Using synthetic data generation to train a cardiac motion tag tracking neural network.

Using synthetic data generation to train a cardiac motion tag tracking neural network.
复制标题

DOI:
10.1016/j.media.2021.102223
复制
发表时间:
2021-12
影响因子:
10.9
通讯作者:
Ennis DB
Ennis DB
中科院分区:
工程技术1区
文献类型:
--
作者:
Loecher M;Perotti LE;Ennis DB

文献摘要

参考文献

被引文献

相似文献

开发并验证了一种基于 CNN 的心脏 MRI 标签跟踪方法。创建了一个合成数据模拟器,用于使用自然图像、布洛赫方程模拟、广泛的组织特性和编程的地面真实运动来生成大量训练数据。使用分析变形心脏模型和具有手动跟踪参考运动路径的体内数据对该方法进行了验证。在分析模型中,研究了相对于 SNR 的误差,并且在 SNR>10(位移误差 <0.3 mm)时看到了准确的结果。体内标签位置(平均位移差 = −0.02 像素,95% CI [−0.73, 0.69])和计算的心脏圆周应变(平均差 = 0.006,95% CI [−0.012, 0.024])具有极好的一致性。使用经过合成数据训练的 CNN 进行自动标签跟踪既准确又精确。
A CNN based method for cardiac MRI tag tracking was developed and validated. A synthetic data simulator was created to generate large amounts of training data using natural images, a Bloch equation simulation, a broad range of tissue properties, and programmed ground-truth motion. The method was validated using both an analytical deforming cardiac phantom and in vivo data with manually tracked reference motion paths. In the analytical phantom, error was investigated relative to SNR, and accurate results were seen for SNR>10 (displacement error <0.3 mm). Excellent agreement was seen in vivo for tag locations (mean displacement difference = −0.02 pixels, 95% CI [−0.73, 0.69]) and calculated cardiac circumferential strain (mean difference = 0.006, 95% CI [−0.012, 0.024]). Automated tag tracking with a CNN trained on synthetic data is both accurate and precise.
DOI: 10.1148/radiology.214.2.r00fe17453
发表时间: 2000-02-01
期刊: RADIOLOGY
影响因子: 19.7
作者:
Moore, CC;Lugo-Olivieri, CH;Zerhouni, EA
通讯作者: Zerhouni, EA
DOI: 10.1109/tmi.2020.2972616
发表时间: 2020-07-01
影响因子: 10.6
作者:
Fechter, Tobias;Baltas, Dimos
通讯作者: Baltas, Dimos
DOI: 10.1007/s10741-017-9621-8
发表时间: 2017-07
影响因子: 4.6
作者:
Scatteia A;Baritussio A;Bucciarelli-Ducci C
通讯作者: Bucciarelli-Ducci C
DOI: 10.1016/j.cviu.2017.12.002
发表时间: 2018-02-01
影响因子: 4.5
作者:
Barbosa, Igor Barros;Cristani, Marco;Theoharis, Theoharis
通讯作者: Theoharis, Theoharis
DOI: 10.1016/j.media.2014.04.007
发表时间: 2014-12-01
影响因子: 10.9
作者:
Heimann, Tobias;Mountney, Peter;Ionasec, Razvan
通讯作者: Ionasec, Razvan