Discovery mi

All study participants underwent two PET scans on a digital GE Discovery MI scanner (General Electric Medical Systems), with an average of 36 ± 35 days between scans. Participants were injected with 341 ± 53 MBq of [¹⁸F]flortaucipir or 365 ± 20 MBq of [¹⁸F]RO948, and LIST mode emission data was acquired for each scan of 80–100 min ([¹⁸F]flortaucipir) or 70–90 min ([¹⁸F]RO948) post injection. Different time frames for image acquisition were chosen due to different pharmacokinetics as described previously [15 (link), 36 (link)].
Low-dose CT scans were performed immediately prior to the PET scans for attenuation correction. PET data was reconstructed using VPFX-S (ordered subset expectation maximization (OSEM) with time-of-flight (TOF) and point spread function (PSF) corrections) with 6 iterations and 17 subsets with 3 mm smoothing, standard Z filter, and 25.6-cm field of view with a 256 × 256 matrix. LIST mode data was binned into 4 × 5-min time frames, and the resulting PET images motion corrected, summed, and co-registered to their corresponding T1-weighted MR images.

Patients need to fast for at least 6 h and blood glucose levels should be controlled to less than 11.1 mmol/L. A dose of 3.70-5.55 MBq/kg ¹⁸F-FDG was administered intravenously. Approximately 60 min after ¹⁸F-FDG injection, a whole-body CT scan from the base of the skull to mid-thigh was performed, followed by a whole-body PET with the same range. Two PET/CT scanners—Discovery ST 8, GE Healthcare, WI, USA (n = 46) and Discovery MI, GE Healthcare, WI, USA (n = 6)—were used to perform all the acquisitions with the same procedures. The acquisition parameters were as follows: 140 kV, 150 mAs, pitch 1.675, 512 × 512 image matrix, slice thickness of 3.75 mm, and a total of six or seven cradle positions with 3.5 min/ cradle position for whole-body acquisition and 120 kV, 150 mA; slice thickness, 2.5 mm/ 1.25 mm for Discovery ST 8 and Discovery MI.

We acquired PET images using the following PET/computed tomography (CT) scanners: Biograph mCT (Siemens) in Seoul,^{32 (link)} Discovery 690 (GE Healthcare) in BioFINDER-1, Discovery MI (GE Healthcare) in BioFINDER-2,^{13 (link),17 (link)} Biograph 6 Truepoint (Siemens) at UCSF and BACS,^{12 (link),33 (link)} and multiple scanners in the multicenter ADNI^{34 (link)} and Avid Radiopharmaceuticals^{23 (link)} cohorts. All PET data were reconstructed at the respective sites into 4 × 5-minute frames within the 80- to 100-minute ([¹⁸F]flortaucipir) and 70- to 90-minute ([¹⁸F]RO948) intervals after injection. Amyloid PET was performed using carbon 11 (¹¹C)–Pittsburgh Compound B (BACS and UCSF), [¹⁸F]florbetapir (Avid Radiopharmaceuticals and ADNI subsets), [¹⁸F]florbetaben (Seoul and ADNI subsets), or [¹⁸F]flutemetamol (BioFINDER-1 and BioFINDER-2). Magnetic resonance images were acquired on the following scanners: 3.0-T Discovery MR750 (GE Healthcare) in Seoul,^{32 (link)} 3.0-T Tim Trio (Siemens) or 3.0-T Prisma (Siemens) in BioFINDER-1 and -2,^{13 (link),17 (link)} 3.0-T Tim Trio or 3.0-T Prisma (Siemens) at UCSF,^{33 (link)} 1.5-T Magnetom Avanto (Siemens) for BACS,^{12 (link)} and multiple 1.5-T and 3-T scanners in the multicenter ADNI^{34 (link)} and Avid Radiopharmaceuticals^{23 (link)} cohorts.

Patients underwent clinical routine ${}^{18}$ F-FDG PET/CT. After a fasting time of at least 6 h, a single injection in bolus of ${}^{18}$ F-FDG (mean adjusted dose 3.5–4.5 MBq/kg) was administered. Patients were scanned after 60 min from ${}^{18}$ F-FDG administration (uptake time), after hydration (500 mL water) and after voiding the bladder. Images were acquired on cross-calibrated GE Discovery MI, Discovery STE, Discovery 710 (General Electric, Milwaukee, WI, USA), acquisition time was 2 min per bed position. In summary, the acquisition protocol was performed according to the European Association of Nuclear Medicine (EANM) procedure guidelines [15 (link)].

All PET/CT and CE-CT examinations followed basic study protocols. For PET/CT, patients fasted for at least four hours, FDG dosage was body-weight adjusted, the uptake time was standardized to 60 minutes in supine position, a non-enhanced CT scan was performed and used for attenuation correction, and data was acquired with arms overhead whenever possible. Blood glucose levels <12 mmol/l were accepted. Body weight, height, and blood glucose level were measured prior to imaging. Five different types of PET/CT scanners were used throughout the study period, i.e. Discovery STE, Discovery LS, Discovery RX, Discovery MI, and Discovery 690 (all GE Healthcare, Waukesha, WI). To compensate for differences in the sensitivity of the different PET/CT scanner generations, we measured the metabolic activity in the mediastinal blood pool and in the liver tissue for reference.
For CE-CT of the abdomen, 80 ml iodinated contrast material (Visipaque® 320, GE Healthcare) were injected, timed for imaging at the portal venous phase with a tube voltage of 120 kV and a tube current–time product of 100–320 mAs. If patients had a recent CE-CT of the region of interest prior to the PET/CT, the CE-CT was not repeated.

Met-PET scanning was performed at Chiba Ryogo Center using a Discovery MI (GE Healthcare, Tokyo, Japan) PET/computed tomography (CT) scanner with spatial resolution of 4.8 mm. Patients fasted for ≥ 6 h before PET scanning, and PET images were acquired with the patient in a resting state. Static scanning was performed for 6 min using 3-dimensional (3D) acquisition, and attenuation-corrected PET images were reconstructed using CT data by means of a 3D-ordered subset expectation maximization algorithm (20 subsets and 2 iterations). A Met dose of 370 MBq was injected intravenously within 1 min, with the scan starting 10 min after Met injection. Summation images covering 20 to 40 min after injection were used for analysis. Met uptake was semiquantitatively evaluated using the ratio T_max/N_ave, which is generated by dividing the maximum pixel count of the standardized uptake value of the tumor by the mean uptake of the contralateral normal frontal-lobe grey matter, avoiding the region affected by the tumor.

All patients underwent clinically indicated PET MPI using 13N-ammonia acquired at rest and during pharmacological stress (adenosine infused at 0.14 mg⋅kg⁻¹⋅min⁻¹ over 6 minutes or single bolus injection of 400 mcg of regadenosone) according to clinical routine. All data were acquired in list-mode on a PET/CT scanner (Discovery DST, Discovery MI or Discovery VCT, all GE Healthcare, Waukesha, WI, USA) as previously reported.^{16 (link)} In brief, a body mass index-adapted dose of 13N-ammonia (i.e., 400-1200 Megabecquerels) was injected and the datasets were reconstructed using ordered subset expectation maximum (OSEM, VUE Point HD or VUE Point FX with 2 iterations and 16 subsets), and a 5 mm Hanning filter and standard decay, scatter and sensitivity corrections (voxel size 2.34, 2.34, 2.80-3.27) were applied. Low-dose unenhanced computed tomography was used for attenuation correction. Dynamic datasets were reconstructed from the first 7 minutes of acquisition and consisted of 9 frames of 10 seconds duration, 6 frames of 15 seconds, 3 frames of 20 seconds, 2 frames of 30 seconds and 1 frame of 120 seconds. MBF at rest (corrected for the rate-pressure-product) and during stress and MFR was calculated using commercially available software (QPET 2017.7 Cedars-Sinai Medical Center, Los Angeles, CA, USA). Static datasets were reconstructed from the following 10 minutes of the acquisition.