Attractor Reconstruction Analysis for Blood Flow Signals

Attractor reconstruction analysis has been previously used to determine changes in the shape and variability of fairly periodic signals such as arterial blood pressure signals and electroencephalogram signals, providing a two-dimensional attractor with features like density and symmetry. Since BF signals are fairly periodic and quasi-stationary, we set out to investigate whether attractor reconstruction method could be applied in signals derived from the microvascular perfusion. We describe the basis and the implementation of attractor reconstruction analysis of the microvascular blood flux (BF) signals recorded from the skin of 15 healthy male volunteers, age 29.2 ± 8.1y (mean ± SD). The efficacy of attractor reconstruction analysis (ARA) as a potential method of identifying changes in the microvascular function is evaluated in two haemodynamic steady states, at 33°C, and during warming at 43°C to generate a local thermal hyperaemia (LTH). Our findings show a significant drop of the maximal density derived from the ARA, during increased flow and that there was good discrimination of the blood flow signals between the two haemodynamic steady states, having good classification accuracy (80%). This study shows that ARA of BF signals can identify different microvascular functional states and thus has a potential for the clinical assessment and diagnosis of pathophysiological condition.