Improved EEG-fMRI integration in epilepsy

In the present work, a new method for the integration of simultaneously recorded EEG and fMRI data in epilepsy is presented. Independent component analysis (ICA) is performed on EEG data, five metrics are applied upon each component spectrum, and a general linear model for each time course is fit to the fMRI data. The proposed methodology was applied to data collected from one patient, systematically producing wider and more significant activation maps when compared with the standard square waveform regressor. Furthermore, the EEG metrics with a frequency content were better predictors of the BOLD signal than global power metrics, supporting previous theoretical predictions and experimental evidence.