Evaluation of PET Scanner Performance in PET/MR and PET/CT Systems: NEMA Tests

Abstract

Objective:The aim of the present study was to compare the performance of positron emission tomography (PET) component of PET/computed tomography (CT) with new emerging PET/magnetic resonance (MR) of the same vendor.Methods:According to National Electrical Manufacturers Association NU2-07, five separate experimental tests were performed to evaluate the performance of PET scanner of General Electric GE company; SIGNATM model PET/MR and GE Discovery 710 model PET/CT. The main investigated aspects were spatial resolution, sensitivity, scatter fraction, count rate performance, image quality, count loss and random events correction accuracy.Results:The findings of this study demonstrated superior sensitivity (~ 4 folds) of PET scanner in PET/MR compared to PET/CT system. Image quality test exhibited higher contrast in PET/MR (~ 9%) compared with PET/CT. The scatter fraction of PET/MR was 43.4% at noise equivalent count rate (NECR) peak of 218 kcps and the corresponding activity concentration was 17.7 kBq/cc. Whereas the scatter fraction of PET/CT was found as 39.2% at NECR peak of 72 kcps and activity concentration of 24.3 kBq/cc. The percentage error of the random event correction accuracy was 3.4% and 3.1% in PET/MR and PET/CT, respectively.Conclusion:It was concluded that PET/MR system is about 4 times more sensitive than PET/CT, and the contrast of hot lesions in PET/MR was ~ 9% higher than PET/CT. These outcomes also emphasize the possibility to achieve excellent clinical PET images with low administered dose and/or a short acquisition time in PET/MR.

Biodistribution and radiation dosimetry of multiple tracers on total-body positron emission tomography/computed tomography

“…Taking into account the partial volume effect, the delineated volume of interest (VOI) was slightly smaller than the actual size of the organ. Tracer kinetics and cumulative organ activity were calculated using VOI to generate time-activity curves to visualize changes in subject’s organ uptake and calculate internal radiation dose ( 16 ). Data and time-activity curves of heart were derived from the left ventricle.…”

Validation of low-dose lung cancer PET-CT protocol and PET image improvement using machine learning

“…Acquisition times and/or injected activities should be proportionally adjusted to obtain the counting statistics presented in table 1 with scanners less sensitive. However, due to faster electronics and increased axial field of view, most of the modern scanners offer similar or higher performance with sensitivities ranging from 13.3 cps/kBq for the Siemens synchronous PET-MR scanner and the GE DMI PET/CT system [23], to more than 20 cps/kBq for the SIGNATM [24] and the DIQ PET/CT system [25] from General Electric. Higher sensitivities would potentially allow further reduction of the acquisition time and/or of the injected dose while preserving the image quality.…”

Impact of brain segmentation methods on regional metabolism quantification in 18F-FDG PET/MR analysis

“…However, in many clinical workflows, PET studies are often performed without the aid of individually matched structural MRI scans, primarily for the sake of convenience in the data collection and brain segmentation possesses. Currently, two commonly employed segmentation strategies for brain PET analysis are distinguished: methods with or without MRI registration and methods employing either atlas-based or individual-based algorithms [ 9 ]. These strategies can be implemented using well-established brain automatic analysis tools, including statistical parametric mapping (SPM) [ 10 ], FMRIB software library (FSL) [ 11 ], and FreeSurfer [ 12 ].…”