PET | Variation of measurements of standardized uptake value (SUV) using dif…

1Tae Yeob Kim,  2Joong Hyun Kim,  1Jeom Jin Lim,  1Hong Jae  Lee, 1Hyun Joo Kim, 1Jae  Sung Lee,

1Dong Soo Lee

1Dept. of Nuclear Medicine, Seoul National Univ. Hospital

2Dept. of Nuclear Medicine, Seoul National U

Objective

Standardized uptake value (SUV) is a widely used semi-quantitative  index in PET image analysis for the estimation of radiotracer accumulation in an interested region or volume. As mechanical characteristics of  scanners are critical  in  the  measurement  of  SUV,  variation  between  different scanners should be considered in big institutes. In this study, SUVs of a phantom  were measured using three different PET/CT scanners, and variation between them was evaluated.

Methods

The PET/CT scanners which were assessed in this study were GEMINI,  GEMINI TF64 (Philips) and Biograph TruePoint 40 (Siemens Medical). A NEMA PET phantom  (Data Spectrum Corp., USA) was used to  measure SUVs.  The phantom  has weight  of  6.8 kg  and three  hot  inserts.  Two sets of activity distributions  for the background and inserts were tested; first, 3.7:3.7:7.4:11.1 MBq (1:1:2:3), and  second,  1.85:7.4:9.25:11.1MBq   (1:4:5:6),  for  background,  insert  1,  insert  2  and  insert  3, respectively. Images were acquired for 2 bed positions with scan time of 2 min for each bed. SUVs for  background  and each insert  were measured using  each PET/CT  scanner and user-provided software.  The measurements of SUV were compared between scanners and with the theoretically calculated true values.

Results

For both  the sets of  activity  distributions,  all the scanners showed similar results.  The variation between each scanner was statistically insignificant. The measurements of SUV were 91.2%, 85.9% and 87.2% of true SUVs, for GEMINI, GEMINI TF64, Biograph TruePoint 40, respectively.

Conclusion

For  all  the  tested  scanners,  variation  between  the  scanners was  insignificant   although   the measured SUVs were slightly  lower than the true values. Therefore,  measurements of SUV using the tested scanners would be interchangeable in big institutes operating diverse PET/CT scanners. However, clinical validation is required as further studies