BrainWave Nets: Are Sparse Dynamic Models Susceptible to Brain Manipulation Experimentation?

Sparse time series models have shown promise in estimating contemporaneous and ongoing brain connectivity. This paper was motivated by a neuroscience experiment using EEG signals as the outcome of our established interventional protocol, a new method in neurorehabilitation toward developing a treatment for visual verticality disorder in post-stroke patients. To analyze the [complex outcome measure (EEG)] that reflects neural-network functioning and processing in more specific ways regarding traditional analyses, we make a comparison among sparse time series models (classic VAR, GLASSO, TSCGM, and TSCGM-modified with non-linear and iterative optimizations) combined with a graphical approach, such as a Dynamic Chain Graph Model (DCGM). These dynamic graphical models were useful in assessing the role of estimating the brain network structure and describing its causal relationship. In addition, the class of DCGM was able to visualize and compare experimental conditions and brain frequency domains [using finite impulse response (FIR) filter]. Moreover, using multilayer networks, the results corroborate with the susceptibility of sparse dynamic models, bypassing the false positives problem in estimation algorithms. We conclude that applying sparse dynamic models to EEG data may be useful for describing intervention-relocated changes in brain connectivity.

A group analysis using the Multiregression Dynamic Models for fMRI networked time series

Connectivity studies of the brain are usually based on functional Magnetic Resonance Imaging (fMRI) experiments involving many subjects. These studies need to take into account not only the interaction between areas of a single brain but also the differences amongst those subjects. In this paper we develop a methodology called the group-structure (GS) approach that models possible heterogeneity between subjects and searches for distinct homogeneous sub-groups according to some measure that reflects the connectivity maps. We suggest a GS method that uses a novel distance based on a model selection measure, the Bayes factor. We then develop a new class of Multiregression Dynamic Models to estimate individual networks whilst acknowledging a GS type dependence structure across subjects. We compare the efficacy of this methodology to three other methods, virtual-typical-subject (VTS), individual-structure (IS) and common-structure (CS), used to infer a group network using both synthetic and real fMRI data. We find that the GS approach provides results that are both more consistent with the data and more flexible in their interpretative power than its competitors. In addition, we present two methods, the Individual Estimation of Multiple Networks (IEMN) and the Marginal Estimation of Multiple Networks (MEMN), generated from the GS approach and used to estimate all types of networks informed by an experiment —individual, homogeneous subgroups and group networks. These methods are then compared both from a theoretical perspective and in practice using real fMRI data.

•

The group-structure (GS) approach models possible heterogeneity between subjects.

•

The search for distinct homogeneous groups is based on the connectivity maps.

•

All four methods for group network were compared using synthetic and real fMRI data

•

IEMN and MEMN estimate individual, homogeneous subgroups and group networks.