The Netherlands protocol for standardisation and quantification of FDG whole body PET studies in multi-centre trials

作者：Ronald Boellaard, Wim J.G. Oyen, Corneline J. Hoekstra, Otto S. Hoekstra, Eric P. Visser, Antoon T. M. Willemsen, Bertjan Arends, Fred J. Verzijlbergen, Josée M. Zijlstra, Anne M. J. Paans, Emile F.I. Comans, Jan Pruim · 发表于：European Journal of Nuclear Medicine and Molecular Imaging · 年份：2008 · DOI：10.1007/s00259-008-0874-2 · 被引用次数：398 · 研究领域：Medical Imaging Techniques and Applications、Radioactive Decay and Measurement Techniques、Radiopharmaceutical Chemistry and Applications

INTRODUCTION: Several studies have shown the usefulness of positron emission tomography (PET) quantification using standardised uptake values (SUV) for diagnosis and staging, prognosis and response monitoring. Many factors affect SUV, such as patient preparation procedures, scan acquisition, image reconstruction and data analysis settings, and the variability in methodology across centres prohibits exchange of SUV data. Therefore, standardisation of 2-[(18)F] fluoro-2-deoxy-D-glucose (FDG) PET whole body procedures is required in multi-centre trials. METHODS: A protocol for standardisation of quantitative FDG whole body PET studies in the Netherlands (NL) was defined. This protocol is based on standardisation of: (1) patient preparation; (2) matching of scan statistics by prescribing dosage as function of patient weight, scan time per bed position, percentage of bed overlap and image acquisition mode (2D or 3D); (3) matching of image resolution by prescribing reconstruction settings for each type of scanner; (4) matching of data analysis procedure by defining volume of interest methods and SUV calculations and; (5) finally, a multi-centre QC procedure is defined using a 20-cm diameter phantom for verification of scanner calibration and the NEMA NU 2 2001 Image Quality phantom for verification of activity concentration recoveries (i.e., verification of image resolution and reconstruction convergence). DISCUSSION: This paper describes a protocol for standardization of quantitat...