Skin blood flow as the first time derivative of the temperature: spectral approach to the blood flow estimation in hands

摘要

Application of thermometry as a method of blood flow estimation in hands is limited by the differences in waveform of the skin temperature and blood flow oscillations. In addition, there is the delay time between simultaneously registered blood flow and temperature signals. We offer the method of blood flow estimation in hands based on the analysis of skin temperature dynamics. Simplification of the Shitzer's temperature dynamics model in hands allowed us to express the relation between the temperature and blood flow oscillations. As it turns out, within using model the skin blood flow is proportionally to the first time derivative of the skin temperature, in other words, the blood flow oscillations determine the rate of skin temperature alterations. We also derived relation between spectral components of the blood flow and temperature oscillations. To extract spectrum of the blood flow oscillations from temperature spectrum it is necessary using frequency dependent multiplication factor and time shift of each of temperature spectral component. In this study the skin temperature was registered by infrared camera and blood flow was estimated by photoplethysmography. First time derivation of the temperature increases correlation between processed signal and blood flow oscillations from ～0.3 to ～0.7 on average. Thus, derived relations between temperature and blood flow in fingertips provide the basis for the application of thermometry in the field of analysis of low-frequency peripheral blood flow oscillations.