Investigation of epileptic structure properties by transfer function.

The efforts aimed at the objective analysis of the electrical activity of the brain have recently become increasingly intensive. The development of mathematical methods and computer science have greatly stimulated this research. Since the first experiments with Fourier analysis, many more complex methods have already been put into practice out of which correlation analysis appeared to be the most convenient one. The latter was recommended for detecting similarities and dependencies between signals, and registrations from various brain structures. The method was expected to facilitate determination of the source of epileptic activity as well as an aid in finding ways of its propagation. But the simple application of these methods, such as Fourier analysis or the correlation analysis may, however, lead to formal conclusions which, apart from evoked potentials, do not show any functional dependencies among the brain structures investigated which, in our case, were epileptic structures. If it has to be found, how in EEG signals represent the structural brain organization from the standpoint of both information processes and the transfer dynamics within the brain, more complicated mathematical procedures have to be assumed.