CardioCurveR {CardioCurveR} | R Documentation |
CardioCurveR: Nonlinear Modeling and Preprocessing of R-R Interval Dynamics
Description
CardioCurveR provides an automated and robust framework for analyzing R-R interval (RRi) signals using advanced nonlinear modeling and preprocessing techniques. The package implements a dual-logistic model to capture both the rapid drop in RRi during exercise and the subsequent recovery phase, following the methodology described by Castillo-Aguilar et al. (2025):
Details
RRi(t) = \alpha + \frac{\beta}{1 + e^{\lambda (t-\tau)}} + \frac{-c \cdot \beta}{1 + e^{\phi (t-\tau-\delta)}}
In this model, \alpha
denotes the baseline RRi, \beta
controls the amplitude of the drop,
\lambda
and \tau
modulate the drop phase, and c
, \phi
, and \delta
govern the recovery
dynamics.
In addition to parameter estimation, CardioCurveR offers state-of-the-art signal preprocessing tools:
CardioCurveR cleans RRi signals by applying zero-phase Butterworth low-pass filtering to remove high-frequency noise while preserving the signal phase. It further employs adaptive outlier replacement, using local regression (LOESS) and robust statistics, to identify and correct ectopic beats without "chopping" dynamic signal features.
These methods ensure that the intrinsic dynamics of RRi signals are maintained, supporting accurate cardiovascular monitoring and facilitating clinical research.