How speech processing can help with beat-to-beat heart rate estimation in ballistocardiograms

Unobtrusive sensors for monitoring patients´ heart rate and other physiological rhythms often produce signals which are challenging to analyze from a signal processing point of view. This is particularly true for bed-mounted ballistocardiography (BCG) sensors. In this paper, we discuss how methods, which are commonly used for pitch tracking in speech processing, can be applied to the problem of beat-to-beat heart rate estimation from BCGs. In particular, we introduce a novel adaptive window autocorrelation method in order to improve the estimation of individual beat-to-beat intervals. Furthermore, we discuss the application of dynamic programming to the extraction of a smooth beat-to-beat interval series from time-varying correlograms. The proposed methods are evaluated with respect to RR-intervals obtained from a reference ECG. Based on the recordings of 5 subjects, the proposed adaptive window method achieved a mean beat-to-beat heart rate interval error of 0.75%.