RT Journal Article SR Electronic T1 Correction of partial volume effect of cerebral glucose metabolism images measured by PET/MRI JF Journal of Nuclear Medicine JO J Nucl Med FD Society of Nuclear Medicine SP 1677 OP 1677 VO 59 IS supplement 1 A1 Hiroshi Ito A1 Takeyuki Nambu A1 Keisuke Matsubara A1 Hitoshi Kubo A1 Noboru Oriuchi A1 Shirou Ishii YR 2018 UL http://jnm.snmjournals.org/content/59/supplement_1/1677.abstract AB 1677Objectives: The integrated design of positron emission tomography/magnetic resonance imaging (PET/MRI) scanner system have potential for clinical and research settings of in dementia and other neurodegenerative diseases. In the present study, the partial volume correction (PVC) of PET images with [F-18]FDG was performed by using MR images acquired with PET/MRI scanner system. The normal database of PET images of glucose metabolism per gray matter tissue was also constructed by using the PVC of PET images. Methods: A PET scanning was performed for 30 minutes with the list-mode data acquisition from 50 min after intravenous bolus injection of [F-18]FDG using a Siemens mMR PET/MRI scanner on five healthy men. MR images were also acquired simultaneously. The PET scanning data simultaneously acquired with MR imaging of three-dimensional T1WI (51 to 56 min after injection of [F-18]FDG) were extracted from PET sinogram data with the list-mode data acquisition. The standardized uptake value (SUV) images of PET were calculated. All PET and MR images were transformed into standard brain format using SPM12. PVC of PET images was performed by the Muller-Gartner method using gray and white matter images segmented from three-dimensional T1WI. Results: The cerebral cortical regions, thalamus and putamen showed increases in SUV by PVC (11-36% increase), however, almost no changes in SUV were observed in the cerebellum and parahippocampal gyrus which showed highest gray matter fractions. A significant negative correlation was observed between the gray matter fraction and the percentage changes in SUV by PVC as compared with SUV without PVC (r=-0.92). The normal database of PET images of glucose metabolism per gray matter tissue could also be constructed by using the PVC of PET images. Lowest SUV were observed in the cerebellum and parahippocampal gyrus with PVC. Conclusions: The regional distribution of SUV of [F-18]FDG must be changed by PVC, and therefore it should be considered in clinical evaluation of brain PET images with PVC. Lowest SUV in the cerebellum and parahippocampal gyrus might be introduced by both lower cerebral glucose consumption and lower lumped constant in these regions. Several methods for PVC of PET using MRI have been reported. Further studies to assess the accuracy of these methods will be needed.