ABSTRACT
Network methods have been successfully used to capture and represent properties of multilevel complex man-made systems and living organisms. The use of network representations in the characterization of time series complexity is a relatively new but quickly developing branch of time series analysis. The most direct method is to map a time series into a graph in which the vertices represent signal values, while edges link values that are consecutive in a signal. Visualization of a transition network is challenging because usually a transition network consists of many vertices which are densely, often completely, interconnected. The plot of such a network may be barely readable. Adjacency matrixes and transition matrixes are standard representations of any network. Network structure methods are able to visualize, describe, and differentiate heart rate dynamics in healthy young subjects and HTX patients. The resulting plots can be considered as an alternative way of assessing heart rate variability.
