Fast Detrending of Unevenly Sampled Series with Application to {HRV}

Detrending RR series is a common processing step prior to HRV analysis. In the classical approaches RR series, which are inherently unevenly sampled, are interpolated and uniformly resampled, thus introducing errors in sub- sequent HRV analysis. In this paper, we propose a novel approach to detrending unevenly sampled series and apply it to RR series. The approach is based on the notion of weighted quadratic variation, which is a suitable mea- sure of variability for unevenly sampled series. Detrending is performed by solving a constrained convex optimization problem that exploits the weighted quadratic variation. Numerical results confirm the effectiveness of the approach. The algorithm is simple and favorable in terms of computational complexity, which is linear in the size of the series to detrend. This makes it suitable for long-term HRV analysis. To the best of the authors’ knowledge, it is the fastest algorithm for detrending RR series.