Si Chinhong, Mok Greta S P, Chen Ling, Tsui Benjamin M W
aBiomedical Imaging Laboratory, Department of Electrical and Computer Engineering, Faculty of Science and Technology, University of Macau, Macau SAR, China bThe Russell H. Morgan Department of Radiology and Radiological Science, Division of Medical Imaging Physics, Johns Hopkins University, Baltimore, Maryland, USA.
Nucl Med Commun. 2016 Mar;37(3):313-21. doi: 10.1097/MNM.0000000000000429.
Pinhole single-photon emission computed tomography provides superior trade-off between resolution and detection efficiency as compared with conventional parallel-hole collimators for imaging small objects. This study aims to design and evaluate an optimized adaptive multipinhole (MPH) collimator for improved clinical myocardial perfusion single-photon emission computed tomography imaging (MPI) and preclinical small-animal imaging (SAI) of rats based on a clinical scanner.
The target resolution and field of view was set to be 1/20 cm for MPI and 0.15/5 cm for SAI, respectively. We determined the design parameters by maximizing the detection efficiency based on system constraints. Point source simulations using Geant4 Application for Emission Tomography were performed for different collimator-to-center of field of view distances to assess the detection efficiency and resolution trade-off. The XCAT phantom with Tc-99m sestamibi distribution and the four-dimensional mouse whole-body phantom with Tc-99m methylene diphosphonate distribution were used to generate noise-free and noisy projections using a three-dimensional analytical MPH projector. Projections were reconstructed using a three-dimensional MPH ordered-subset expectation maximization algorithm. Noise and bias were assessed on the reconstructed images for different collimators.
The design parameters are (i) 14 pinholes with 3.42 mm aperture size, 14.5 cm collimator-to-detector distance for MPI; (ii) six pinholes with an aperture size of 0.94 mm, 21.2 cm collimator-to-detector distance for SAI. For MPI, the projected full width at half maximum values were 10.68 and 8.19 mm for low energy high resolution (LEHR) and MPH, respectively, whereas MPH had double detection efficiency. For SAI, the projected full width at half maximum values for LEHR and MPH were 4.93 and 1.20 mm, respectively, whereas the detection efficiency of MPH showed 17.5% improvement as compared with LEHR. The noise-bias trade-off improved for MPH as compared with LEHR for both MPI and SAI. The proposed collimator will have adjustable collimator-to-detector distances - that is, 14.5 cm for MPI and 21.2 cm for SAI.
The new collimator yields substantial improvement in image quality as compared with current MPI using LEHR with extra capability for SAI, bridging the clinical and preclinical imaging based on the same platform.
与传统平行孔准直器相比，针孔单光子发射计算机断层扫描在成像小物体时，在分辨率和探测效率之间提供了更好的权衡。本研究旨在设计并评估一种优化的自适应多针孔（MPH）准直器，以改善基于临床扫描仪的临床心肌灌注单光子发射计算机断层扫描成像（MPI）以及大鼠的临床前小动物成像（SAI）。
将MPI的目标分辨率和视野分别设置为1/20 cm，SAI的目标分辨率和视野分别设置为0.15/5 cm。我们基于系统约束通过最大化探测效率来确定设计参数。使用Geant4发射断层扫描应用程序对不同的准直器到视野中心距离进行点源模拟，以评估探测效率和分辨率之间的权衡。使用具有Tc-99m司他米比分布的XCAT体模和具有Tc-99m亚甲基二膦酸盐分布的四维小鼠全身体模，通过三维解析MPH投影仪生成无噪声和有噪声的投影。使用三维MPH有序子集期望最大化算法对投影进行重建。针对不同的准直器，在重建图像上评估噪声和偏差。
设计参数为：（i）MPI时14个针孔，孔径大小为3.42 mm，准直器到探测器的距离为14.5 cm；（ii）SAI时6个针孔，孔径大小为0.94 mm，准直器到探测器的距离为21.2 cm。对于MPI，低能高分辨率（LEHR）和MPH的投影半高宽值分别为10.68和8.19 mm，而MPH的探测效率提高了一倍。对于SAI，LEHR和MPH的投影半高宽值分别为4.93和1.20 mm，而MPH的探测效率与LEHR相比提高了17.5%。与LEHR相比，MPH在MPI和SAI中的噪声-偏差权衡均得到改善。所提出的准直器将具有可调节的准直器到探测器的距离，即MPI为14.5 cm，SAI为21.2 cm。
与当前使用LEHR的MPI相比，新型准直器在图像质量上有显著提高，并且具有用于SAI的额外能力，在同一平台上弥合了临床和临床前成像之间的差距。
