Real-time estimation of heart rate variability parameters from passband filtered interbeat interval series

Real-time heart rate variability (HRV) analysis provides means to measure neural system responsiveness and has recently been found useful in cardio-resynchronization therapy optimization. Computationally efficient methods for LF and HF power estimation are proposed. The methods are based on interbeat interval series decomposition into LF and HF components using short, infinite impulse response filters. The LF and HF instantaneous powers are computed for time windows of adaptive lengths of duration of LF and HF cycles respectively. The proposed methods outperform considerably the spectral methods in applications requiring fast tracking of HRV parameters.