|
|
|
 EVENT-RELATED SPECTRAL ANALYSIS (ESA) ELECTRICAL CORTICAL STIMULATION (ECS) MAPPING Electrocorticography (ECoG) The electrical potentials that are generated by the brain are recorded from small disk-shaped electrodes that are embedded in a soft plastic sheet that is surgically implanted on the surface of the cerebral cortex, underneath the dura mater (hence the name, subdural electrodes). This invasive procedure is done for clinical purposes, most commonly in patients with intractable epilepsy, prior to surgical removal of brain tissue near functionally important brain regions (e.g. language cortex). The electrodes are implanted by a neurosurgeon such that they are connected to wires that exit the skull and scalp so that they can be connected to amplifiers after the patient recovers from surgery. After the first surgery the patient is monitored in a specially equipped hospital room for several days. During this time the electrodes may be used to record the patient's seizures to decide which part of the brain should be removed to alleviate the seizures. The electrodes may also be used to map language and other important brain functions using electrical cortical stimulation (ECS, see below). In addition, the electrodes may be used to record the normal activity of the brain while they are performing language or other cognitive tasks. These electrocorticographic (ECoG) recordings may then be analyzed for changes that indicate which parts of the brain were active during the tasks. The patient must undergo a second operation to remove the electrodes, and at this time the neurosurgeon may remove the brain tissue that was identified to cause seizures. Event-Related Potentials The most commonly used index of cortical activation in electrophysiological recordings is the event-related potential (ERP), derived from averaging the signal in the time domain. The theoretical basis for this method is that the signal is composed of two elements: (1) an ERP that is time and phase-locked with an exogenous or endogenous event, and (2) signal components that are not phase-locked with the event. With averaging, non-phase-locked signals are cancelled out. The basic assumption of this approach is that all relevant aspects of the signal are not only time-locked, but also phase-locked with the event under study. Event-Related Spectral Analysis An alternative approach to translating an electrophysiological signal consists of examining event-related changes in its power spectrum (Pfurtscheller, 1999). In this approach the signal is first converted into the frequency domain, and then the signal's power within a particular frequency band is averaged. To see if this band power changes significantly in response to the event under study, averaging is temporally anchored to the event, and band power in epochs following the event are compared with band power in a baseline period preceding the event. Like averaging in the time domain, the detection of an event-related power change requires that it is time-locked, but phase-locking is not necessary. Electrical Cortical Stimulation (ECS) Mapping ECS mapping is often done in patients who have been surgically implanted with subdural electrodes for intractable epilepsy. It consists of passing a weak electrical current through a portion of the brain while the patient is asked to perform a language task. If the current interferes with the language task, it is interpreted as evidence for that part of the brain being at least partly responsible for that language task. The drawback of ECS is that it takes more time to test each part of the brain, one at a time, than it does to test all parts of the brain at once, as can be done with electrocorticographic mapping with ESA (see above). In addition, ECS sometimes precipitates a seizure, which can disrupt and delay further language mapping. Therefore, we hope that in the future ESA may be used as a "first pass" to map all brain regions simultaneously. Then ECS could be used to verify the function of brain tissue that might be removed. |
© Copyright 2002 | All Rights Reserved
Clinical Neurophysiology Laboratory
The Johns Hopkins University |
Department of Neurology
Meyer 2-147 | 600 N. Wolfe Street
Baltimore, Maryland
21287 USA
410-955-6772 (tel) | 410-955-0751 (fax)
