Biograph mct flow pet ct

The 18F-florbetapir PET scans were performed with a PET/CT system (Biograph mCT Flow PET/CT, Siemens, Erlangen, Germany) 50 min after the intravenous injection of 7.4 MBq/kg 18F-florbetapir and lasted for 20 min. PET images were reconstructed by filtered back projection algorithm with corrections for decay, normalization, dead time, photon attenuation, scatter and random coincidences. In brief, images were coregistered to the individual structural MRI and further warped into the standard Montreal Neurological Institute (MNI) stereotactic space. Standard uptake value ratios (SUVR) were calculated for the cortical regions of interest (ROIs) using cerebellar crus as a reference^{36 (link)}. The mean cortical SUVR scores were calculated by weighted averaging of these ROIs. PET Image interpretation was performed by three nuclear medical physicians with specialized training according to the guidelines of visual rating^{37 (link)}.

First, 18F-florbetapir was prepared by the PET center of Huashan Hospital, Fudan University. After resting for 15 min in a quiet environment, the participants received an intravenous injection of 7.4 MBq/kg 18F-florbetapir according to body weight, and PET/CT imaging of the brain was performed 50 min later using a PET/CT scanner (Biograph mCT Flow PET/CT, Siemens, Erlangen, Germany) for 20 min. The PET data were reconstructed using a filtered inverse projection algorithm. Individual PET images were rigidly fused to MRI T1 images with six parameters and spatially normalized. The PET image interpretation was performed independently by three PET diagnostics experts, and the results depended on the agreement of more than two experts.
Aβ-PET semi-quantitative analysis was performed to assess the intracerebral Aβ deposition load. Using the cerebellar peduncle as the reference area for Aβ-PET semi-quantitative analysis, the standardized uptake value ratios (SUVR) of Aβ deposition in eight gray matter cortical brain regions of interest (frontal, lateral parietal, lateral temporal, medial temporal, occipital, basal ganglia, posterior cingulate gyrus, precuneus) were defined based on automatic anatomical localization mapping (a higher SUVR value represents a heavier load of Aβ deposition in the brain) [27 (link)].

One hundred and ninety-five participants were scanned by 18F-florbetapir PET (Biograph mCT Flow PET/CT; Siemens, Erlangen, Germany) at the PET Center of Huashan Hospital of Fudan University within 1 month of recruitment into the study. The subjects received an intravenous injection of 18F-AV-45 at a dose of about 10 mCi (370 MBq) and rested for 50 min. Then, PET imaging was performed for 20 min using low-dose CT. After the acquisition, the filtered back-projection algorithm reconstructed the PET image; attenuation, normalization, dead time, photon attenuation, scattering, and random coincidence were corrected. The results were determined independently by three clinicians who were blinded to the clinical diagnosis. Any differing opinions were resolved using the criterion that the global amyloid standardized uptake value ratio (SUVR, whole gray matter/bilateral cerebellar calf uptake value) was <1.29.

The reconstructed CT scans are with two spatial resolutions of 0.98 × 0.98 × 5 mm³ and 1.37 × 1.37 × 5 mm³, and the reconstructed PET scans are with 4.06 × 4.06 × 5 mm³ and 4.07 × 4.07 × 5 mm³. For all CT slices, the matrix size is 512 × 512, whereas the PET slices have two types 200 × 200 and 168 × 168. Thus, all PET slices were up-sampled in the axial plane, leading to the size of 512 × 512 via the bicubic interpolation algorithm (24 (link)). The reason that we choose the bicubic interpolation algorithm for interpolation lies in its advantage of conserving detailed information, which is vital in the segmentation step. As for the spatial resolution, we remain its diversity unchanged to enhance robustness of the segmentation network. Next, to improve the contrast between lesion area and surrounding soft tissue in CT images, pixel values outside of −150 to 150 were set to −150 and 150. Then PET and CT images were all normalized to the interval of [0, 1]. Last, though PET/CT images had been registered by the hardware of the PET/CT scanner (Siemens Biograph mCT Flow PET/CT), there is slight deviation caused by involuntary respiratory movement of the patient during the image acquisition process. As the focus of this work is not on the registration, here we simply use the multi-mode intensity registration algorithm to correct the deviation (25 ).

Phantom acquisitions were performed on a time-of-flight (TOF) capable whole-body PET/CT; Siemens Biograph mCT Flow PET/CT (Siemens Medical Solutions USA, Inc). We used a PET/CT to facilitate a direct comparison with the reference standard: PET images corrected for attenuation with CT. Our approach uses a custom TX source, detailed in Section II–B. We then describe the supplemental transmission aided attenuation map reconstruction algorithm, including its theory (Section II–C.1), PET data pre-processing (Section II–C.2), method for updating and estimating the emission images (Section II–C.3), approach for attenuation map updates (Section II–C.4), the combined update equation (Section II–C.5), and data corrections (Section II–C.6). Pseudo-code for the complete sTX-MLAA scheme, including scatter corrections, is shown in Algorithm 2. Phantom experiments, data processing, and analysis are detailed in Section II–D. ¹⁸F-fluorodeoxyglucose (FDG) was used for filling the TX source in all experiments.