Performance characteristics of the digital uMI550 PET/CT system according to the NEMA NU2-2018 standard

Background The aim of this study is to conduct physical performance evaluation on the uMI550 whole-body PET/CT system according to the NEMA NU2-2018 standard. Methods According to the NEMA NU2-2018, spatial resolution, sensitivity, scatter fraction, count-rate performance, accuracy of count losses and random corrections, image quality, and timing resolution were evaluated. Spatial resolution was measured by using a(22)Na point source. System sensitivity was measured by inserting an(18)F line-source in six concentric aluminum sleeves with varying diameters. Scatter fraction, count-rate performance, accuracy of count loss, and timing resolution were all calculated by analyzing dynamically acquired data of an(18)F line-source inside a polyethylene cylinder in 20 cm diameter and 70 cm length. Image quality was assessed using a NEMA IEC body phantom with a 4:1 ratio of activity concentration of spheres to the warm background. Additionally, three patient studies were performed, with one brain scan and two whole-body scans, separately. The patient images were evaluated to get a visual first impression of uMI550. Results The tangential, radial, and axial spatial resolutions were measured as 2.91 mm, 2.98 mm, and 2.97 mm FWHM, respectively, at 1 cm radial offset. The total system sensitivity to line source at center was 10.24 cps/kBq. A NECR peak was measured as 124.4 kcps at 18.85 kBq/mL. The scatter fraction at NECR peak was 36.65%, and the maximum count-rate error at and below NEC peak was 1.55%. Contrast recovery coefficients were from 46.5 (10 mm) to 83.9% (37 mm). The timing resolution was measured as 372 ps at low count rate. Conclusion NEMA NU-2 2018 testing was performed on the new SiPM-based uMI550 PET/CT system. The uMI550 shows a high-spatial resolution of less than 3 mm and a good timing resolution of 372 ps. It shows clinical significances on improving potentially diagnostic ability on small lesions.