Motion compensated self supervised deep learning for highly accelerated 3D ultrashort Echo time pulmonary MRI.
Motion compensated self supervised deep learning for highly accelerated 3D ultrashort Echo time pulmonary MRI.
复制标题
DOI：
10.1002/mrm.29586
复制
发表时间：
2023-06
影响因子：
3.3
通讯作者：
Johnson KM
中科院分区：
文献类型：
--
作者：
Miller Z;Johnson KM
研究运动补偿、自我监督、基于模型的深度学习（MBDL）作为重建自由呼吸、3D肺超短回波时间（UTE）采集的方法。开发了一种自监督的超维MBDL架构（XD-MBDL），该架构结合呼吸状态以重建单个高质量3D图像。非刚性的，基于GPU的运动场被纳入到这个架构中，通过估计运动场从一个较低的分辨率运动分辨（XD-GRASP）迭代重建。在有和没有造影剂的肺UTE数据集上评价了运动补偿的XD-MBDL，并将其与不考虑动态呼吸状态或利用运动校正的自监督MBDL的约束重建和变体进行了比较。使用XD-MBDL重建的图像表现出改善的图像质量，如通过表观SNR、CNR和视觉评估测量的，相对于不考虑动态呼吸状态的自监督MBDL方法、XD-GRASP和最近提出的运动补偿迭代重建策略（iMoCo）。此外，相对于XD-GRASP和iMoCo，XD-MBDL减少了重建时间。开发了一种允许自监督MBDL联合收割机组合多个呼吸状态以重建单个图像的方法。将该方法与基于GPU的图像配准相结合，进一步提高了重建质量。该方法显示了从自由呼吸3D肺部UTE采集重建用户选择的呼吸相位的有希望的结果。
To investigate motion compensated, self-supervised, model based deep learning (MBDL) as a method to reconstruct free breathing, 3D Pulmonary ultrashort echo time (UTE) acquisitions. A self-supervised eXtra Dimension MBDL architecture (XD-MBDL) was developed that combined respiratory states to reconstruct a single high-quality 3D image. Non-rigid, GPU based motion fields were incorporated into this architecture by estimating motion fields from a lower resolution motion resolved (XD-GRASP) iterative reconstruction. Motion Compensated XD-MBDL was evaluated on lung UTE datasets with and without contrast and was compared to constrained reconstructions and variants of self-supervised MBDL that do not account for dynamic respiratory states or leverage motion correction. Images reconstructed using XD-MBDL demonstrate improved image quality as measured by apparent SNR, CNR and visual assessment relative to self-supervised MBDL approaches that do not account for dynamic respiratory states, XD-GRASP and a recently proposed motion compensated iterative reconstruction strategy (iMoCo). Additionally, XD-MBDL reduced reconstruction time relative to both XD-GRASP and iMoCo. A method was developed to allow self-supervised MBDL to combine multiple respiratory states to reconstruct a single image. This method was combined with GPU-based image registration to further improve reconstruction quality. This approach showed promising results reconstructing a user-selected respiratory phase from free breathing 3D pulmonary UTE acquisitions.
登录
查看更多内容
影响因子：
3.3
作者：
Sandino CM;Lai P;Vasanawala SS;Cheng JY
通讯作者：
Cheng JY
影响因子：
3.3
作者：
Hammernik K;Klatzer T;Kobler E;Recht MP;Sodickson DK;Pock T;Knoll F
通讯作者：
Knoll F
影响因子：
3.3
作者：
Buehrer, Martin;Pruessmann, Klaas P.;Kozerke, Sebastian
通讯作者：
Kozerke, Sebastian
影响因子：
3.3
作者：
Higano, Nara S.;Hahn, Andrew D.;Tkach, Jean A.;Cao, Xuefeng;Walkup, Laura L.;Thomen, Robert P.;Merhar, Stephanie L.;Kingma, Paul S.;Fain, Sean B.;Woods, Jason C.
通讯作者：
Woods, Jason C.
影响因子：
28.1
作者：
Christodoulou AG;Shaw JL;Nguyen C;Yang Q;Xie Y;Wang N;Li D
通讯作者：
Li D