Multiple sparse volumetric priors for distributed EEG source reconstruction

Abstract

We revisit the multiple sparse priors (MSP) algorithm implemented in the statistical parametric mapping software (SPM) for distributed EEG source reconstruction (Friston et al., 2008). In the present implementation, multiple cortical patches are introduced as source priors based on a dipole source * Corresponding author -phone number: +32484777651 space restricted to a cortical surface mesh. In this note, we present a technique to construct volumetric cortical regions to introduce as source priors by restricting the dipole source space to a segmented gray matter layer and using a region growing approach. This extension allows to reconstruct brain structures besides the cortical surface and facilitates the use of more realistic volumetric head models including more layers, such as cerebrospinal fluid (CSF), compared to the standard 3-layered scalp-skull-brain head models.We illustrated the technique with ERP data and anatomical MR images in 12 subjects. Based on the segmented gray matter for each of the subjects, cortical regions were created and introduced as source priors for MSP-inversion assuming two types of head models. The standard 3-layered scalp-skullbrain head models and extended 4-layered head models including CSF. We compared with the current implementation by assessing the free energy corresponding with each of the reconstructions using Bayesian model selection for group studies. Strong evidence was found in favor of the volumetric MSP approach compared to the MSP approach based on cortical patches for both types of head models. Overall, the strongest evidence was found in favor of the volumetric MSP reconstructions based on the extended head models including CSF. These results were verified by comparing the reconstructed activity. The use of volumetric cortical regions as source priors is a useful complement to the present implementation as it allows to introduce more complex head models and volumetric source priors in future studies.

Improved Localization of Seizure Onset Zones Using Spatiotemporal Constraints and Time-Varying Source Connectivity

“…Sensor noise and uncertainty on the propagation model are represented by Ξ ∈ℝ C × T . Here, the source space projection is performed via the Multiple Sparse Prior ( MSP ) algorithm that includes a two level-hierarchical parametric empirical Bayesian model, for which the source covariance is the weighted sum of multiple empirical priors with compact spatial support so that they are independent but locally determined on the basis of brain anatomy (Strobbe et al, 2014). Hence, the weighted parameters are optimized just based on the available data by maximizing the so-termed Free Energy cost function and using standard variational schemes such as Expectation Maximization (Wipf et al, 2010).…”

“…The proposed methodology for SOZ localization, based on source connectivity analysis, comprises the following procedures: – Head model construction : Individual head models are constructed for each patient by using the Finite Difference Method (FDM), making use of their presurgical anatomical MR images as performed in Strobbe et al (2014). From the MR images, the nested meshes representing the scalp, outer skull, and inner skull are extracted and additionally converted to volumes by using the SPM Matlab toolbox.…”

“…– Source reconstruction : To estimate the cortical sources, we use the time adaptive version of the Multiple Sparse Volumetric Priors algorithm (Strobbe et al, 2014) 2 . The main variation of the volumetric MSP version is the incorporation of patches based on the neighboring information extracted from the gray matter, allowing the application of FDM forward models.…”

Electrical source imaging of interictal spikes using multiple sparse volumetric priors for presurgical epileptogenic focus localization

“…In order to apply the MSVP technique, the orientations of the dipoles were determined based on the curvature of the segmented white matter (see Phillips et al (2002) and Strobbe et al (2014b) for more details). Based on the dipole source space in each patient, we constructed 100 possible sets of 256 sparse volumetric regions.…”

“…In this paper we evaluate an ESI technique that allows to estimate the activity of sources distributed in the brain of the patient corresponding with a specific time epoch of the interictal spike activity. It is an application of our previous work in which we suggested to use multiple sparse volumetric priors (MSVP) for ESI using the hierarchical Bayesian framework implemented in the statistical parametric mapping software 1 ( Strobbe et al, 2014a , Strobbe et al, 2014b ). Compared to the more traditional approaches, where the sources are typically estimated that correspond to the spike peak, or to 50% of the spike peak during the rising phase of the spike ( Boon et al, 1997a , Boon et al, 1999a , Brodbeck et al, 2011 , Birot et al, 2014 ), the choice of the time epoch in order to localize the origin of the activity using the MSVP method is not clear.…”