次方样本熵自适应加权的超声合成孔径成像算法  

Ultrasonic synthetic aperture imaging algorithm based on power sample entropy as adaptive weight

作　　者：郑驰超[1] 成娟[1] 彭虎[1] 

机构地区：[1]合肥工业大学生物医学工程系,合肥230009

出　　处：《声学学报》2017年第1期109-114,共6页Acta Acustica

基　　金：国家自然科学基金青年科学基金项目(61201060);国家自然科学基金项目(61172037)资助

摘　　要：利用相位相干系数(PCF)和广义相干系数(GCF)对波束形成后的结果进行加权,能有效提高超声成像的质量,但存在背景组织亮度降低,对比度不高,以及远处目标成像强度降低等问题。本文提出一种基于次方样本熵的合成孔径成像算法,将单个孔径发射时的低质量成像结果作为元素,根据孔径位置排列,构成空间向量。根据不同成像点对应的空间向量的随机性不同,计算每个点的空间向量的次方样本熵,并将该熵值作为权系数进行加权成像。采用FieldⅡ仿真数据成像结果表明,相比于传统的DAS算法,次方样本熵方法能够提高成像的分辨率和对比度;相比于PCF和GCF算法,次方样本熵方法能够在不损失组织背景强度的情况下,进一步改善了成像质量。Phased Coherence Factor （PCF） and General Coherence Factor （GCF） algorithms can enhance the quality of ultrasound imaging, by means of weighting on the results derived from beamforming algorithm. These weighted imaging algorithms have some drawbacks, such as that the brightness of tissue background is reduced, the contrast is not high, and imaging intensity of distant target is declined. The ultrasonic synthetic aperture imaging algorithm based on power sample entropy （PSampen） is proposed. The low-quality imaging results transmitted by a single aperture, can be treated as one element, and a spacial vector can be formed according to the aperture location information. The randomness of spacial vector that corresponds to different imaging points is different and this randomness can be characterized by PSampen, so the PSampen of each spacial vector is calculated as the adaptive weight. Field II simulation results demonstrated that the imaging contrast and resolution based on our proposed method could be effectively improved compared with the traditional DAS. Compared with the PCF and GCF, the tissue background brightness will not be lost. Besides, the quality of ultrasound image is improved.

关 键 词：自适应加权 成像算法 合成孔径 加权系数 空间向量 仿真数据 超声成像 点对应 目标成像 成像质量 

分 类 号：TB559[理学—物理]