Lebedeva N.N., Sydorova O.A., Maragey R.A.

Natural sciences, biology and medicine in particular, are becoming one of the most important areas of applications of the fractal analysis. Fractal information effects, i.e., transmission of biologically significant information between the processes at the temporal and spatial levels determined by their mutual similarity, are in the focus of attention. Dynamic chaos of the 1/f-type fluctuations (random chaotic changes) in physical processes determines the fractal structure of biological systems. Other explicit examples of the chaotic dynamics on a shorter time scale have been found in physiological experiments. Fractal attractors in dynamical functions reflected in changes in electrical fields (functions of the brain, heart, and electrically active cells) have been studied in greater detail. The operation of the cardiovascular system and the brain activity are critical with respect to the dynamic chaos. It has been established that the normal heart rate is subjected to chaotic variations whose topology in the phase space demonstrates all features of a fractal attractor. Investigation of the role of the fluctuations with the 1/f-type spectrum showed that similar fluctuations are characteristic of the signals of the brain electrical activity (EEG). Multiple correlations of the measured EEG parameters with the clinical, neurophysiological, and neuropsychological brain indices revealed by the multivariate correlation analysis allowed us to highly appreciate the application of the multifractal method of EEG processing. Changes in the chaotic structure of the brain dynamics were found to be associated with perception of not only sensory but nonsensory stimuli as well. The problem of discrimination of the types of human emotions on the basis of acoustical speech characteristics is of interest both in theoretical and applied aspects. The authors of the article have developed the program package for the analysis of different emotionally colored speech signals. This software makes it possible to calculate the spectral speech characteristics such as peak frequencies (formants, obertones) and spectral power density and to plot the noise approximant of the spectrum to obtain the fractal number of dimension of a speech signal. Primary processing of actors - speech records during reproducing different emotional states made it possible to detect the informational features of emotions on the basis of the results of the fractal analysis. The application of the fractal analysis to the estimation of experimental results not only supplements them but may be valuable for the construction of operative express methods of the estimation of the human functional state.