|
Project of the Scientific Grant Agency VEGA 2/4026/04
Duration of the project: 01/2004 - 12/2006
Head of the project: Doc. RNDr. Viktor Witkovský, CSc.
Co-operating institutions: Faculty of Mathematics, Physics and Informatics of CU, Faculty of Philosophy of CU, and Institute of Measurement Science SAS
Financial support from VEGA: 182 thou. SKK
Project Summary
The project is focused on application and development of advanced methods of mathematical statistics, artificial neural networks, and non-linear dynamical systems to analyse data measured in order to better understand the human brain activity. Objective lies in usage of traditional methods and in investigation of new original algorithms which are capable to characterize and predict specific states of brain based on measured EEG signal. Prime areas of interest concern sleep stages, relaxation and pathologies (e.g. epilepsy). The results may be applicable in neuro-diagnostics and therapy as well as in design of effective strategies aimed at improvement of cognitive abilities.
Results
The results of the project contribute to clarification of the effects of the short-term and the long-term audio-visual stimulation (AVS) of the human brain. The methods, utilized in the project, are based on the current research results in the field of analysis of the EEG signals of the human brain. In particular, the methods for analysis of non-linear dynamical systems, the methods based on a self-organized artificial neural networks, and the statistical methods. We have analyzed EEG data of three different types - whole-night sleep records, records measured during relaxation, and data from 20-minute AVS program, recorded repeatedly during 25 days for more healthy volunteers. Impact of AVS was researched in the frame of direct, transient, and long-term effects. The results of the project contribute to definition of the relaxation and to objective quantitative characterization of relaxation state. Further, in the project, we have identified the measurable effects of AVS and supported the hypothesis of describing EEG signal as 1/f noise. We have highlighted the classification and prediction abilities of selected complexity measures used in the field of analysis of nonlinear dynamical systems (especially, the fractal coefficient).
Selected Publications
- TEPLAN, M. - KRAKOVSKÁ, A. - ŠTOLC, S.: EEG responses to long-term audio-visual stimulation, International Journal of Psychophysiology, Volume 59, Issue 2, February 2006, Pages 81-90
- TEPLAN, M. - KRAKOVSKÁ, A. - ŠTOLC, S.: Short-term effects of audio-visual stimulation on EEG. Measurement Science Review, Volume 6, Section 2, No. 4, 2006, Pages 67-70 KRAKOVSKÁ, A. - ŠTOLC, S.: Fractal complexity of EEG signal. Measurement Science Review, Volume 6, Section 2, No. 4, 2006, Pages 63-66
- ŠUŠMÁKOVÁ, K.: Correlation Dimension versus Fractal Exponent During Sleep Onset, Measurement Science Review, 6, Section 2, No. 4, 2006, 58-62 TEPLAN, M.: Audio-visual stimulation and relaxation. Linear and nonlinear EEG measures, Dissertation thesis, Institute of Measurement Science, Slovak Academy of Sciences, 2006
- ŠUŠMÁKOVÁ, K.: Finding the most sensitive measures for sleep stages detection, Neuromath '06, Conference on Mathematical Neuroscience, Sant Juliá de Lória, Principat d'Andorra, 2006
- BAJLA, I. - RUBLÍK, F. - ARENDACKÁ, B. - FARKAŠ, I. - HORNIŠOVÁ, K. - ŠTOLC, S. - WITKOVSKÝ, V.: Segmentation and Supervised Classification of Image Objects in Epo Doping-Control. Submitted to Machine Vision and Applications.
|
