Abstract: 

The electroencephalogram is an average recording of brain electric potentials that, despite its limited spatial resolution, continues to provide a real-time window into the underlying brain dynamics. It, therefore, comes with a unique set of advantages, and challenges, when used to model and understand brain rhythms. On the one hand, detailed models of single cells, ubiquitous in computational neuroscience, are often tuned to result in nonlinear, time varying dynamics. On the other hand, average models of activity in neural populations, popular in clinical applications, are often approximately tuned to macroscale recordings (EEG, fMRI) with basic nonlinearities and time-invariant dynamics. In this talk I will discuss our recent efforts to introduce more detailed laminar population activity models with known connection topologies and dynamics. It is seen that, with the ensuing increase in dynamic richness, the principal challenge of identifying model parameters from macroscale recordings can be approached with new Kalman-based estimation techniques that allow for identifying nonlinear, time-varying dynamics.

Biography:

Fadi N Karameh is an Associate Professor in the Electrical and Computer Engineering Department at the American University of Beirut (AUB) in Beirut, Lebanon. Prof Karameh joined AUB in 2003 shortly after graduating from the Laboratory of Information and Decision Systems at the Massachusetts Institute of Technology (MIT) in Cambridge, USA. With an emphasis on neurophysiological signals and systems, his interdisciplinary research brings together detailed understanding of cellular mechanisms in the area of computational neuroscience and system-theoretic approaches in identification, estimation and signal processing in the area of electrical engineering. His current efforts use EEG recordings for the analysis, detection and modeling of the initiation and propagation of stimulus-induced therapeutic seizures in cortical networks.