融合空时符号相干和特征空间波束形成的超声成像

Ultrasound imaging algorithm combining eigenspace-based beamforming and spatio-temporal sign coherence factor

为了进一步提高超声成像的质量,提出融合特征空间最小波束形成和空时符号相干系数的成像方法。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信号子空间,得到特征空间方法的加权向量,同时采用空时平滑方法计算符号相干系数;最后用空时符号相干系数作为加权系数对特征空间最小方差波束形成的结果进行优化。为了验证算法的有效性,对医学成像上常用的点目标和斑目标进行成像。仿真实验结果表明,与特征空间最小方差算法和融合特征空间与相干系数的算法相比,所提方法提高了对比度和稳健性,其代价是略微降低了成像分辨率。

An imaging method combining eigenspace-based minimum variance（ESBMV）beamforming and spatio-temporal sign coherence factor（STSCF） is proposed to improve the quality of ultrasound imaging. The minimum variance beamforming method is used to calculate the covariance matrix and weight vector of the echo data. The feature decomposition is performed for the covariance matrix to obtain the signal subspace,and then the weight vector is projected into the signal subspace to obtain the weight vector of the eigenspace method. The spatio-temporal smoothing method is used to calculate the sign coherence factor.The spatio-temporal sign coherence factor is deemed as the weighting coefficient to optimize the results of eigenspace-based minimum variance beamforming. The point scatters and cyst phantom commonly used in medical imaging are imaged to verify the validity of the algorithm. The simulation results show that,in comparison with the eigenspace-based minimum variance algorithm and the algorithm combining eigenspace and coherence factor,the proposed algorithm can improve the contrast and robustness,whose shortcoming is to reduce the imaging resolution slightly.