Yavorska E. Mathematical models of rhythmocardiosignal and methods of heart rhythmic characteristics determination with a predictable confidence

In the dissertation the important scientific task is solved -mathematical models of rytmocardiosignals are improved by taking into account of their nonstationary and developed methods of heart rate rhythmic characteristics determination with the predictable confidence.The known means and systems for de-scription and determination of heart rate rhythmic are analyzed. The new mathematical model of rhythmocardiograms as periodi-cally correlated stochastic sequence is on the first applied. Ex-pressions of estimations of rhythmocardiogram characteristics are obtained and digital methods of their evaluation in the bio-medical systems are developed. Methods of digital processing of rhythmocardiograms are developed and verification of the results had been made.Possibility of simplification and automation of the analysis of the non-stationary rhythmocardiogram and the pre-diction of the confidence of the analysis results had been gained by application of 2-th order statistics for the analysis of non-stationaryrhythmocardiogram on the difference as in the known methods of analysis. The methods of statistical theory of making decisions are first applied for determination of confidence of results of evaluation of a power spectral density of the rhythmo-cardiogram. The methods of the automation of rhythmocardio-gram processing are developed are suitable for modification of the known biomedical systems and methods of analysis of other biomedical signals (electroencephalograms, electromyograms and others look like). Results of the research were applied for improving of systems like look as CardioSence system.