Frequency-pattern functional tomography of magnetoencephalography data allows new approach to the study of human
Rodolfo R Llinás1, Mikhail N Ustinin2
1Department of Neuroscience and Physiology, New York University School of Medicine New York, NY, USA.
Abstract:
A method based on a set of new theorems for the analysis of multichannel time series is described, based on precise Fourier transform and coherence analysis of the restored signals from a detailed set of frequency components. Magnetic field recordings of spontaneous and evoked activity by means of magnetic encephalography demonstrated that multichannel precise Fourier spectrum contains a very large set of harmonics with high coherence. The inverse problem can be solved with great precision based on coherent harmonics, so the technique is a promising platform of general analysis in brain imaging. The analysis method makes it possible to reconstruct sites and timing of electrical activity generated by both spontaneous and evoked brain function at different depths in the brain in the millisecond time range.
