Using the method of nonlinear recursive analysis for typifying electroencelography time series data
DOI:
https://doi.org/10.34185/1562-9945-2-145-2023-09Keywords:
recurrent analysis, electroencephalography, time series, recurrent diagram, delay parameter, nesting space dimension, JRQA analysis, Matlab environmentAbstract
This paper considers the issue of using the method of nonlinear recurrent analysis to the problem of typing information provided in the form of time series data of electro-encephalograms (EEG) taken from a patient. A technique for determining hidden infor-mation for this series and its use for constructing the corresponding recurrence diagram (RP) at the points of information retrieval are described. It is shown that the use of RP has significant drawbacks associated with the visualization of information on a com-puter monitor screen, so another way of research is proposed - the calculation of nu-merical indicators of RP. Their calculation must be carried out for each point of the sev-enth information, for which it was proposed to take the points (O1, O2, Pz) - these are the right and left occipital and parietal taps. The given RP indicators made it possible to typify the obtained data and determine the type of which was called "HEALTHY-RP", which distinguishes epileptic and non-epileptic EEG types.
References
Belozyorov V.Ye., Zaytsev V. G. Mathematical modeling of parkinson’s illiness by chaotic dynamics methods // Bulletin of DNU. Series "Modeling". 2017. Volume 25, No. 8, Vip. 9. P. 21–39. DOI 10.15421/
Belozyorov V.Ye., Pohorielov O.V., Serdiuk V.N., Zaytsev V. G. New Approach to Problem of Diagnostics of Cerebral Cortex Diseases Using Chaotic Dynamics Meth-ods// 7 th The international conference “Social Science and Humanity”. 23–29, Sep-tember, London, 2017. - №2.--P.7 – 28.
Torse Dattaprasad, Veena Desai, Rajashri Khanai.“Classification of Epileptic Sei-zures using Recurrence Plots and Machine Learning Techniques”. https://www.researchgate.net/publication/332675878.
Recurrence Quantification Analysis. [Online] Available: http://www.recurrence-plot.tk/rqa.php.
Systems with Emphasis on Multi-domain Feature Extraction and Classification using Machine Learning," BRAIN Broad Research in Artificial Intelligence and Neu-roscience 8.4. 2017., p. 109-129.
Kirichenko L.O., Stepanenko Y.D., Yandukov D.Y. Time series classification using recurrence charts// System technologies. N 5(136) - Dnipro,2021.- P.81 – 87. DOI 10.34185/1562-9945-5-136-2021-08
Nervous ailments / S.M. Vinichuk, E.G. Dubenko, E.L. Macheret and in.; For red. CM. Vinichuk, E.G. Dubenka. - K .: Health, 2001. - 696 p.
Mekler A.A. Application of the Apparatus for Nonlinear Analysis of Dynamic Sys-tems for EEG Signal Processing // Actual Problems of Modern Mathematics: Scien-tific Notes. p / ed. prof. Kalashnikova E.V., ed. LGU them. A.S. Pushkin, St. Peters-burg, 2004. T. 13 (issue 2), p. 112-140.
Potsdam Institute for Climate Impact Research (PIK), Germany
Belozyorov V.Ye., Zaytsev V.G. Influence of the recurrence threshold and the de-lay parameter of a time series on the information content of its recurrent diagram (on the examples of chaotic attractors). System technologies for modeling complex systems/Monograph under the general. ed. prof. A. I. Mikhaleva. - Dnepr: NMetAU_IVK “System Technologies”, 2016. -608 p. - p. 67-90.
Bodyansky E.V., Kucherenko E.I., Mikhalev A.I. Neuro-fuzzy Petri nets in the problems of modeling complex systems / Monograph. - Dnipropetrovsk: System Technologies, 2005. - 311 p.
Richard O. Duda, Peter E. Hart, David G. Stork Pattern Classification, 2nd Edi-tion. Wiley-Interscience, 2001. 688 p.
