
A statistical algorithm for fast estimation of heart rate and localization of cardiac sounds in phonocardiograms recorded by electronic stethoscopes
Author(s) -
Valery Oliynik,
AUTHOR_ID
Publication year - 2021
Publication title -
vìsnik. serìâ fìziko-matematičnì nauki/vìsnik kiì̈vsʹkogo nacìonalʹnogo unìversitetu ìmenì tarasa ševčenka. serìâ fìziko-matematičnì nauki
Language(s) - English
Resource type - Journals
eISSN - 2218-2055
pISSN - 1812-5409
DOI - 10.17721/1812-5409.2021/3.14
Subject(s) - phonocardiogram , cardiac cycle , algorithm , time domain , computer science , mathematics , cardiology , artificial intelligence , medicine , computer vision
A draft version of the algorithm for fast determination of duration of the systolic and diastolic phases of the cardiac cycle based on a statistical analysis of a digital phonocardiogram is presented. The developed technique includes localization of the cardiac tones in the record by a quantile dichotomy and modal analysis of the calculated time intervals between the adjacent detected peaks. Next, the scatter plot is constructed in terms of time intervals before and after each peak. The stability of the cardiac rhythm is determined by the density of the cluster occurring near the cross-section of two modal values plotted along the axes. At dominance of a cardiac component in the signal, the discussed approach allows quite reliable determination of the moments of the emergence of the first and second cardiac sounds tones practically on each of the cardiac cycles. This opens the possibility to analyze the variation in the duration of separate phases of heart contractions over time. Due to the work directly in the time domain and low computational costs, the algorithms of this class are suitable for application in portable diagnostic systems with limited processor power.