基于K空间的MR截断伪影去除算法

摘要

Objective To design a truncation artifact removal in MR images based on K-space. Methods We extract K-space high-frequency data after artifact processing,add the high-frequency data to the original K-space of artifact images in order to obtain the final complete K-space,and then get the artifact removing effect after reconstruction. Results Through simulated data and clinical picture experiments,MR truncation artifacts are improved. Conclusions The algorithm based on K-space has certain removal effect on truncation artifacts,yet relatively weak,which need to be improved.%目的：根据 MR 截断伪影产生的原理，设计了一种基于 K 空间的截断伪影去除算法。方法对伪影图像进行处理后提取 K 空间高频数据，将高频数据与原伪影图像的 K 空间融合得出最终完整的K 空间，再经过重建达到去除伪影的目的。结果在模拟数据试验和临床图片实验中，MR 截断伪影都得到改善。结论本文提出的基于 K 空间的算法对截断伪影有一定的去除效果，但图像域效果比较微弱，算法有待改进。