The short-term regulation of blood pressure is influenced by many influences, some being represented by cardiovascular signals and their changes. Because of the complexity of this system, the linear methods for its analysis are not sufficient. Non-linear methods for time series analysis have been devised quite a lot. In the course of this work, the analysis of the relations of several signals was evaluated as the most suitable conditional entropy and the resulting indexes of affinity and directionality. This method was applied to a set of heart-rate signals and systolic and diastolic pressure signals measured on eight adults and nine children. Relationships were analyzed but unfortunately after the statistical test was held the expected information links between the individual signals were not established. The indices were very small - in the hundredths of bits. Finally, optimization of the algorithm of the whole method has been performed and the newly modified function already shows significantly better results, for example strong information binding from the time-series of systolic pressure to a series of diastolic pressures.
| Identifer | oai:union.ndltd.org:nusl.cz/oai:invenio.nusl.cz:378149 |
| Date | January 2018 |
| Creators | Lokaj, Jiří |
| Contributors | Ronzhina, Marina, Kolářová, Jana |
| Publisher | Vysoké učení technické v Brně. Fakulta elektrotechniky a komunikačních technologií |
| Source Sets | Czech ETDs |
| Language | Czech |
| Detected Language | English |
| Type | info:eu-repo/semantics/masterThesis |
| Rights | info:eu-repo/semantics/restrictedAccess |
Page generated in 0.0153 seconds