针对合成孔径连续声束合成(SASB)算法渡越时间存在误差,会带来严重伪影的问题,提出线性约束最小方差(MV)准则和SASB算法相结合的内镜超声成像(SASB-MV)算法。该算法首先对各个虚拟源点进行固定聚焦,获取成像点的超声回波信号矢量,根据成像点的超声回波信号矢量构建样本协方差矩阵;然后由线性约束最小方差准则构建权值矢量;最后,通过权值矢量和超声回波信号矢量得到成像点的高分辨率回波信号值。在Filed Ⅱ中进行散射点仿真实验和圆形暗斑仿真实验。实验结果表明本文算法的横向分辨率相比动态接收聚焦算法(DRF)和SASB算法分别提高了59.1%和31.2%,同时图像对比度分别提高了44.7%和16.2%。

Considering the deviation in the time-of-flight of synthetic aperture sequential beamforming (SASB), which causes serious artifacts, this study proposed an endoscopic ultrasound-imaging algorithm (SASB-MV) combining the linear constrained minimum variance (MV) criterion with the SASB algorithm. The algorithm first focused on each virtual source point, obtained the ultrasonic echo signal vector of the imaging point, and constructed the sample covariance matrix based on the ultrasonic echo signal vector of the imaging point. The algorithm then constructed the weight vector based on the linear constraint MV criterion and finally obtained the high-resolution echo signal value of the imaging point through the weight and ultrasonic echo signal vectors. Moreover, this study conducted simulation experiments of scattering point and circular dark spot in Filed II. Experimental results show that compared with the dynamic receive focusing (DRF) and SASB algorithms, the lateral resolution of the proposed algorithm improved by 59.1% and 31.2%, and the contrast improved by 44.68% and 16.23%, respectively.