@misc{oai:repo.qst.go.jp:00066498,
 author = {赤松, 剛 and 井狩, 彌彦 and 脇坂, 秀克 and 山谷, 泰賀 and 木村, 裕一 and 織田, 圭一 and 千田, 道雄 and 赤松 剛 and 脇坂 秀克 and 山谷 泰賀 and 木村 裕一 and 織田 圭一},
 month = {Oct},
 note = {Brain PET imaging technique is valuable for clinical researches as well as for clinical practices. However, image quality and quantitative capability of PET data depend on the PET camera model and the details of acquisition protocol, which makes it a challenge to acquire reliable data in a multicenter clinical study. To make multicenter brain PET data meaningful and to contribute to establishing brain PET as a verified imaging biomarker, we have proposed methods of evaluating absolute quantitative capability, resolution, contrast, uniformity, image noise, etc., using appropriate phantoms for standardization among different camera models. At this moment, we have issued phantom test procedures for brain PET with 11C-methionine, 18F-FDG, and amyloid agents (11C-PiB, 18F-florbetapir, 18F-flutemetamol, and 18F-florbetaben). We first defined the phantom models and the elements of quality that are essential for brain methionine, FDG, and amyloid PET images. Those phantoms that are commercially available and easy to use have been selected: so-called brain tumor phantom for brain methionine PET and Hoffman 3D brain phantom and uniform cylindrical phantom for brain FDG and amyloid PET. Subsequently, we have determined a physical performance indicator and criteria for the phantom tests for each PET drug, which was evaluable and achievable with most of the PET camera models. As a result, these phantom test procedures have been used in a number of multicenter PET studies including Japanese Alzheimer’s Disease Neuroimaging Initiative (J-ADNI). The proposed phantom test and criteria facilitate standardization of brain PET imaging and are useful to validate brain PET scanning performance of the imaging sites., 2017 IEEE NSS/MIC},
 title = {Phantom test procedures and criteria for standardization of brain PET imaging across different cameras},
 year = {2017}
}