Evaluation of the quality of rate adaptation algorithms for cardiac pacing

The aim of this study was a comparative evaluation of some commonly used pacemaker control parameter combinations and algorithms for heart rate (HR) reconstruction. The quality of heart rate reconstruction was estimated comparing the reconstructed heart rate and intrinsic heart rate on the basis of a healthy person model. The model was developed on the basis of simultaneous measurement of intrinsic HR and different control parameters, which can electrically be recorded: QT-interval, (QT), ventilation (V)- and body movement activity (ACT). Relative mean square error (RMSE) between natural and reconstructed HR was used as a measure of HR reconstruction quality. It was shown that the results of linear and polynomial approximations for HR reconstruction practically coincided inside the range of interest for pacing in clinical practice 60-120 beats.min(-1) and there was no need to implement more complex algorithms. As for single-parameter systems the QT parameter had better reconstruction accuracy compared with ACT and V. All double-parameter systems presented HR reconstruction error level of around 10% that were within the limits suitable for clinical practice. The application of more than two parameters for HR reconstruction seems to be unnecessary, as the triple-parameter models appeared not to improve remarkably the quality of HR reconstruction.