基于光流的组织应变估计算法
Algorithm for Estimating Tissue Strain Based on Optical Flow
-
摘要: 该研究针对光流法的平移假设不能准确地描述组织压缩后的运动形式,提出了一种新的超声弹性成像算法。当探头对组织施压时假设组织的形变为仿射变换,把先验估计与光流法结合,同时估计出位移和应变分布。为了验证该算法的有效性,用仿真的超声射频回波信号比较该文算法和另外一种成像算法的成像质量,结果表明,新提出的算法在8%压缩量时,信噪比、对比度噪声比和算法时长都优于对比算法。结果证明,该文算法可以有效地估计大压缩量情况下的轴向位移和轴向应变。Abstract: Because the translation hypothesis of optical flow method can not accurately describe the form of motion after tissue compression, so we proposed a new ultrasonic elastic imaging algorithm. It was assumed that the deformation of the tissue was affine transformation when the probe was pressed to the tissue, and the displacement and strain distribution were estimated simultaneously by the optical flow method combined with the prior estimation. In order to verify the effectiveness of the algorithm, the imaging quality of the algorithm and the other imaging algorithm were compared with the simulated radio frequency echo signal. The results show that the new algorithm is higher in signal to noise ratio (SNRe), contrast to noise ratio (CNRe) and running speed than the contrast algorithm under 8% compression. The results show that the new proposed algorithm can effectively estimate axial displacement and axial strain in the case of large compression.