Deep learning-based velocity antialiasing of 4D-flow MRI.

Deep learning-based velocity antialiasing of 4D-flow MRI.
复制标题

基于深度学习的4D-flow MRI速度抗锯齿。

DOI:
10.1002/mrm.29205
复制
发表时间:
2022-07
影响因子:
3.3
通讯作者:
Markl, Michael
Markl, Michael
中科院分区:
医学3区
文献类型:
--
作者:
Berhane, Haben;Scott, Michael B.;Barker, Alex J.;McCarthy, Patrick;Avery, Ryan;Allen, Brad;Malaisrie, Chris;Robinson, Joshua D.;Rigsby, Cynthia K.;Markl, Michael

文献摘要

参考文献

被引文献

相似文献

开发一种卷积神经网络(CNN),用于4D-flow MRI中速度混叠的稳健和快速校正。 本研究纳入了667例成人受试者,其主动脉4D-flow MRI数据存在速度混叠(n = 362)和无速度混叠(n = 305)。此外,10名对照组接受背靠背4D血流扫描,速度编码灵敏度(vencs)在60、100和175 cm/s时发生全身变化。无混叠数据集用于模拟速度混叠,方法是将venc降低到原始值的40%-70%,同时使用地面实况定位所有混叠体素(153个训练,152个测试)。152个模拟和362个现有的混叠数据集用于测试,并与传统的速度抗混叠算法进行了比较。计算Dice分数以量化CNN性能。对于对照组,venc 175 cm/s扫描被用作基础事实,并与CNN校正的venc 60和100 cm/s数据集进行比较。 与传统算法的162 ± 14 s相比,CNN需要176 ± 30 s来执行。与传统算法相比,CNN在模拟数据上表现出了优异的性能(Dice评分CNN的中位数范围:[0.89-0.99],常规算法:[0.84-0.94],p < 0.001,在所有模拟vencs中),并在现有速度混叠数据集中检测到更多混叠体素(检测到的CNN中值:159个体素[31-605],常规算法:65 [7-417],p < 0.001)。对于对照组,CNN显示venc = 60 cm/s和100 cm/s时的Dice评分分别为0.98 [0.95-0.99]和0.96 [0.87-0.99],而流量比较显示一致性为中度至极好。 深度学习在4D-flow MRI中实现了快速和强大的速度抗混叠。
To develop a convolutional neural network (CNN) for the robust and fast correction of velocity aliasing in 4D‐flow MRI. This study included 667 adult subjects with aortic 4D‐flow MRI data with existing velocity aliasing (n = 362) and no velocity aliasing (n = 305). Additionally, 10 controls received back‐to‐back 4D‐flow scans with systemically varied velocity‐encoding sensitivity (vencs) at 60, 100, and 175 cm/s. The no‐aliasing data sets were used to simulate velocity aliasing by reducing the venc to 40%–70% of the original, alongside a ground truth locating all aliased voxels (153 training, 152 testing). The 152 simulated and 362 existing aliasing data sets were used for testing and compared with a conventional velocity antialiasing algorithm. Dice scores were calculated to quantify CNN performance. For controls, the venc 175‐cm/s scans were used as the ground truth and compared with the CNN‐corrected venc 60 and 100 cm/s data sets The CNN required 176 ± 30 s to perform compared with 162 ± 14 s for the conventional algorithm. The CNN showed excellent performance for the simulated data compared with the conventional algorithm (median range of Dice scores CNN: [0.89–0.99], conventional algorithm: [0.84–0.94], p < 0.001, across all simulated vencs) and detected more aliased voxels in existing velocity aliasing data sets (median detected CNN: 159 voxels [31–605], conventional algorithm: 65 [7–417], p < 0.001). For controls, the CNN showed Dice scores of 0.98 [0.95–0.99] and 0.96 [0.87–0.99] for venc = 60 cm/s and 100 cm/s, respectively, while flow comparisons showed moderate‐excellent agreement. Deep learning enabled fast and robust velocity anti‐aliasing in 4D‐flow MRI.
DOI: 10.1148/rg.2019180091
发表时间: 2019-05-01
期刊: RADIOGRAPHICS
影响因子: 5.5
作者:
Azarine, Arshid;Garcon, Philippe;Zins, Marc
通讯作者: Zins, Marc
DOI: 10.1002/jmri.20871
发表时间: 2007-04-01
影响因子: 4.4
作者:
Markl, Michael;Harloff, Andreas;Frydrychowicz, Alex
通讯作者: Frydrychowicz, Alex
DOI: 10.1002/mrm.1910330517
发表时间: 1995-05-01
影响因子: 3.3
作者:
WANG, Y;RIEDERER, SJ;EHMAN, RL
通讯作者: EHMAN, RL
DOI: 10.1002/jmri.25045
发表时间: 2016-04-01
影响因子: 4.4
作者:
Loecher, Michael;Schrauben, Eric;Wieben, Oliver
通讯作者: Wieben, Oliver
DOI: 10.1016/j.media.2019.101537
发表时间: 2019-12-01
影响因子: 10.9
作者:
Zhuang, Xiahai;Li, Lei;Yang, Guang
通讯作者: Yang, Guang