Matsumoto Keiichi, Kitamura Keishi, Mizuta Tetsuro, Shimizu Keiji, Murase Kenya, Senda Michio
Department of Image-based Medicine, Institute of Biomedical Research and Innovation.
Nihon Hoshasen Gijutsu Gakkai Zasshi. 2006 Jan 20;62(1):78-85. doi: 10.6009/jjrt.62.78.
The SET-3000 G/X (Shimadzu Corp., Kyoto, Japan) has a large aperture and functions as a three-dimensional (3D) dedicated PET scanner. However, the large number of line of responses in the SET-3000 G/X scanner creates a large volume of sinogram data and prolongs reconstruction time in iterative reconstruction. The purpose of this study was to optimize basic acquisition parameters (maximum ring difference and span) for sensitivity and spatial resolution for 3D whole-body (18)F-FDG PET.
Detector rings and image planes numbered 50 and 99, respectively. In sensitivity measurement, the maximum ring difference (MRD) was changed from 1 to 49. In the measurement of spatial resolution, the span was changed from 3 to 21. For sensitivity and spatial resolution measurements, the standard protocols defined by the Japan Radioisotope Association (JRIA) 1994 and the National Electrical Manufacturers Association (NEMA) NU 2-2001 were used. We also evaluated the corresponding image noise by placing identical ROI on the reconstructed images.
The total sensitivity of MRD=49 was 85.7 cps/Bq/ml in a uniform phantom (15 cm diameter, 30 cm tall cylinder) filled with (18)F. This was approximately two times higher than MRD=13. The image noise in the center of the axial FOV decreased with increasing MRD. Spatial resolution was slightly decreased as MRD increased, but axial resolution deteriorated with a span of more than 11.
Optimum basic data-acquisition parameters for whole-body (18)F-FDG PET were MRD 49 to obtain maximum sensitivity and span 9 to avoid decreasing spatial resolution. Additionally, it was concluded that the basic data-acquisition parameters should be carefully selected for 3D whole-body (18)F-FDG PET in order to maximize the efficiency of PET measurement.
SET - 3000 G/X（日本京都岛津公司）具有大孔径，可作为三维（3D）专用正电子发射断层扫描（PET）扫描仪。然而，SET - 3000 G/X扫描仪中大量的响应线会产生大量的正弦图数据，并延长迭代重建中的重建时间。本研究的目的是优化用于3D全身（18）F - FDG PET灵敏度和空间分辨率的基本采集参数（最大环差和跨度）。
探测器环和图像平面分别编号为50和99。在灵敏度测量中，最大环差（MRD）从1变化到49。在空间分辨率测量中，跨度从3变化到21。对于灵敏度和空间分辨率测量，使用了日本放射性同位素协会（JRIA）1994年和美国国家电气制造商协会（NEMA）NU 2 - 2001定义的标准协议。我们还通过在重建图像上放置相同的感兴趣区（ROI）来评估相应的图像噪声。
在充满（18）F的均匀体模（直径15 cm、高30 cm的圆柱体）中，MRD = 49时的总灵敏度为85.7 cps/Bq/ml。这大约是MRD = 13时的两倍。轴向视野（FOV）中心的图像噪声随着MRD的增加而降低。随着MRD的增加，空间分辨率略有下降，但当跨度超过11时轴向分辨率变差。
全身（18）F - FDG PET的最佳基本数据采集参数是MRD为49以获得最大灵敏度，跨度为9以避免空间分辨率降低。此外，得出结论，对于3D全身（18）F - FDG PET，应仔细选择基本数据采集参数，以最大限度提高PET测量效率。
