Extend brilliance workstation 3

All participants were required to fast for at least 6 hours and the levels of serum glucose were less than 110 mg/dL before the ¹⁸F-FDG PET/CT scan. PET/CT images were acquired by a hybrid PET/CT scanner (GEMINI TF 64, Philips). For attenuation correction, a low-dose plain CT scan was performed from the skull base to the middle of thigh, with the following parameters: tube voltage=120 kV, tube current=249 mA, detector collimation=64 × 0.625 mm, pitch=0.829, tube rotation time=0.5 sec, slice thickness=5.0 mm. A three-dimension mode PET scan was performed approximately 1 hour later, after intravenous injection of ¹⁸F-FDG with a dose of 3.7 MBq/kg. The PET scan parameters were as follows: field of view (FOV) of 576 mm, matrix of 144 × 144, slice thickness and interval of 5.0 mm. The PET images were reconstructed using the ordered subset expectation maximization (OSEM) method (33 subsets per iteration). All collected data were transferred into Philips Extend Brilliance Workstation 3.0 to reconstruct PET, CT, and PET/CT fusion images, respectively.

The FDG PET/CT imaging protocol followed the EANM procedure guidelines [16 (link)]. All patients were required to have a fasting period of at least 6 h and a blood glucose level below 110 mg/dl before administration of ¹⁸F-FDG. PET/CT images were acquired 60 min after intravenous injection of 3.7 MBq/kg dose of ¹⁸F-FDG using a hybrid PET/CT scanner (GEMINI TF 64, Philips, Netherlands). A nonenhanced CT scan was performed from the skull base to the middle of the thigh for precise anatomical location and attenuation correction, and the corresponding scanning parameters were as follows: 120 kVp, 80 mAs, pitch of 0.829, tube rotation time of 0.5 s per rotation, and both slice thickness and interval of 5.0 mm. A 3-dimensional mode PET scan, which matched the CT slice thickness, was then performed. PET datasets were iteratively reconstructed using the ordered subset expectation maximization algorithm.
Two experienced nuclear medicine physicians analyzed all PET, CT, and PET/CT fusion images using Philips Extend Brilliance Workstation 3.0. SUVmax was determined as the highest voxel value within the VOI. We delineated the volume of interest (i.e., tumor region) using the SUV threshold method. MTV represents the sum of voxels with SUV no less than 40% of the SUVmax. SUVmean represented the mean SUV in the VOI. TLG was the product of MTV and SUVmean.