of the University of Texas in Houston. Patients aren't subjected to radiation or strong magnetic fields. Tests can be repeated without safety concerns, making MEG an especially attractive option for children and infants.
While functional magnetic resonance imaging (fMRI), positron-emission tomography (PET), and single-photon emission computed tomography (SPECT) assess brain function indirectly, MEG takes direct measurement of the brain's electrical function in real time.
With its high temporal resolution, MEG can be used to measure events lasting only milliseconds. fMRI, PET, and SPECT have much longer time scales. MEG also has excellent spatial resolution, localizing sources of activity with millimeter precision.
Changes in electrical activity in the brain affect the associated magnetic fields. These changes in the magnetic fields are captured by the MEG machine's array of superconducting detectors and amplifiers, said Dr. Castillo.
The equipment is housed in a specially shielded room to isolate the sensor from external noise produced by vibration and from electrical devices that produce magnetic fields.
In this case, MEG provided two types of information on the location of the abnormal electrical activity (i.e., epileptiform activity) and location of his speech centers, both of which were necessary for planning the boy's epilepsy surgery.
The yellow triangles in the image on the left located the site of interictal epileptiform electrical activity. For measurement of interictal epileptiform activity, the boy was placed in the helmet-shaped sensor, resting with his eyes closed. MEG measurement of interictal epileptiform activity was done in tandem with EEG to zero in on the abnormal activity, said Dr. Castillo.
The second type of information (indicated by red dots in the image on the right) is functional activity, recorded while the child listened to a series of words, to locate language function within the brain. When mapping functional activity, such as the ability to recognize words, the patient is subjected to repetitions of specific stimuli. Brain activity is averaged across all of the repetitions, which filters out any background brain activity that is not related to the task. The language function measurement takes about 30 minutes—long enough to repeat the task twice.
Typically children older than 5 years don't need to be sedated, but younger children do in order to remain still for the duration of the test.
After the MEG-derived map of the epileptogenic zone was intraoperatively confirmed, the area was resected, sparing areas of the eloquent cortex within the dominant hemisphere language-specific cortex. The postsurgery map on the right shows that the boy's language-specific cortex was spared by the surgeon. After surgery his linguistic skills were intact, and he is currently seizure free, said Dr. Castillo. In addition, the boy has regained some of the cognitive abilities he had lost.
MEG also is used currently to map cognitive and sensory functions prior to surgery to remove brain tumors. In addition, the technique is being investigated to track the effect of different interventions following stroke, when the brain reorganizes the location of functions to compensate for the areas lost due to stroke. “We try to track changes in the organization of functions in the brain after stroke and understand how different types of interventions can modulate those changes,” said Dr. Castillo.
The group at the University of Texas in Houston is also conducting research into dyslexia and ADHD using MEG. Other groups are using MEG to better understand the progression of Alzheimer's disease.
Currently there are nine facilities in the United States that are using MEG clinically to prepare for surgery due to epilepsy or brain tumors, said Dr. Castillo.
MEG data are overlaid on an MRI to allow resection planning; yellow triangles mark interictal activity, and red dots localize language activity (left). Postsurgical image confirms sparing of language cortex (middle). The sensor array covers the head only (right). Photo Courtesy Dr. Eduardo M. Castillo
