Diffuse Correlation Spectroscopy Measurement And Topological Imaging Method

Blood flow at the microcirculatory level is responsible for the delivery of oxygen and nutrients to tissue and the removal of metabolic by-products.Measurement of tissue blood flow can provide crucial information for both the diagnosis and therapeutic monitoring of many vascular/cellular diseases.Blood flow velocity is an important indicator of tissue blood flow,and its non-invasive continuous detection is of great significance in clinical diagnosis and treatment.Diffuse correlation spectroscopy(DCS)is a near-infrared method for non-invasive detection of deep tissue blood flow changes.It can detect deep tissue information,and is easy to operate.It also can realize quantitative calculation of tissue blood flow.The DCS topography system can obtain the intensity temporal autocorrelation curves by multi-source and multi-exploration measurement,and then reconstruct the blood flow image in the tissue with fitting algorithm,so that it can provide tissue metabolic information and functional heterogeneity.Most current DCS imaging systems use the commercial digital correlators.These imported products are expensive and have poor flexibility(for example,for some of them the delay-time grid and the number of bits per channel are fixed).To overcome these limitations,we develop a DCS topography system based on software correlator that consists of a long coherence length laser,the fiber detection module,photon correlator and computer.The system measures the intensity temporal autocorrelation curves of twelve source channels and twelve detector channels with optical switch,and then obtains the intensity temporal autocorrelation curves of all channels.Then,the fitting algorithms of correlation curve based on correlation diffusion equation(CDE)and Monte Carlo(MC)are discussed.The constrained nonlinear optimization algorithm based on the analytical solution of CDE seeks data matching between measurement and analytical solution in semi-infinite geometry.The algorithm is simple and has average convergence speed.Nth-order linear algorithm does not seek for the solution to CDE.Instead,it combines the integral form of autocorrelation function with Nth-order Tylor polynomial to obtain the blood flow index iteratively and is more advantageous in the complex geometry.Finally,in order to evaluate the performance of the system,this paper also designs a dynamic phantom and does a series of topological imaging experiments.The experimental results show that the system can distinguish different flow rates of liquid medium,and reconstruct a two-dimensional image of liquid flow velocity distribution to locate the flow targets accurately.