Computerized Analysis Of The Cross Calibration Correction Factor (Cccf) Test For Enhanced Follow-Up Of The Performance Of Pet/Ct Systems

1509 Objectives: The NEMA acceptance test, a well-established tool to assess the performance of PET/CT systems, is dependent on the operator skills and the quality of test technique and requires expensive and not commonly available phantoms. This test is therefore performed usually once in a life time of a system or once a year at the most. Cross Calibration Correction Factor (CCCF) test is currently used only for the periodic calculation of CCCF to assure calibration of the PET system to the dose calibrator that measures patient injected dose, and to assess the uniformity of the reconstructed PET/CT image. The current study suggests a standardized computerized methodology for measurement and analysis of CCCF test results, which provides extended data of additional parameters for enhanced follow-up of the performance of PET/CT systems.\n Methods: The CCCF test was performed following the vendor9s protocol on a clinical PET/CT system (Biograph mCT 40S, 4R, Siemens Medical Solutions USA, Inc.). Extended data derived from the CCCF test results were analyzed to calculate: a. Single-to-noise ratio (SNR) of the images as an index of homogeneity of FDG distribution and measurement, b. Sensitivity of the PET system along the axial plane and overall, and c. The quality of attenuation correction. Computerized Software was developed to ensure standardization of calculations and consistency of extended test results. Results: After performing CCCF test with the vendor9s cylindrical water phantom (6283ml, radius 100mm,~1.6 mCi) additional parameters were derived from the test results using the developed software: a. The average standard deviation (SD) of activity concentration (KBq/ml) in each slice was calculated and plotted (Figure 1A). The average SD of activity concentration (KBq/ml) in the ROIs divided by the average was 0.14% which is the SNR (Figure 1B, red circle). The histogram of activity concentration distribution in the phantom showed a normal Gaussian curve (Figure 1C). Both parameters indicate high homogeneity of FDG distribution and measurement. b. The proportional sensitivity of the system along the axial plane, which is the inversed square root of SD per slice, was plotted (Figure 1D). Overall sensitivity of the system was calculated by dividing the number of counts by time duration of the acquisition (cps) and by the acquisition calibrated injected dose, and was 7.85 (cps/KBq). This value is proportional to the NEMA sensitivity (Figure 1B, red box). c. The quality of attenuation correction was evaluated by measuring activity concentration (KBq/ml) along a midline cross section image of the phantom. Visual evaluation of the derived plot confirmed a normal step function curve (Figure 1E). The developed Software performed all calculations and provided an automatic structured report of results (figure 1B).Conclusions: Software was developed and used for extended analysis of CCCF test results of a PET/CT system. This simple, readily available and standardized methodology provided additional information on homogeneity of FDG distribution and measurement, on PET/CT system sensitivity and on the quality of the attenuation correction of images. The extended data derived from CCCF test results indicate the potential use of this methodology for a continuous long term follow-up of the performance of PET/CT systems in clinical routine.