Blood Flow Estimation From DCS Measurements Using a Frequency-Domain Approach in Experimental Animal Models

Diffuse correlation spectroscopy (DCS) is a relatively new noninvasive methodology that provides blood flow (BF)-related measurements from tissue in a safe, continuous, and portable way. The BF index (<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${\text {BF}}_{i}$ </tex-math></inline-formula>) calculated through the normalized temporal intensity autocorrelation function of scattered light as measured by DCS has been proven to be proportional to the absolute BF. This time-domain method requires knowledge of the optical properties (absorption and scattering coefficients) of the tissue medium for reliable <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${\text {BF}}_{i}$ </tex-math></inline-formula> estimation which are either assumed to be static and known a priori or require another device with the capability to continuously measure and simultaneously feed them into the <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${\text {BF}}_{i}$ </tex-math></inline-formula> estimation pipeline of the DCS system. In this study, we propose a novel approach to calculate a BF-related parameter using frequency-domain information obtained from the power spectral density (PSD) of DCS measurements based on its observed power-law behavior, as an alternative to the traditional time-domain methods. We aimed to systematically compare the <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${\text {BF}}_{i}$ </tex-math></inline-formula> estimations obtained via the proposed frequency-domain method and the traditional time-domain approach under different scenarios of animal experiments, modeling hemorrhagic shock, hypoxia, and sham control. Our results demonstrated that the <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${\text {BF}}_{i}$ </tex-math></inline-formula> estimates obtained by the two methods followed the same behavior and hence the proposed frequency-domain feature is a viable metric of actual BF as would be desired. Notably, the frequency-domain approach not only reduced the variance in the <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${\text {BF}}_{i}$ </tex-math></inline-formula> estimates but also eliminated the need for the knowledge of optical properties of the tissue as in the time-domain approach.