Wavelet analysis of skin temperature fluctuations: a novel method of microcirculation diagnostics.


Microvascular tone assessment as implemented in the Microtest device is based on the spectral analysis of temperature oscillations on the skin surface. These low-amplitude (~ 0.01 °C) fluctuations occur due to the periodic changes of surface blood vessels’ tone and correlate with them in amplitude. Within the skin blood flow oscillation range, five sub-bands correspond to the different factors involved in the regulation of vascular tone: a pulse wave (0.5-2 Hz), a respiration wave (0.14-0.5 Hz), myogenic oscillations (0,05-0.14 Hz), neurogenic activity (0.02-0.05 Hz) and the endothelial mechanisms (0.0095-0.02 Hz). The amplitudes of the high-frequency temperature oscillations (the respiratory and the pulse waves) on the surface of the skin are comparable to thermal noise, and cannot be reliably measured, but the other sub-bands offer valuable insight into the mechanisms of vascular tone regulation by myogenic, neurogenic and endothelial activity.

In the studies of the vascular system reactions to external stimuli via functional tests it is necessary to monitor the changes in the spectral content of the detected signal, in particular, the changes occurring within the timeframes comparable with the period of oscillation. Such mathematical problems go beyond the traditional spectral analysis, but are handled well by the methods of wavelet analysis, which, among other applications, are used for analysis of laser Doppler flowmetry (LDF) data. The main advantage of wavelet analysis over other spectroscopic methods is the ability to obtain information about the frequency and time characteristics of the signal at the same time. A characteristic feature of the processed signals is their short duration – at the frequency of interest only a few fluctuations take can place during the observation. This requires special care in the analysis of the spectral characteristics, taking into account the boundary effects, and makes it necessary to use special algorithms (adaptive wavelets). The Microtest software uses wavelet filtering based on adaptive wavelet analysis that enables processing of relatively short duration data series, or fragments thereof.

Comparison of the data obtained by Microtest with the results from laser Doppler flowmetry (Moor Instr.) showed a high correlation (0.7) of the LDF and the temperature signals within the frequency range of 0.1 - 0.01 Hz corresponding to the myogenic, neurogenic and endothelial mechanisms of vascular tone regulation.


ЛДФ-грамма и колебания кожной температуры в частотном диапазоне 0.1...0.01Гц

(P. Frick et al. Biomed Signal Proces 08/2015; 21)