Neural mass models: An approach to model EEG

- Bachelor and master thesis -

                                                                                              Copyright by Kriwy

Project description:

The electroencephalogram (EEG), since its introduction in 1924, has been a widely used tool to investigate the electrical activity of the brain by placing surface electrodes on the scalp. Although it is already known that EEG reflects extracellular activity of synchronous neurons and that it can be divided into distinct frequency bands the neurophysiological mechanisms behind that collective behaviour are still unknown.

Thus several modelling attempts were proposed to shed light into that mystery. These strategies cover different scales of the brain, ranging from very detailed at the single neuron level to more coarse-grained models of cortical columns, i.e. large assemblies of neurons. In particular the simulation of the latter has attracted a lot of attention lately, as it is more feasible and allows easier conclusions of the EEG. One of these so-called neural mass models has been proposed by Jansen et al. and presents the focus of this work.

It should be emphasized that there are several alternative neural mass models to choose from if desired (for references see below). 

At first an implementation of the model followed by the reproduction of the results of Jansen et al. and others will be necessary to obtain a comprehensive understanding of the model's properties and capabilities. Only then will the application to current scientific issues be possible. The focus then will be in particular on sleep, as this work will be within the framework of the collaborative research centre SFB 654 "Plasticity and Sleep".

Promising work will thus lead to a cooperation with the experimental sleep research conducted at the Institute of Neuroendocrinology by Prof. Dr. Lisa Marshall.

Requirements:

The theoretical work requires substantial skills in mathematics in order to understand the mathematical model and programming skills in Matlab, C/C++ or Java for its implementation.

A prior knowledge of neurology and sleep is not required (albeit desired), but should be acquired through the course of the thesis. Thus a great interest in such interdisciplinarity between the fields of computer science and medicine should exist beforehand.

References:

• Electroencephalogram and visual evoked potential generation in a mathematical model of coupled cortical columns
  B. H. Jansen, V. G. Rit
  Biological Cybernetics 73, 357-366 (1995)
• A neural mass model for MEG/EEG: coupling and neuronal dynamics
  O. David, K. J. Friston
  NeuroImage 20, 1743-1755 (2003)
• Realistically Coupled Neural Mass Models Can Generate EEG Rhythms
  R. C. Sotero, N. J. Trujillo-Barreto, Y. Iturria-Medina
  Neural Computation 19, 478–512 (2007)
• A continuum model for the dynamics of the phase transition from slow-wave sleep to REM sleep
  J.W. Sleigh, M.T. Wilson, L.J. Voss, D.A. Steyn-Ross, M.L. Steyn-Ross and X. Li
  in D. A. Steyn-Ross & M. Steyn-Ross (Eds), Modeling Phase Transitions in the Brain (pp. 203-221), Springer (New York) (2010).

Contact:

Dipl.-Phys. Hong-Viet V. Ngo
Institute for Neuro- and Bioinformatics
E-Mail: ngo@inb.uni-luebeck.de