Weick Stefan, Völker Michael, Hemberger Kathrin, Meyer Cord, Ehses Philipp, Polat Bülent, Breuer Felix A, Blaimer Martin, Fink Christian, Schad Lothar R, Sauer Otto A, Flentje Michael, Jakob Peter M
Department of Radiation Oncology, University of Würzburg, Germany.
Department of Experimental Physics 5, University of Würzburg, Germany.
Magn Reson Med. 2017 Feb;77(2):787-793. doi: 10.1002/mrm.26159. Epub 2016 Mar 10.
To demonstrate that desynchronization between Cartesian k-space sampling and periodic motion in free-breathing lung MRI improves the robustness and efficiency of retrospective respiratory self-gating.
Desynchronization was accomplished by reordering the phase (k ) and partition (k ) encoding of a three-dimensional FLASH sequence according to two-dimensional, quasi-random (QR) numbers. For retrospective respiratory self-gating, the k-space center signal (DC signal) was acquired separately after each encoded k-space line. QR sampling results in a uniform distribution of k-space lines after gating. Missing lines resulting from the gating process were reconstructed using iterative GRAPPA. Volunteer measurements were performed to compare quasi-random with conventional sampling. Patient measurements were performed to demonstrate the feasibility of QR sampling in a clinical setting.
The uniformly sampled k-space after retrospective gating allows for a more stable iterative GRAPPA reconstruction and improved ghost artifact reduction compared with conventional sampling. It is shown that this stability can either be used to reduce the total scan time or to reconstruct artifact-free data sets in different respiratory phases, both resulting in an improved efficiency of retrospective respiratory self-gating.
QR sampling leads to desynchronization between repeated data acquisition and periodic respiratory motion. This results in an improved motion artifact reduction in shorter scan time. Magn Reson Med 77:787-793, 2017. © 2016 International Society for Magnetic Resonance in Medicine.
证明在自由呼吸肺部磁共振成像中,笛卡尔k空间采样与周期性运动之间的失步可提高回顾性呼吸自门控的稳健性和效率。
通过根据二维准随机(QR)数重新排序三维快速低角度激发(FLASH)序列的相位(k)和分区(k)编码来实现失步。对于回顾性呼吸自门控,在每条编码的k空间线之后分别采集k空间中心信号(直流信号)。QR采样导致门控后k空间线的均匀分布。使用迭代广义自校准部分并行采集(GRAPPA)重建门控过程中产生的缺失线。进行志愿者测量以比较准随机采样与传统采样。进行患者测量以证明QR采样在临床环境中的可行性。
与传统采样相比,回顾性门控后均匀采样的k空间允许更稳定的迭代GRAPPA重建并改善了鬼影伪影的减少。结果表明,这种稳定性可用于减少总扫描时间或重建不同呼吸相位的无伪影数据集,两者均提高了回顾性呼吸自门控的效率。
QR采样导致重复数据采集与周期性呼吸运动之间的失步。这导致在更短的扫描时间内改善运动伪影的减少。《磁共振医学》77:787 - 793, 2017。© 2016国际磁共振医学学会。