Modeling of ventricular repolarisation time series by multi-layer perceptrons

Abstract

The QT interval measured on the body-surface Electrocardiogram (ECG), corresponds to the time elapsed between the depolarization of the first myocardial ventricular cell (beginning of the Q wave) and the end of the repolarisation of the last ventricular cell (end of the T wave). Predictive models of the QT dynamical behavior are believed to be a useful tool to detect abnormalities in QT adaptation for heart rate changes. In this paper, patient specific predictive models based on a Multi Layer Perceptrons are presented and their predictive performances are tested on real and artificial data.