Cardiological Data Analysis Algorithms Based on Wavelet Theory

Abstract

The report presents tools for analyzing cardiac data using various wavelet transforms. Algorithms for the analysis of heart rate variability, which is a dynamic, non-stationary variable, are presented. Heart rate analysis with mathematically based methods is a topical issue. Electrocardiography and long-term Holter recordings have established themselves as non-invasive medical methods for assessing cardiovascular activity. In the time domain parameters established in practice are determined. Spectral analysis of heart rate variability makes it possible to assess the work of the heart and to assess its condition in the coming days. Spectral analysis is usually performed in three frequency bands and can be done by different mathematical methods. The analyzes were performed on real long-term Holter records for patients with proven heart disease diagnosed by a cardiologist and for people without cardiovascular problems. The presented numerical and graphical results were obtained with the help of the MATLAB software program. Comparative analyzes show differences in the studied frequency parameters between patients with heart disease and healthy individuals. The research performed and the results obtained can be useful in the clinical practice of cardiologists.