Department of Biomedical Engineering, Stony Brook University, Stony Brook, New York.
Department of Psychiatry, Stony Brook University Medical Center, Stony Brook, New York.
J Nucl Med. 2019 Apr;60(4):555-560. doi: 10.2967/jnumed.118.214320. Epub 2018 Aug 30.
Attenuation correction is a notable challenge associated with simultaneous PET/MRI, particularly in neuroimaging, where sharp boundaries between air and bone volumes exist. This challenge leads to concerns about the visual and, more specifically, quantitative accuracy of PET reconstructions for data obtained with PET/MRI. Recently developed techniques can synthesize attenuation maps using only MRI data and are likely adequate for clinical use; however, little work has been conducted to assess their suitability for the dynamic PET studies frequently used in research to derive physiologic information such as the binding potential of neuroreceptors in a region. At the same time, existing PET/MRI attenuation correction methods are predicated on synthesizing CT data, which is not ideal, as CT data are acquired with much lower-energy photons than PET data and thus do not optimally reflect the PET attenuation map. We trained a convolutional neural network to generate patient-specific transmission data from T1-weighted MRI. Using the trained network, we generated transmission data for a testing set comprising 11 subjects scanned with C-labeled -[2-]4-(2-methoxyphenyl)-1-piperazinyl]ethyl]--(2-pyridinyl)cyclohexanecarboxamide) (C-WAY-100635) and 10 subjects scanned with C-labeled 3-amino-4-(2-dimethylaminomethyl-phenylsulfanyl)benzonitrile (C-DASB). We assessed both static and dynamic reconstructions. For dynamic PET data, we report differences in both the nondisplaceable and the free binding potential for C-WAY-100635 and distribution volume for C-DASB. The mean bias for generated transmission data was -1.06% ± 0.81%. Global biases in static PET uptake were -0.49% ± 1.7%, and -1.52% ± 0.73% for C-WAY-100635 and C-DASB, respectively. Our neural network approach is capable of synthesizing patient-specific transmission data with sufficient accuracy for both static and dynamic PET studies.
衰减校正对于同时进行的 PET/MRI 是一个显著的挑战，特别是在神经影像学中，存在空气和骨体积之间的明显边界。这一挑战导致人们对 PET 重建的视觉效果，更具体地说是定量准确性产生了担忧，因为这些重建是基于 PET/MRI 获得的数据。最近开发的技术可以仅使用 MRI 数据来合成衰减图，这些技术可能足以满足临床应用的需求；然而，几乎没有工作评估这些技术对于在研究中经常使用的动态 PET 研究的适用性，这些研究旨在获取生理信息，如一个区域中神经受体的结合潜能。与此同时，现有的 PET/MRI 衰减校正方法基于合成 CT 数据，这并不理想，因为 CT 数据是用比 PET 数据能量低得多的光子采集的，因此不能最佳地反映 PET 衰减图。我们训练了一个卷积神经网络，从 T1 加权 MRI 生成患者特定的透射数据。使用训练好的网络，我们为一个测试集生成了透射数据，该测试集包括 11 名使用 C-标记的-[2-]-4-（2-甲氧基苯基）-1-哌嗪基]乙基]-[（2-吡啶基）环己烷甲酰胺]（C-WAY-100635）和 10 名使用 C-标记的 3-氨基-4-（2-二甲基氨甲基-苯基硫代）苯甲腈（C-DASB）进行扫描的受试者。我们评估了静态和动态重建。对于动态 PET 数据，我们报告了 C-WAY-100635 的不可置换和自由结合潜能以及 C-DASB 的分布容积的差异。生成的透射数据的平均偏差为-1.06%±0.81%。静态 PET 摄取的全局偏差分别为-0.49%±1.7%和-1.52%±0.73%，分别用于 C-WAY-100635 和 C-DASB。我们的神经网络方法能够以足够的精度合成用于静态和动态 PET 研究的患者特定的透射数据。
