Research On Laser Speckle Auto-Inverse Covariance For The Detection Of Blood Flow

Abnormal blood flow velocity is one of the signs of microvascular dysfunction.Therefore,continuous monitoring of the changes in blood flow velocity is of great significance for assessing microvascular function and deciphering pathological mechanisms in basic research.As a non-contact,wide-field,high temporal and spatial resolution and low-cost technology for the detection of blood flow velocity,laser speckle imaging is widely used in clinical diagnosis and treatment.However,several factors affect the accuracy of laser speckle detection of blood flow,such as coherence loss and detector noise associated with the imaging system,the form of the auto-correlation function of the electric field related to the properties of biological tissue,non-ergodic components associated with static scattering,and finite statistical sample size.To improve the accuracy of the detection of blood flow velocity,accomplish the quantitative laser speckle imaging of blood flow,and make the blood flow velocity measured in different application scenarios comparable,the above problems were discussed in this thesis.The influence of the statistical sample size on the mean of the speckle contrast was deduced by the probabilistic and statistical methods.A method for unbiased estimation of flow velocity in the range of 0.1～30 mm/s was proposed.The influence of the properties of biological tissue and the imaging system on the new method was discussed.The quantitative laser speckle auto-inverse covariance model for imaging blood flow velocity was established and verified in phantom and animal experiments.The main research contents include:(1)Aiming at the problem of inaccurate detection of blood flow velocity by laser speckle contrast method under limited statistical sample size,the quantitative relationships between the mean and signal-to-noise ratio of the speckle contrast and statistical sample size were deduced by establishing the gamma distribution model of the squared speckle contrast.The accuracy of blood flow velocity detection by the laser speckle contrast method was improved.The guidance for the trade-off between the signal-to-noise ratio and Spatiotemporal resolution of laser speckle contrast imaging was provided.(2)Aiming at the problem of inaccurate detection of slow blood flow velocity by laser speckle contrast method under limited statistical sample size,a novel method for the detection of blood flow velocity,named speckle auto-inverse covariance,was proposed.Compared with the traditional speckle contrast method,the speckle auto-inverse covariance method had unbiasedness,consistency,higher computational efficiency,higher signal-tonoise ratio,and greater linear goodness of fit.(3)Aiming at the influence of statistical sample size,imaging system parameters,and biological tissue characteristics on the accuracy of laser speckle imaging of blood flow,the effects of the static scattering,the polarization and coherence of laser light source,the magnification and aperture size,exposure time,the noise of photodetector and the formation of the auto-correlation function of the electric field on the estimation of blood flow velocity were analyzed.Their parametric forms in the laser speckle auto-inverse covariance method were derived.The quantitative model of laser speckle auto-inverse covariance imaging of blood flow was established and applied to the observation of the changes in blood flow velocity in the process of middle cerebral artery occlusion and cortical spreading depression through the intact skull of the mice.The signal-to-noise ratio of the images of blood flow velocity distribution and the accuracy of the detection of blood flow velocity was improved.Under the non-ergodic condition,the essential difference between the autocorrelation function of the intensity in the time domain and space domain was analyzed,and the necessity of spatial averaging of the results of temporal laser speckle imaging of blood flow was pointed out.