An implementation of a quality assurance method using a cylindrical 68GE Phantom for measuring spatial resolution in PET

Abstract

Background: Important information regarding spatial resolution can be ob- tained from the daily uniformity quality control test performed on a Positron Emission Tomography (PET) scanner using a cylindrical 68Ge phantom. This information can be used to assess scanner spatial resolution over time and to pro- vide multicenter clinical trials with further data for comparing scanners to ensure inter-center image quality. Objectives: To compare two methods for obtaining spatial resolution estimates from the daily uniformity quality control image of a cylindrical 68Ge phantom and to investigate the short and long term repeatability and robustness to small changes in acquisition and reconstruction parameters. Research Design: The value of spatial resolution obtained for each of two scanners (GE Discovery 690 and Philips Gemini TF) using the NEMA NU2 (2007) protocol was used as a reference value for comparison to those obtained from the trial methods. Of the two methods investigated, one included applying Fourier analysis to a transaxial image of the cylindrical 68Ge phantom, and the other applying an edge spread function to a coronal image slice from the same phantom acquisition. Results and Conclusions: Using the Discovery 690's daily quality control pa- rameters, the Fourier analysis method gave a spatial resolution estimate of 5.39 0.09 mm, whilst the edge spread function method gave 9.67 0.47 mm. Both methods were found to be repeatable and robust. They were also found to be sensitive to changes in acquisition and reconstruction parameters, and degrada- tion in scanner spatial resolution. The Fourier analysis technique was found to be independent of position in the FOV, and dependent on count density.