Farouk Nathoo

Statistical Modeling of Electromagnetic Neuroimaging Data


I will present some recent work involving the development of statistical methods for the analysis of MEG and EEG data: (I) A skew-t space-varying regression model for the spectral analysis of resting state brain activity; (II) A sparse functional linear model for solving the ill-posed neuroelectromagnetic inverse problem based on spatial spike-and-slab priors; (III) A high-dimensional state-space model for the combined analysis of MEG and EEG data. I will discuss model formulations and computational problems, along with solutions based on mean field approximations and variational Bayes.