Forsberg F, Healey A J, Leeman S, Jensen J A
Department of Medical Engineering and Physics, King's College School of Medicine and Dentistry, Dulwich Hospital, London, UK.
Phys Med Biol. 1991 Nov;36(11):1539-49. doi: 10.1088/0031-9155/36/11/013.
A consequence of employing coherent detection methods in medical ultrasound imaging systems is the occurrence of interference effects in the received echo field, which produce the speckle artefact. Speckle can severely degrade the information content of the image, and its efficient removal from ultrasound pulse-echo images is the focus of a number of research projects. Traditionally, the approach towards speckle reduction in pulse-echo images has been based on two classes of technique, either employing some form of spatial/frequency compounding or a data (image) filter. Both approaches have inherent shortcomings, and two alternative techniques are suggested here: 'local frequency diversity' and 'frequency differencing'. These algorithms deterministically identify where speckle occurs, and correct for speckle only within short, localized, corrupted segments of the A-line. This provides the potential for real-time implementation. Simulated and clinical in vivo images have been obtained, and the capabilities of the alternative speckle reduction algorithms are assessed against the more conventional approaches.
在医学超声成像系统中采用相干检测方法的一个后果是，在接收到的回波场中会出现干扰效应，从而产生散斑伪像。散斑会严重降低图像的信息含量，从超声脉冲回波图像中有效去除散斑是许多研究项目的重点。传统上，脉冲回波图像中减少散斑的方法基于两类技术，要么采用某种形式的空间/频率复合，要么采用数据（图像）滤波器。这两种方法都有固有的缺点，这里提出了两种替代技术：“局部频率分集”和“频率差分”。这些算法可以确定散斑出现的位置，并仅在A线的短的、局部的、损坏的段内校正散斑。这为实时实现提供了可能性。已经获得了模拟和临床体内图像，并将替代散斑减少算法的能力与更传统的方法进行了评估。
