Evaluation of co-registration for PET/CT quality control

2243 Objectives: Fusion of PET and CT data on a hybrid camera must be highly accurate in order to provide precise collaboration of data. The objective of this study is to determine the necessity of a quality control method that would verify the PET/CT scanner’s functionality by looking at co-registration of the fused data. Methods: The NEMA IEC Body Phantom Set (Biodec, Inc.) was used. The Torso cavity weight allowed for the volume to be calculated using the density of water. The volume helped in the determination of using 2 mCi of FDG in the Torso cavity according to the typical clinical activity concentration of 0.14 mCi/mL. The spheres in the Torso cavity were filled with about 90 mCi of FDG and 60 mL of water/Iodinated CT contrast. A Philips Gemini PET/CT scanner (Milipites, CA) was used with the Biopsy CT Abdomen protocol. The Osiris 4(Hospital Universitaires de Geneva) program helped determine the diameter of the spheres using the histogram function and the full-width at half max concept. Results: Upon visual inspection of the data, the CT diameters, on average, were found to be closer to the actual diameters than that of the PET diameters. The CT diameters had between a 30 and 40 percent error, where the PET diameters had an error between 50 and 80 percent. When the 50/50 fused image was evaluated, the error was between 50 and 60 percent. Conclusions: The co-registration of the PET/CT Phantom is poor, which means the fusion of the two sets of data is not very accurate, possibly leading to misread studies. When the 50/50 fused image was analyzed, the image was better than just the PET data, but was worse than the CT data. The poor co-registration observed brings up the issue of using this phantom on a regular basis to ensure that co-registration does not deteriorate.