Magnetoencephalography (MEG) is the noninvasive magnetic measurement of electrical current within the brain. It is believed to give significant new information about brain function that complements other modalities such as EEG and pet. Using MEG, it may be possible to localize sources of normal and pathological brain activity to a few millimeters in three dimensions. The MEG bandwidth extends down to dc to allow evaluation of injury currents and very low frequency changes in neuronal activity. Environmental noise is frequently the limiting factor for MEG measurements. The goal of this project is to improve the MEG signal-to-noise ratio by a combination of shielding, instrument design, and signal processing. Successful outcome of this work will be especially beneficial to measurements of spontaneous MEG where signal averaging is not feasible and to detection of signals from subcortical sources. This project has immediate significance in the clinical evaluation of epilepsy patients being considered for surgical treatment. Other potential areas of benefit include evaluation of neuromuscular disorders and observation of dc currents associated with spinal cord injuries.
