Sharpir

¹⁸F-FDG was synthesized with an in-house cyclotron and automated synthesis system (F200; Sumitomo Heavy Industries, Shinagawa, Tokyo, Japan). PET/CT images were acquired 60 min after an intravenous injection of ¹⁸F-FDG, fixed at 5.0 MBq/kg. Patients were instructed to urinate before scanning to reduce tracer accumulation in the bladder. All PET/CT images were obtained using a Discovery PET/CT 600 (GE Healthcare, Pewaukee, WI) with a multi-detector-row CT component (16 detectors). Scanning covered an area from the head to the mid-thigh. Low dose CT with shallow breathing was performed first and used for attenuation correction and image fusion. CT acquisition was performed with 120 kVp using an auto exposure control system, beam pitch of 0.938, slice thickness of 3.75 mm. Emission images were acquired in three-dimensional mode for 2.5 min per bed position. The 3D-OSEM reconstruction method (VUE point HD; GE Healthcare) was used both for (a) conventional PET (16 subsets; 3 iterations), and (b) PSF-PET [(16 subsets; 5 iterations) + PSF algorithm (Sharp IR, GE Healthcare)]. For both reconstructions, the matrix size was 192 × 192, resulting in a 3.65 × 3.65 × 3.65 mm voxel size, and a 4 mm Gaussian filter was used.

PET/CT imaging was performed with a hybrid PET/CT system (Discovery 710, GE Healthcare). First, a scout view and a low-dose 64-slice CT scan was performed for attenuation correction of PET emission data and localisation of the phantom structures. CT scan was acquired with a tube voltage of 140 kV in the helical mode with current modulation in the range of 40–120 mA. The rotation speed was1.25 s^-1, helical thickness 3.75 mm, standard reconstruction, slice thickness 1.25 mm, matrix size 512x512.
Following CT, three-dimensional PET images were acquired. Emission scan time was 30 min per bed position (15.7 cm with 23% bed overlap). The number of bed positions was two for the Jaszczak and one for the NEMA phantom. Emission data was corrected for geometrical response, detector efficiency, system dead time, random coincidences, scatter and attenuation.
For non-attenuation corrected images the 3D iterative reconstruction technique (GE VUE Point HD) with 2 iterations/24 subsets and a filter cut-off of 6.4 mm was used. The matrix size was 192x192. Attenuation corrected images were obtained with the use of 3D-OSEM iterative reconstruction method. It was conducted with TOF PET reconstruction algorithm (GE VUE Point FX) and a resolution recovery algorithm (GE SharpIR) with 4 iterations/32 subsets and a filter cut-off of 3.0 mm. The matrix size was 256x256.

Imaging was performed using a Discovery 690 PET/CT scanner (GE Healthcare, Chicago, Illinois, USA) 20 min (range 19–21 min) after administration of 382 MBq (range 293–411 MBq) of ¹¹C-Methionine. One 15 cm bed position centered on the subject’s pituitary gland was acquired for 20 min. The images were reconstructed with OSEM iterative reconstruction using 3 iterations and 24 subsets, 128 × 128 matrix size, 2 mm Gaussian filter, scatter correction, CT-measured attenuation correction, time of flight and point spread function correction (SharpIR, GE). An unenhanced CT scan was acquired with 140 kV, fixed mA of 220, a rotation speed of 0.5 s, a pitch of 0984:1, 30 cm field of view, a slice thickness of 1.25 mm and a 1.25 mm spacing interval and reconstructed using filtered back projection.