基于绝对离差统计系数加权的超声成像方法

An adaptive weighted ultrasound imaging method based on absolute dispersion statistical factor

摘要:

目的 为了改善超声成像设备的图像质量，提高超声图像的对比度和分辨率，增强噪声抑制能力。方法 首先将超声探头接收到的回波信号进行延迟处理，并将延迟后的数据作为成像矢量进行存储，然后计算该向量的平均值，得到绝对偏差向量。根据偏差设定阈值进行统计分析，推导出绝对偏差统计系数。然后将该系数与原始成像结果相乘，得到基于绝对偏差统计系数加权的超声图像。结果 与传统采用延时叠加（DAS）波束形成方式进行成像的算法相比，该算法横向半峰值宽度（FWHM）明显减小，噪声对比度（CNR）明显增加。与基于相干系数（CF）算法相比，横向半峰值宽度（FWHM）略有减小，对比度（CR）和背景信噪比（sSNR）均有所增加。与基于广义相干系数（GCF）算法相比，横向FWHM略有减小，CNR有所增加。结论 基于绝对离差统计系数加权的算法计算量简单，易于实现，且可抑制较强的噪声，在一定程度上提高了超声图像的对比度和分辨率，有效改善背景组织的成像质量。

Abstract:

Objective To improve the image quality of ultrasound imaging equipment, enhance the contrast and resolution of ultrasound images, and enhance the ability to suppress noise.Methods Firstly, we delayed the echo signal received by the ultrasonic probe and stored the post-delay data as an imaging vector. Subsequently, we calculated the average value of the vector to obtain the absolute deviation vector. Based on the deviation, we set a threshold for statistical analysis and derived the absolute deviation statistical coefficient. We then multiplied this coefficient with the original imaging result to obtain an ultrasound image weighted based on the absolute deviation statistical factor.Results Compared with traditional algorithms that use delay-and-sum (DAS) beamforming for imaging, this algorithm significantly reduces the lateral peak full width at half maximum (FWHM) and increases the contrast-to-noise ratio (CNR). Compared with the coherence factor (CF) based algorithm, the lateral FWHM is reduced, while contrast ratio (CR) and speckle signal-to-noise ratio (sSNR) are both increased. Compared with the generalized coherence factor (GCF) based algorithm, the lateral FWHM is reduced and CNR is increased.Conclusion The algorithm based on absolute dispersion statistical factor weighting has a simple computational complexity, is easy to implement, and can suppress strong noise, which to some extent improves the contrast and resolution of ultrasound images and effectively improves the imaging quality of background tissues.