TY - JOUR T1 - Multicenter phantom study of characteristics of gamma cameras with LEHR collimators in cardiac <sup>123</sup>I-MIBG imaging JF - Journal of Nuclear Medicine JO - J Nucl Med SP - 1916 LP - 1916 VL - 57 IS - supplement 2 AU - Koichi Okuda AU - Kenichi Nakajima AU - Chiemi Kitamura AU - Masayo Hori AU - Yumiko Kirihara AU - Shinro Matsuo AU - Junichi Taki AU - Mitsumasa Hashimoto AU - Seigo Kinuya Y1 - 2016/05/01 UR - http://jnm.snmjournals.org/content/57/supplement_2/1916.abstract N2 - 1916Objectives We have proposed a standardization method for the determination of 123I-metaiodobenzylguanidine (MIBG) heart-to-mediastinum ratio for the calibration of the characteristics of gamma cameras and collimators using a dedicated phantom. This standardization activity has been extended to include over 500 hospitals in Japan. Accordingly, we have been accumulating the calibration factors obtained from different combinations of gamma cameras and collimators. The aim of this study was to validate the calibration factors in various gamma cameras with low-energy high-resolution (LEHR) collimators.Methods A total of 302 planar phantom images were acquired by using six different gamma cameras: Infinia (n = 27) and Discovery/Optima (n = 16) by GE Healthcare; Symbia/e.cam by Siemens Healthcare (n = 122); Symbia/e.cam by Toshiba Medical Systems (n = 62); BrightView by Philips Healthcare (n = 45); and PRISM by Shimadzu Medical Systems (n = 30). The phantom images were all acquired by using the selected cameras with LEHR collimators, except for the BrightView camera, with a cardiac high-resolution collimator, which has equivalent characteristics to LEHR collimators. The mean acquisition time was 305 ± 102 s (95% range, 60-600 sec). Planar imaging was performed with 128 × 128 (n = 11), 256 × 256 (n = 226), and 512 × 512 (n = 62) matrices. The calibration factor, defined as the conversion coefficient K, was computed by using a dedicated software program. The standard K value lies within an approximate range of 0.5 to 1.0, corresponding to low-energy to medium-energy collimators, and enables cross-calibration among various gamma cameras with collimators.Results The mean K value for all of the gamma cameras with low-energy-type collimators was 0.547 ± 0.036 (95% range, 0.485-0.629). When the conversion coefficients were more individually evaluated for each of the vendors, the mean K values for the Symbia/e.cam cameras by Siemens and Toshiba were 0.533 ± 0.033 (95% range, 0.480-0.616) and 0.553 ± 0.034 (95% range 0.483-0.624), respectively (P = 0.0002). The mean K values for the two GE gamma cameras, Discovery/Optima (0.539 ± 0.035; 95% range, 0.489-0.597) and Infinia (0.561 ± 0.034; 0.486-0.642, P = 0.072), were equivalent. The mean K value for the BrightView camera was 0.547 ± 0.021 (95% range, 0.510-0.595). The PRISM gamma camera had the highest mean K value of the six cameras evaluated in this study (0.587 ± 0.037; 95% range, 0.519-0.683, P &lt; 0.034).Conclusions The characteristics of the gamma cameras and collimators, in combination, could be determined by using the conversion coefficient K, derived from the multicenter phantom study. Because the K values in the present study differed significantly, depending on the vendor, even among the LEHR collimators of the same type, cross-calibration might be required to obtain reliable results. ER -