【摘要】：為了提高醫(yī)學超聲成像的空間分辨率,提出一種融合了特征空間最小方差與符號相干系數(shù)的波束形成方法。首先利用最小方差法計算回波數(shù)據(jù)的協(xié)方差矩陣和加權(quán)向量;然后對協(xié)方差矩陣進行特征分解得到信號子空間,并將加權(quán)向量投影到該空間上;最后計算符號相干系數(shù),用于優(yōu)化特征空間法得到的回波信號,最終獲得超聲成像數(shù)據(jù)。為驗證算法的有效性,對醫(yī)學超聲成像中常用的點目標、斑目標進行仿真,對點目標仿體和人體頸動脈組織進行超聲成像實驗。結(jié)果表明:所提出的方法在分辨率、對比度以及穩(wěn)健性等方面都優(yōu)于傳統(tǒng)的延時疊加算法、最小方差算法、特征空間最小方差法以及特征空間與相干系數(shù)融合的方法。
[Abstract]:In order to improve the spatial resolution of medical ultrasonic imaging, a beamforming method combining the minimum variance of feature space and symbolic coherence coefficient is proposed. Firstly, the covariance matrix and weighted vector of echo data are calculated by the method of minimum variance, then the signal subspace is obtained by eigenfactorization of the covariance matrix, and the weighted vector is projected onto the space. Finally, the symbolic coherence coefficient is calculated to optimize the echo signal obtained by the eigenspace method, and finally the ultrasonic imaging data are obtained. In order to verify the effectiveness of the algorithm, the point targets and spot targets commonly used in medical ultrasound imaging were simulated, and ultrasound imaging experiments were carried out on the point target imitating body and human carotid artery tissue. The results show that the proposed method is superior to the traditional delay superposition algorithm, the minimum variance algorithm, the feature space minimum variance method and the fusion method of feature space and coherence coefficient in terms of resolution, contrast and robustness.
【作者單位】： 浙江大學生物醫(yī)學工程教育部重點實驗室;
【基金】：中央高�；究蒲袠I(yè)務(wù)費專項資金(2014FZA5019,2015FZA5019) 國家科技支撐計劃(2011BAI12B02)資助
【分類號】：R445.1

【參考文獻】

相關(guān)期刊論文 前2條

1 吳文燾;蒲杰;呂q，

本文編號：2334047