基于迭代修正法的CT金属伪影消除

Reduction of Metal Artifacts in Computed Tomography Using Iterative Correction Method

人体内的金属植入物会导致CT重建图像中出现金属伪影。传统的方法利用插值技术替换掉原始投影数据中受金属物污染的数据,以此来达到去除金属伪影的目的,但是容易产生次生伪影。为了解决传统方法的缺陷,提出了一种基于迭代修正的金属伪影消除方法。首先,利用插值修正法获取到一张预处理图像;然后,以该预处理图像为基础进行迭代修正,每一次迭代过程中利用全变分滤波来优化图像质量;经过多次迭代后得到最终图像。实验表明,相比于传统的插值修正法,提出的方法能够更加有效地去除金属伪影,同时较好地抑制次生伪影的产生。

A computed tomography(CT) image including metallic objects suffers from metal artifacts.onventional methods for metal artifacts reduction(MAR) replace the metal-affected data by interpolation in the sinogram domain.However,they often lead to secondary artifacts.To overcome this limitation,an iterative correction method for MAR is proposed.In this method,an interpolation-based MAR method is first used to obtain an pre-corrected image.Then this pre-corrected image is inputted to an iterative correction operation as a base.At each iteration,total variation(TV) based filter is applied to improve the image quality.After several iterations,final corrected image is obtained.It is proved that the proposed method provides noticeably better performance in reducing metal artifacts and preventing the introduction of secondary artifacts compared with conventional interpolation-based methods.