Sparse Focusing High Frame Rate Algorithm Based On Compressed Sensing In Ultrasonic Super Harmonic Imaging

Cardiovascular structures include aortic circulation and myocardial microcirculation.High resolution and noninvasive imaging of them is an urgent technical problem to be solved.Ultrasound has potential applications in noninvasive,reliable and high-resolution imaging of cardiovascular structures in vivo.In this regard,the ultrasound imaging system has been built in the early stage of this project.On this basis,this paper designs focused ultrasound super harmonic imaging algorithm for the imaging system,so that the imaging system has the feasibility of noninvasive and high-resolution imaging of cardiovascular structureFirstly,the high-resolution and high contrast ultrasound imaging methods are deeply studied.Due to the micron scale characteristics of myocardial microvessels,the imaging system is required to have ultra-high resolution.Super harmonic imaging technology has the characteristics of low frequency transmission and high frequency reception,which can effectively improve the resolution;second harmonic has been widely used in clinical medical imaging,and higher harmonic has better imaging effect.In addition,synthetic aperture focused ultrasound imaging can also effectively improve the imaging resolution and contrast.In this paper,a multi-channel linear array ultrasonic probe is used to transmit low-frequency ultrasound(5MHz)and receive echo signal containing high-order harmonic components for synthetic aperture focusing imaging based on ultra harmonic ultrasonic imaging technology.Secondly,the high frame rate ultrasound imaging technology is deeply studied.Because the heart is in a state of rapid motion,it needs high sampling rate to image it.However,the traditional focused ultrasound has low sampling efficiency,and the frame rate can only reach about 30 frames / s.This paper innovatively combines synthetic aperture focused ultrasound with compressed sensing technology to reduce the focus of sampling,and then uses ADMM algorithm to complete and reconstruct the matrix in the data processing stage to improve the sampling frame rate by sparse focusing.Then,for the ultrasonic super harmonic sparse focusing imaging method proposed in this paper,a complete imaging process and related imaging algorithm are designed.Combined with GPU parallel computing technology,the CPU + GPU parallel computing architecture is adopted to improve the efficiency of data processing for the parallel computing processing steps such as filtering,beamforming and matrix completion.Finally,we design the simulation and in vivo experiments,through a large number of tests and algorithm optimization,we achieve the matrix completion of 25% sparse space-time sequence of ultrasonic signals,the average absolute error is less than 5,the structure similarity is 90%;the average signal-to-noise ratio of static simulation imaging results is 20.16,the contrast to noise ratio is 1.82,which proves the feasibility of the imaging method and related imaging algorithm designed in this paper High resolution image can be obtained;small animal cardiogram experiment tests the imaging effect of real dynamic scene,non invasively images the heart of living mice in vitro,and obtains high-resolution video data of heart,which is 4times higher than the frame rate of traditional focused ultrasound imaging,and verifies the high frame rate performance of the algorithm.