Performance Evaluation of the FLEX Triumph X-PET Scanner Using the National Electrical Manufacturers Association NU-4 Standards

The purpose of this work was to evaluate the performance characteristics of the preclinical X-PET subsystem of the FLEX Triumph PET/CT scanner based on the NU 4-2008 standards of the National Electrical Manufacturers Association (NEMA). Methods: The performance parameters evaluated include the spatial resolution, scatter fraction, count losses and random coincidences, sensitivity, and image-quality characteristics. The PET detector array consisted of 11,520 individual bismuth germanate crystals arranged in 48 rings and 180 blocks, with an axial field of view (FOV) of 11.6 cm and a inner ring diameter of 16.5 cm. The spatial resolution was measured with a small Na-22 point source (diameter, 0.25 mm) at different radial offsets from the center. Sensitivity was calculated using the same source by stepping the source axially through the axial FOV of the scanner. Scatter fraction and counting-rate performances were determined using a mouse-and rat-sized phantom with an F-18 line source insert. The NEMA image-quality phantom and rodent imaging were also performed to access the overall imaging capabilities of the scanner. Results: Tangential spatial resolution in terms of full width at half maximum varied between 2.2 mm at the center of the FOV and 2.3 mm at a radial offset of 2.5 cm. The radial spatial resolution varied between 2.0 at the center and 4.4 mm at a radial offset of 2.3 cm. The peak system absolute sensitivity was 5.9% at the center of the FOV. The absolute system sensitivity was 0.67 counts/s/Bq, and the relative total system sensitivity was 73.9%. The scatter fraction for the mouse-sized phantom was 7.9%, with a peak true counting rate of 168 kilocounts per second (kcps) at 0.3 MBq/mL and a peak noise-equivalent counting rate of 106 kcps at 0.17 MBq/ mL. The rat-sized phantom had a scatter fraction of 21%, with a peak true counting rate of 93 kcps at 0.034 MBq/mL and a peak noise-equivalent counting rate of 49 kcps at 0.02 MBq/mL. Recovery coefficients for the image-quality phantom ranged from 0.13 to 0.88. Conclusion: The performance of the X-PET scanner based on the NEMA NU 4-2008 standards was fully characterized. The overall performance demonstrates that the X-PET system is suitable for preclinical research.