Methods and apparatus for cardiac r-wave sensing in a subcutaneous ecg waveform

Abstract

The present invention uses a R-wave sensing algorithm that uniquely combines an automatic threshold adjustment method with a new noise rejection technique. This algorithm has significant advantages in avoiding the sensing of T-waves, P-waves, and noise/artifacts. Detecting the presence of noise bursts uses features that determine if an R-R interval adjacent to or within the noise signal is valid. Circuitry that discriminates noise signals from R-waves can use any one of several features including, but not limited to, the following: detection events occurring so close together that they are outside normal physiologic heart rates; frequency content that is wider than that of QRS complexes; amplitudes that are different than the adjacent or encompassing R-waves; and amplitudes that display greater than normal variability. The present invention employs multiple discrete thresholds optionally with different decay constants, alone or in combination with one or more substantially constant magnitude sensing threshold.