Research On Blood Flow Imaging Method Based On Laser Speckle Contrast Analysis

The blood in the organism directly participates in the process of material energy and information transmission between organs,tissues and cells.Blood flow speed,blood vessel morphology,and blood vessel density can reflect the pathological conditions of certain chronic diseases.Therefore,it is very important of detection of diseases,diagnosis and monitoring of the treatment process.However,some blood vessels exist in the epithelial layer and organ tissues and the diameter of which is small,and the blood flow in the small blood vessels is slow,which brings great difficulties to disease diagnosis and monitoring during treatment.Therefore,it is of great significance to study accurate and efficient blood flow index monitoring methods.Laser Speckle Contrast Analysis(LASCA)is a technology that can provide realtime imaging of blood vessel structure and blood flow function,and can quickly image without direct contact with objects.However,there are still two problems with this technology.First,the interference of imaging system noise and static scattering of skin tissue are serious,which makes the calculated contrast value smaller in dynamic range and is not conducive to image visualization.Second,the imaging parameters,such as matching exposure time and system gain,are different for different experimental objects,so it is necessary to explore the influence of system parameters on image imaging quality.In order to solve the above problems,this thesis proposes a method of imaging for the blood vessel structure and the relative velocity of blood flow based on LASCA.The method is divided into three parts,i.e.image preprocessing,blood flow imaging and image enhancement.In the image preprocessing,a temporal-spatio singular value decomposition filter(TSSVD)is proposed to eliminate the influence of skin tissue scattering on blood flow images.Then the time-domain laser speckle contrast analysis method was used to image the blood flow.Finally,a combination of anisotropic diffusion coupling filter,adaptive block-matching and 3D filtering and high-low cap transform was used to connect the vascular edge to further filter the background noise and improve the image contrast.The above method is referred to as TSSVD-TLASCA algorithm.The reliability of the speckle acquisition system and the feasibility of the TSSVD-TLASCA algorithm are verified by numerical simulation and simulation experiments.It is proved that the algorithm can effectively suppress the influence of background scattering on imaging and improve the contrast between vascular area and skin background.It enlarges the dynamic range of contrast ratio and improves the image quality.Finally,for the blood flow imaging problem of human fingers,the exposure time and system gain coefficient of the system in the speckle acquisition system,as well as the related parameters in the TSSVD-TLASCA algorithm(the number of filtered principal components and the number of batch processing frames)are explored,and the optimal range of imaging parameters is found.