Rabj Order Based Decomposition and Classification of Heart Rate Signals

摘要

A new technique for decomposing composite signals into trend and cyclic components is addressed. Contrary to other techniques, it does not require any a priori knowledge about the functional form or distribution of the signal components. The algorithm is applied to extract cyclic components of heart rate time series retrieved from implanted defibrilators. It is shown that after the decomposition of these signals, better classification can be performed between halthy subjects and patients suffering from ventricular tachyarrhythmias.