The conversion of one cardiac rhythm or electrical pattern to another, almost always from an abnormal to a normal one. This conversion can be pharmacologic or electrical.
Published in Chapter:
Time-Frequency Analysis for EGM Rhythm Classification
Hamid Sheikhzadeh (ON Semiconductor, Canada) and Robert L. Brennan (ON Semiconductor, Canada)
Copyright: © 2008
|Pages: 7
DOI: 10.4018/978-1-59904-889-5.ch166
Abstract
In this article, we study the problem of rhythm classification and event detection based on intracardiac electrogram (EGM) signals. At present, only a very limited scope of signal processing is possible in implantable cardioverter defibrillators (ICDs), due to the scarcity of available resources. As a result, relatively simple beat-by-beat time-domain analysis of the EGM signal(s) is employed for rhythm detection. Recently, researchers have attempted to exploit more sophisticated signal processing methods, such as wavelet transforms and template matching (Astrom, Olmos, & Sornmo, 2006; Brown, Christensen, & Gillberg, 2002; Koyrakh, Gillberg, & Wood, 1999). However, the new methods have rarely been employed in practical systems because of their computational and power demands.