Assessment of Cardiovascular Dynamics Using Periodicity Attributes Derived from Peripheral Blood Flow Signals

A periodic process can be characterized in terms of three periodicity (or p-) attributes: the periodicity (or period-length), the periodic wave-shape or pattern and the wave-magnitude or the scaling factor; all three attributes can be time varying in a real-life situation. In this report, we hypothesize that an analysis of the dynamics underlying a nearly periodic physiological process, such as appearing in a rhythmic blood wave pattern, can be quantified in terms of the dynamics of its periodicity attributes. This report analyzes data obtained from archival studies in which the photo-plethysmograph signal (PPS) is recorded from the finger. Each specific blood wave signal is decomposed into a regular component, which is nearly periodic and an irregular residual process. The dynamics of the PPS p-attributes of the regular part are analyzed individually as well as collectively to assess the general cardiovascular state. A new class of surrogate series based on the shuffling of the p-attributes is proposed to detect the nonlinear determinism in the PPS. The dynamics is further studied by mapping the variations of the p-attributes in a novel p-space, defined by the three orthogonal periodicity-attribute components; each point in the p-space represents one nearly periodic segment. Novel complexity measures based on global and temporal variations 0 dynamics in the p-space are proposed. A correlation is explored between the complexity measures derived from the p-space mapping of PPS that closely matches the cardiovascular state of a typical human subject. The mathematical algorithms derived from a simple blood flow wave pattern can be easily applied for assessing other physiologic signals in the cardiovascular system obtained during perturbations caused by dynamic exercise, thermal stress, and potentially high terrestrial physiologic effects during hypobaric stress.