PhD Thesis Final Defense to be held on July 10, 2019, at 12:00

Photo credit: Adrianos Katsouris

The examination is open to anyone who wishes to attend (Multimedia Amphitheatre, Central Library of NTUA).


Abstract: In the present PhD dissertation, algorithmic techniques for calculating the equivalent far-field radiation generated by the electrical activity of the brain during cognitive processes are presented. The aim is to introduce the study of the far-field radiation produced by Electroencephalography signals as a new method for analysis, complementing the existing conventional methods, both in time- and frequency- domains. The starting point of this research was the development of an innovative Electroencephalography inverse problem technique, utilizing models composed of multiple distributed elementary dipoles, determined by Electromagnetic Theory. The emergence of the far-field radiation as a new metric for the assessment of the brain’s electrical activity is verified by developing new far-field radiation calculation techniques, which extend the well-established sLORETA method. An indicative application of the developed algorithms to real experimental data demonstrates the validity of the methods as well as the prospects for their use for the assessment of behavioral characteristics, both in clinical and research applications.

The 1st Chapter presents the scientific field of Cognitive Neuroscience, which is expected to significantly benefit from the development of methods calculating the far-field radiation, emphasizing its historical evolution from antiquity to modern times. Also presented in detail is the related "Protagoras" experiment, which was recently conducted by the 1st University Psychiatric Clinic of the Eginition Hospital and aimed at the study of mental time travel. The experimental test data was used both to be the scope and implementation of the algorithms that follow, and to extract psychophysiological conclusions for the purposes of the experiment.

Subsequently, in Chapter 2, the anatomy of the brain, the structure and function of the neurons, as well as the electric potentials’ propagation through the nerve fibers are presented as generating causes of the electrical dynamics on the surface of the head. These potentials are recorded with the Electroencephalography method and are the product of brain activity that occurs during cognitive processes. Also, relevant techniques of digital processing of Electroencephalography data currently in use, as well as other methods of recording and visualizing brain functions are described.

In Chapter 3, the Electromagnetic Theory as well as elements from Antenna Theory are developed, forming a theoretical background for the following modeling techniques. Starting from Maxwell's classical equations, the relations determining the electric and magnetic fields produced by elementary sources (dipoles) in both the near- and the far-regions are developed. As such, the differences between the two regions reflected in the form of surface dynamics of the brain as compared to its radiation pattern are identified.

Subsequently, the 4th Chapter develops the concept of the Electroencephalography inverse problem, which concerns the determination of a set of equivalent elementary sources that can reproduce the recordings. The difficulties encountered in solving it as well as relevant techniques in the bibliography are presented in detail, with emphasis on the assumptions and hypotheses of each. Furthermore, the new method of solving this inverse problem developed during the research is presented, utilizing elemental magnetic dipoles of appropriate topology. Techniques are presented to reduce the complexity of the problem as well as for calculating the far-field produced by the determined multiple dipole models. Also, data from the "Protagoras" experiment are input to the algorithms and the results of the new method’s application for extracting psychophysiological conclusions are presented.

Finally, Chapter 5 describes extensively the developed sLORETA extension techniques concerning the determination of the far-field of the Electroencephalography brain recordings, both in frequency- domain and in time- domain. Applying the “Protagoras” data to these makes it possible to statistically investigate the alteration in the brain radiation pattern during the cognitive tests of the experiment, offering new possibilities for extracting behavioral norms.

Supervisor: Christos Capsalis, Professor

PhD student: Adrianos Katsouris