A neural network for tracking the prevailing heart rate of the electrocardiogram

An artificial neural network (ANN) with feedback for tracking the prevailing heart rate of the electrocardiogram (EKG) is presented. The ANN accurately tracks the change of rate over a wide range of heart rates, and is robust in the presence of arrhythmic and anomalous conditions. Such a network has potential application in the development of a robust heart rate monitor or in the enhancement of the rhythm monitoring system. The ANN was trained using the backpropagation learning algorithm. The performance of the trained network was evaluated using an independent set of R-R intervals. Of the 270 test exemplars, in 226 cases the predicted prevailing R-R interval was within 1% of the observed prevailing R-R interval, in 38 cases the prediction was within 2% of the observed, and in the remaining six cases the prediction was within 4% of the observed.<>