Experimental Research Of Adaptive Beamforming Oriented To SSI

In medical ultrasound imaging systems, beamforming design is a core issue that determines the quality of image, which also determines the effect of diagnosis. At present, the Delay-and-Sum beamforming (DAS) has been widely used in medical ultrasound imaging. However, the image has a wide main lobe and large sidelobes, which leads to poor performance on the resolution and contrast. In practice it needs to be combined with the variable aperture and apodization technologies to solve these problems. Since the different speed, different propagation characteristics and different attenuations when ultrasound translate in different tissue of humen body, if the delay time of DAS algorithm is based on a fixed speed of 1540m/s to calculate, it will lead to errors of the image, which can have an influence of the diagnose.In recent decades, the adaptive beamforming has seen broad applicaion in the area of array signal processing such as wireless communications, radar and sonar. The most common method is the adaptive minimum variance (MV) beamforming algorithms, proposed by Capon in 1969. The weighting factors at all channels can be dynamically adjusted based on the echo signal, thereby making it possible to obtain more accurate beam, improved the image quality.Based on the principles of both DAS beamforming method and adaptive beamforming method, this thesis designs an experiment to investigate the difference in image quality between these two methods. Field II is used to simulate and compare three different focus modes of synthesis, as well as compare the traditional DAS beamforming and minimum variance adaptive beamforming imaging. Supersonic Shear Imaging (SSI) is a new technique studying real-time visualization of ultrasound technology of soft tissue. The beamform characteristic is parallel-transmit-parallel-receive beamforming. In this thesis, an adaptive beamforming method based on sparse representation was reported, which could be applied to the high-frame-rate plan wave image. Then real plan-wave-approximate ultrasound data was collected using ARFI (Acoustic Radiation Force Impulse) system designed by SIAT (Shenzhen Institude of Advanced Technology) and processed in MATLAB. Fixed focus method and subsection dynamic focus mothod are realized using DAS and MV respectively.Based on the simulation and experimental results, it can be concluded that the minimum variance (MV) adaptive beamforming algorithm can get a better image than the traditional DAS beamforming both in simulation and the actual application.