@misc{oai:repo.qst.go.jp:00075608,
 author = {赤松, 剛 and 田島, 英朗 and 吉田, 英治 and 岩男, 悠真 and 脇坂, 秀克 and 前田, 貴雅 and 山谷, 泰賀 and Akamatsu, Go and Tashima, Hideaki and Yoshida, Eiji and Iwao, Yuma and Wakizaka, Hidekatsu and Maeda, Takamasa and Yamaya, Taiga},
 month = {Apr},
 note = {【Purpose】Brain PET imaging plays important roles in neurology, neuro-oncology and molecular imaging research. There is a potential demand for a brain PET scanner with high spatial resolution and high sensitivity to achieve more accurate brain imaging. Therefore, we are developing a brain-dedicated helmet-type PET scanner with 4-layer depth-of-interaction (DOI) detectors. In this study, we evaluated performance characteristics of the helmet-type PET prototype. 

【Method】The helmet-type PET prototype was composed of 54 detector blocks. The 47 detector blocks were arranged with a hemispherical helmet detector unit and the other 7 detector blocks were located at the neck position. The detector block itself was composed of 4-layer GSOZ crystals. We measured performance characteristics (spatial resolution, sensitivity, scatter fraction, count rate characteristics, accuracy of corrections for count losses and randoms, and image quality) of the helmet-type PET prototype based on the NEMA NU2-2018 standard. In addition to the original NEMA standard, some modified measurements were performed to predict performance of brain imaging. 

【Result】The helmet-type PET prototype showed great spatial resolution (1.6 mm at 1 cm offset, OSEM reconstruction), great sensitivity (32.7 kcps/MBq obtained by the modified setup) and promising image quality. With the original 70-cm-long scatter phantom, the scatter fraction, peak noise-equivalent count rate and maximum absolute bias were 37%, 34.9 kcps@4.0kBq/mL and 17.4%.

【Conclusion】We confirmed the high performance of the helmet-type PET prototype. The helmet-type PET scanner has the high potential to achieve more accurate brain PET imaging, even with a small number of detectors., 第75回日本放射線技術学会総会学術大会},
 title = {NEMA NU2 performance evaluation of a helmet-type PET prototype with four-layer DOI detectors},
 year = {2019}
}