Department of Nuclear Medicine, Essen University Hospital, Duisburg-Essen University, Essen, Germany -
Department of Nuclear Medicine, Essen University Hospital, Duisburg-Essen University, Essen, Germany.
Q J Nucl Med Mol Imaging. 2023 Mar;67(1):57-68. doi: 10.23736/S1824-4785.21.03300-8. Epub 2021 Jul 26.
The aim of this phantom study was to optimize the [Ga]Ga-PSMA PET/CT examination in terms of scan time duration and image reconstruction parameters, in combination with PSF and TOF modelling, in a digital Biograph Vision PET/CT scanner.
Three types of phantoms were used: 1) soft-tissue tumor phantom consisting of six spheres mounted in a torso phantom; 2) bone-lung tumor phantom; 3) resolution phantom. Phantom inserts were filled with activity concentrations (ACs) that were derived from clinical data. Phantom data were acquired in list-mode at one bed position. Images with emission data ranging from 30 to 210 s in 30-s increments were reconstructed from a reference image acquired with 3.5-min emission. Iterative image reconstruction (OSEM), point-spread-function (PSF) and time-of-flight (TOF) options were applied using different iterations, Gaussian filters, and voxel sizes. The criteria for image quality was lesion detectability and lesion quantification, evaluated as contrast-to-noise ratio (CNR) and maximum AC (peak AC), respectively. A threshold value of CNR above 6 and percentage maximum AC (peak AC) deviation range of ±20% of the reference image were considered acceptable. The proposed single-bed scan time reduction was projected to a whole-body examination (patient validation scan) using the continuous-bed-motion mode.
Sphere and background ACs of 20 kBq/mL and 1 kBq/mL were selected, respectively. The optimized single-bed scan time was approximately 60 s using OSEM-TOF or OSEM-TOF+PSF (four iterations, 4.0-mm Gaussian filter and almost isotropic voxel size of 3.0-mm side length), resulting in a PET spatial resolution of 6.3 mm for OSEM-TOF and 5.5 mm for OSEM-TOF+PSF. In the patient validation, the maximum percentage difference in lesion quantification between standard and optimized protocol (whole-body scan time of 15 vs. 5 min) was below 19%.
A reduction of single-bed and whole-body scan time for [Ga]Ga-PSMA PET/CT compared to current recommended clinical acquisition protocols is postulated. Clinical studies are warranted to validate the applicability of this protocol.
本研究旨在通过结合 PSF 和 TOF 建模，在数字 Biograph Vision PET/CT 扫描仪中，针对扫描时间持续时间和图像重建参数，对 [Ga]Ga-PSMA PET/CT 检查进行优化。
使用了三种类型的体模：1）软组织肿瘤体模，由安装在躯干体模中的六个球体组成；2）骨肺肿瘤体模；3）分辨率体模。体模插入物充满了源自临床数据的活性浓度（AC）。在一个床位位置以列表模式采集体模数据。从参考图像中以 30 秒的增量重建发射数据范围为 30 至 210 秒的图像，参考图像的发射时间为 3.5 分钟。使用不同的迭代次数、高斯滤波器和体素大小应用迭代图像重建（OSEM）、点扩散函数（PSF）和时间-of-flight（TOF）选项。图像质量的标准是病灶的可检测性和病灶的定量，分别表示为对比度噪声比（CNR）和最大 AC（峰值 AC）。接受的标准为 CNR 超过 6，以及参考图像的最大 AC（峰值 AC）偏差范围在±20%以内。使用连续床位运动模式，将建议的单床位扫描时间减少量投影到全身检查（患者验证扫描）。
选择了 20 kBq/mL 和 1 kBq/mL 的球体和背景 AC。使用 OSEM-TOF 或 OSEM-TOF+PSF（四次迭代，4.0-mm 高斯滤波器和几乎各向同性的 3.0-mm 边长体素大小），优化后的单床位扫描时间约为 60 秒，得到 OSEM-TOF 的 PET 空间分辨率为 6.3mm 和 OSEM-TOF+PSF 的 5.5mm。在患者验证中，标准和优化方案（全身扫描时间 15 分钟与 5 分钟）之间的病灶定量最大百分比差异低于 19%。
与当前推荐的临床采集方案相比，[Ga]Ga-PSMA PET/CT 的单床位和全身扫描时间减少是可能的。需要进行临床研究来验证该方案的适用性。
