特征空间和符号相干系数融合的最小方差超声波束形成  

Eigenspace-based minimum variance beamforming combined with sign coherence factor for ultrasound beamforming

作　　者：刘婷婷[1] 周浩[1] 郑音飞[1] 

机构地区：[1]浙江大学生物医学工程教育部重点实验室,杭州310027

出　　处：《声学学报》2015年第6期855-862,共8页Acta Acustica

基　　金：中央高校基本科研业务费专项资金(2014FZA5019;2015FZA5019);国家科技支撑计划(2011BAI12B02)资助

摘　　要：为了提高医学超声成像的空间分辨率,提出一种融合了特征空间最小方差与符号相干系数的波束形成方法。首先利用最小方差法计算回波数据的协方差矩阵和加权向量;然后对协方差矩阵进行特征分解得到信号子空间,并将加权向量投影到该空间上;最后计算符号相干系数,用于优化特征空间法得到的回波信号,最终获得超声成像数据。为验证算法的有效性,对医学超声成像中常用的点目标、斑目标进行仿真,对点目标仿体和人体颈动脉组织进行超声成像实验。结果表明:所提出的方法在分辨率、对比度以及稳健性等方面都优于传统的延时叠加算法、最小方差算法、特征空间最小方差法以及特征空间与相干系数融合的方法。To improve the spatial resolution of medical ultrasound imaging, a bemforming method which combines the eigenspace-based minimum variance （ESBMV） with sign coherence factor （SCF） is proposed. Firstly, minimum variance beamforming is used to obtain covariance matrix and weight vector; then the weight vector is projected onto the signal subspace constructed from the eigenstructure of covariance matrix; finally, sign coherence factor is computed and used to optimize echo data obtained from ESBMV. Simulations of point scatterers and cyst phantom and in vivo experiments based on point scatterers and common artery are used to verify the proposed method. The results show that the proposed method is better than delay and sum （DAS）, the minimum variance （MV）, ESBMV, and ESBMV combined with coherence factor （ESBMV＋CF） beamformer in the aspect of resolution, contrast and robustness.

关 键 词：最小方差法 特征空间 超声成像 相干系数 波束形成 空间分辨率 协方差矩阵 加权向量 

分 类 号：R445.1[医药卫生—影像医学与核医学]