ePrints.FRI - University of Ljubljana, Faculty of Computer and Information Science

Denis Mašič (1997) . Prešeren awards for students.

Full text not available from this repository.

Abstract

Electrocardiogram (ECG) represents graphical record of time alternations of the electrical activity of the heart. On the ECG we can observe different pathological changes of the heart. Beside time-domain analysis of the ECG signals , we can pefom frequency-domain analysis. Various influences of autonomic nerve system (ANS) on myocardium are reflected in the estimates of power spectrum of instantaneousheart rate (IHR). Reflections of two main different parts of ANS are observed in two seperated bands of IHR power spectrum: the low frequency band and high frequency band, reflecting the influence of sympathetic and parasympathetic part of the ANS. This study is intended to characterize power spectrum estimates of the IHR during transient ishemic episodes of the ECG. Atherosclerosis of coronary vessels, which convey blod to the heart muscle, is the main cause for ishemic disease. Ischemia is important disease which can often lead to myocardial infarction. Fast and accurate diagnosis is thus even more important . In this research we try to explain ishemic related pathophysiologic changes and their correlates with the IHR power spectrum. We use different tehniques to estimate power spectrum: The Fourier periodogram , the Lomb periodogram , an avtoregressive method and an adaptive avtoregressive method. We try to get various parameters such as computational requirement, convenience and accuracy of the methods with regard to the power spectrum estimates. In study we also test the hypothesis established in some previous studies which could explain connections betwen distinctly different types of ischemia and different temporal patterns of ishemic changes in the ECG. Hypothesis is saying that isolated episodes are due to increased oxygen demand resulting from sporadic physical activity and that salvo episodes are due to reduced oxygen suppply resulting from quasi-periodic vasoconstriction or vasospasm. Differentiating betwen these two types of ishemia on the basis of the ECG only would be significant clinical impotance in the light of economy and non-invasiveness in comparison to current clinical methods. For the test material we used European Society of Cardiology ST-T database which consists of 90 two-hour in duration ambulatory electrocardiographic records. Records contain annotaded transient ST segment episodes compatible with ischemia. We found that the best tehnique for power spectrum estimation is the adaptive avtoregressive method. By using this method we avoid moving window tehnique which is necessary in other three methods due to non-stationary behaviour of the IHR. We got the most inconsistent power spectrum estimates by uisng the Lomb periodogram. Weakness of this method lies in the inconviennce of the IHR function for this method thus not leaving correct estimates in the high frequency part of the spectrum. Statistical results confirms the hypothesis for isolated episodes. On the other hand, beside ANS, endothelial factors play an important role in salvo episodes. As we expected, the results for this group are less significant. Our future goal is to characterize spectrum of the IHR using 24-hour records of newly developed 24-hour ambulatory record database.

Item Type: Thesis (Prešeren awards for students)
Keywords:
Number of Pages: 63
Language of Content: Slovenian
Mentor / Comentors:
Name and SurnameIDFunction
prof. dr. Franc Jager235Mentor
Link to COBISS: http://www.cobiss.si/scripts/cobiss?command=search&base=51012&select=(ID=4733012)
Institution: University of Ljubljana
Department: Faculty of Computer and Information Science
Item ID: 3746
Date Deposited: 18 Jan 2017 12:34
Last Modified: 14 Feb 2017 08:07
URI: http://eprints.fri.uni-lj.si/id/eprint/3746

Actions (login required)

View Item