Multiscale Analysis of Heart Rate Variability Using Subsymmetry and Effort-to-Compress Complexity Measures

Analysis of heart rate variability using complexity measures can enable early detection of abnormalities. However, structure and correlations that exist in the RR tachograms (cardiac inter-beat time series) at different scales are not char-acterized by conventional techniques that analyze the data at a single scale only. Recently, researchers have extended these complexity measures and applied them across multiple scales which enables better characterization and discrimination. In this work, we introduce a novel multiscale extension of a recently proposed complexity measure known as Subsymmetry (SubSym) that measures the number of sub-symmetries in the time series, and analyze the effect of scaling on the Subsymmetry of the RR intervals of healthy young and old subjects. We further compare multiscale Subsymmetry (MS-SubSym) measure with a multiscale version of Effort-To-Compress complexity measure (MS-ETC) on the same data set. It is found that both the complexity measures are able to capture the differences in the complexity values efficiently (p < 0.05) at higher scales as compared to the commonly used scale of 1.