SPATIOTEMPORAL DECOMPOSITION OF THE EEG - A GENERAL-APPROACH TO THE ISOLATION AND LOCALIZATION OF SOURCES

The principal-component method of source localization for the backgrou nd EEG is generalized to arbitrary spatio-temporal decompositions. It is shown that as long as the spatial patterns of the decomposition spa n the same signal space as the principal spatial components, the compu tational process of attempting to localize the sources is the same. De compositions other than the principal components are shown to be super ior for the EEG in that they appear to enable individual sources to be better isolated. An example is given using the common spatial pattern decomposition and using a raw varimax rotation of a subset of the com mon spatial patterns. The results show that the principal component de composition is almost ineffective for isolating spike and sharp wave a ctivity in an EEG from a patient with epilepsy, that the common spatia l pattern decomposition is significantly better and that the varimax r otation is better yet. That the varimax rotation is best is demonstrat ed by attempting to locate dipole sources inside the brain which accou nt for the spike and sharp wave activity on the scalp. The question wh ich remains is whether there exists some oblique rotation of the basis vectors of the EEG signal space which is optimal for isolating indivi dual sources.