The full text of this item is not available at this time because the student has placed this item under an embargo for a period of time. The Libraries are not authorized to provide a copy of this work during the embargo period, even for Texas A&M users with NetID.
A Modal Approach to the Space Time Dynamics of Cognitive Biomarkers
Abstract
High resolution mappings of brain activity have historically been restricted to functional magnetic resonance imaging (fMRI) measures. Brain wave recordings via electroencephalography (EEG) have nonstationary statistical information and display nonlinear dynamics but are known to contain useful information for clinical and human performance applications. As a result, a modeling technique which makes no assumptions about the stationarity of the data and accounts for the spatio-temporal dynamics is needed for clinical and human performance applications. This work presents a detailed analysis of EEG brain waves through the lens of dynamical systems, with a focus on modal analysis for mapping the brain wave patterns in real time. Here, modern system identification techniques are modified and applied to brain wave recordings. The resultant state space models cast the brain wave dynamics as a set of discrete spatio-temporal modes which recreate the original data through a weighted superposition of the modes. Modal state space models are shown to be highly inter-individual. To account for the nonlinear, nonstationary effects of the signal, an adaptive unknown input estimator is created and leveraged to update the modes in real time when nonlinear or nonstationary influences are observed. These models are shown to be relevant to cognitive modeling outcomes of interest. This work creates a new tool for the analysis of general biomarker data which can be understood by dynamic systems engineers and applied in modern neuroscience efforts.
Citation
Griffith, Tristan D. (2022). A Modal Approach to the Space Time Dynamics of Cognitive Biomarkers. Doctoral dissertation, Texas A&M University. Available electronically from https : / /hdl .handle .net /1969 .1 /197729.