Beta 1


Title A mesoscopic view of neural processing, coincidence detection, and rhythms of the brain
Author Klinkby, Kristian Tjalfe (Intelligent Signal Processing, Department of Informatics and Mathematical Modeling, Technical University of Denmark, DTU, DK-2800 Kgs. Lyngby, Denmark)
Supervisor Hansen, Lars Kai (Intelligent Signal Processing, Department of Informatics and Mathematical Modeling, Technical University of Denmark, DTU, DK-2800 Kgs. Lyngby, Denmark)
Institution Technical University of Denmark, DTU, DK-2800 Kgs. Lyngby, Denmark
Thesis level Master's thesis
Year 2009
Abstract In the present work, it has been attempted to incoorporate neuro-scientific knowledge into a mathematical model of EEG generation. This does not make the model correct, however, it may be less wrong than prior attempts, which assume too little or too much about the macroscopic process of brainwave generation. The fundamental assumption providing the theoretical support for the present models, is the validity of treating GABAergic neurons as merely providing an extension to the synaptic dynamics between excitatory neurons, i.e. the EI- and EIB-partnerships. This, in turn, has paved the way for the fundamental result of the present work, that simple relations for weighting the significance of excitatory stimulation may provide approximate peak postsynaptic potential and thus simple evaluation against the activation threshold. The theoretical support of such an EI-partnership has in part been argued for through analysis of the histological makeup of cortical tissue and from the empirical evidence found in [22] and [28]. Another important assumption of the present work, is that excitatory stimulation arrive in fairly well defined spike packages, for which spike timing dependent plasticity provide improved temporal focus. If serious arguments against one of these assumption arise, the presented model should be taken up for serious reconsideration. However, providing these assumptions hold, an efficient method of evaluating cortical activity has been achieved. Based on empirical evidence of neural connectivity and synaptic variability, a probalistic model of neural population excitablity may be quickly constructed, evaluated and used as statistical priors for neural EEG sources.
Imprint Technical University of Denmark (DTU) : Kgs. Lyngby, Denmark
Series IMM-M.Sc.-2009-45
Fulltext
Original PDF ep09_45_net.pdf (3.43 MB)
Admin Creation date: 2009-08-19    Update date: 2009-10-27    Source: dtu    ID: 248262    Original MXD