A new multimodal imaging strategy for integrating fMRI with EEG.

For the multimodal functional neuroimaging combining fMRI and EEG, it is of significance to correct the distortion due to the fMRI-EEG mismatches, which arise from the fundamental difference between brain electric and hemodynamic activities. In the present study, we proposed a novel two-step approach (referred to as "Twomey algorithm") for the fMRI-constrained cortical source imaging. In the first step, a "hard" spatial prior derived from the fMRI was imposed to solve the EEG inverse problem with a reduced source space; in the second step, the fMRI constraint was removed and the source estimate from the first step was re-entered as the initial guess of the desired solution into an EEG least squares fitting procedure with Twomey regularization. We evaluated the performance of Twomey algorithm in comparison with the well-known 90 % fMRI-constrained Wiener filter and the EEG-alone weighted minimum norm methods in a randomized computer simulation setting. We examined the effects of invalid fMRI priors including fMRI invisible sources, fMRI extra regions and fMRI displacement. The present results in terms of the point spread function (PSF) and the localization error (LE), demonstrated that 1) relative to the EEG-alone solution, a higher spatial resolution can be achieved by using the fMRI priors in cortical source imaging; 2) the occurrence of fMRI invisible sources is the most problematic factor responsible for the error of the fMRI-EEG integrated cortical source imaging; and 3) the Twomey algorithm is more robust than the Wiener filter approach under invalid fMRI priors. The proposed Twomey approach may make an important contribution to the multimodal neuroimaging integrating fMRI with EEG.