Synthetic CT generation from non-attenuation corrected PET images for whole-body PET imaging.

Synthetic CT generation from non-attenuation corrected PET images for whole-body PET imaging.
复制标题

DOI:
10.1088/1361-6560/ab4eb7
复制
发表时间:
2019-11-04
影响因子:
3.5
通讯作者:
Yang X
Yang X
中科院分区:
工程技术2区
文献类型:
--
作者:
Dong X;Wang T;Lei Y;Higgins K;Liu T;Curran WJ;Mao H;Nye JA;Yang X

文献摘要

参考文献

被引文献

相似文献

PET/MRI 的衰减校正 (AC) 面临的挑战包括扫描间运动、图像伪影(例如截断和失真)以及结构体素强度到 PET mu-map 值的错误转换。我们提出了一种基于深度学习的方法,从非衰减校正 PET (NAC PET) 图像中导出合成 CT (sCT) 图像,用于全身 PET/MRI 成像上的 AC。采用 3D 循环一致生成对抗网络 (CycleGAN) 框架从 NAC PET 合成 CT 图像。该方法学习一种转换,最大限度地减少由 NAC PET 生成的 sCT 与真实 CT 之间的差异。它还学习逆变换,使得从 sCT 生成的循环 NAC PET 图像接近真实的 NAC PET 图像。还利用自注意力策略来识别信息最丰富的组件并减轻噪声的干扰。我们对总共119台全身PET/CT进行了回顾性研究,其中80台用于训练,39台用于测试和评估。使用该方法生成的全身 sCT 图像与真实 CT 图像非常相似,并且在软组织、肺和骨组织上显示出良好的对比度。 sCT 相对于真实 CT 的平均绝对误差 (MAE) 小于 110 HU。使用sCT进行全身PET AC,PET定量的平均误差小于1%,归一化均方误差(NMSE)小于1.4%。全身平均归一化互相关接近于1,PSNR大于42dB。我们提出了一种基于深度学习的方法,从全身 NAC PET 生成 sCT,用于 PET AC。使用所提出的方法生成的 sCT 在定性和定量上都与真实 CT 图像非常相似,并且在缺乏结构信息的情况下展示了全身 PET AC 的巨大潜力。
Attenuation correction (AC) of PET/MRI faces challenges including inter-scan motion, image artifacts such as truncation and distortion, and erroneous transformation of structural voxel-intensities to PET mu-map values. We propose a deep-learning-based method to derive synthetic CT (sCT) images from non-attenuation corrected PET (NAC PET) images for AC on whole-body PET/MRI imaging. A 3D cycle-consistent generative adversarial networks (CycleGAN) framework was employed to synthesize CT images from NAC PET. The method learns a transformation that minimizes the difference between sCT, generated from NAC PET, and true CT. It also learns an inverse transformation such that cycle NAC PET image generated from the sCT is close to true NAC PET image. A self-attention strategy was also utilized to identify the most informative component and mitigate the disturbance of noise. We conducted a retrospective study on a total of 119 sets of whole-body PET/CT, with 80 sets for training and 39 sets for testing and evaluation. The whole-body sCT images generated with proposed method demonstrate great resemblance to true CT images, and show good contrast on soft tissue, lung and bony tissues. The mean absolute error (MAE) of sCT over true CT is less than 110 HU. Using sCT for whole-body PET AC, the mean error of PET quantification is less than 1% and normalized mean square error (NMSE) is less than 1.4%. Average normalized cross correlation on whole body is close to one, and PSNR is larger than 42 dB. We proposed a deep learning-based approach to generate sCT from whole-body NAC PET for PET AC. sCT generated with proposed method shows great similarity to true CT images both qualitatively and quantitatively, and demonstrates great potential for whole-body PET AC in the absence of structural information.
DOI: 10.1186/s40658-018-0225-8
发表时间: 2018-11-12
期刊: EJNMMI physics
影响因子: 4
作者:
Liu F;Jang H;Kijowski R;Zhao G;Bradshaw T;McMillan AB
通讯作者: McMillan AB
DOI: 10.1002/mp.13656
发表时间: 2019-09
期刊: Medical physics
影响因子: 3.8
作者:
Harms J;Lei Y;Wang T;Zhang R;Zhou J;Tang X;Curran WJ;Liu T;Yang X
通讯作者: Yang X
DOI: 10.2967/jnumed.107.049353
发表时间: 2008-11-01
影响因子: 9.3
作者:
Hofmann, Matthias;Steinke, Florian;Pichler, Bernd J.
通讯作者: Pichler, Bernd J.
DOI: 10.1118/1.4941014
发表时间: 2016-03-01
期刊: MEDICAL PHYSICS
影响因子: 3.8
作者:
Mehranian, Abolfazl;Arabi, Hossein;Zaidi, Habib
通讯作者: Zaidi, Habib
DOI: 10.1109/tmi.2013.2272660
发表时间: 2013-11-01
影响因子: 10.6
作者:
Schramm, G.;Langner, J.;van den Hoff, J.
通讯作者: van den Hoff, J.