AUTHOR=Nascimento Diego C. , Pinto-Orellana Marco A. , Leite Joao P. , Edwards Dylan J. , Louzada Francisco , Santos Taiza E. G. 

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

JOURNAL=Frontiers in Systems Neuroscience

VOLUME=Volume 14 - 2020

YEAR=2020

URL=https://www.frontiersin.org/journals/systems-neuroscience/articles/10.3389/fnsys.2020.527757

DOI=10.3389/fnsys.2020.527757

ISSN=1662-5137

ABSTRACT=Sparse time series models have shown promise for estimating contemporaneous and ongoing brain connectivity. This paper was motivated by a neuroscience experiment, using EEG signal data as the outcome of our established interventional protocol, a new method in neurorehabilitation towards the development of a treatment for visual verticality disorder in post-stroke patients. To analyze the complex outcome measure (EEG) that reflects the neural-network  functioning and processing in more specific way in relation to traditional analyses, we present the comparison among sparse time series models (classic VAR, GLASSO, TSCGM, and TSCGM-modified with nonlinear and iterative optimizations) combined with the graphical approach, as a Dynamic Chain Graph Model. These dynamic graphical models were useful to assess the role of estimating the brain network structure and describing its causal relationship. Additionally, this method allowed visualization and comparison across experimental conditions and across the 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 application of sparse dynamic models to EEG data may be useful for describing intervention-relocated changes in brain connectivity.