Clinical implications of different image reconstruction parameters for interpretation of whole-body PET studies in cancer patients

摘要: The standardized uptake value (SUV) is the most commonly used parameter to quantify intensity of radiotracer in tumors. Previous studies suggested that measurements 18F-FDG accumulation tissue might be affected by image reconstruction method, but clinical relevance these findings has not been assessed. Methods: Phantom were performed and whole-body PET images 85 cancer patients analyzed. All reconstructed using either filtered backprojection (FBP) with measured attenuation correction (MAC) or iterative (IR) segmented (SAC). In a subset 15 patients, all 4 combinations IR+SAC, IR+MAC, FBP+SAC, FBP+MAC. For phantom studies, sphere containing was placed water-filled cylinder activity concentration FBP IR combinations. Clinical displayed simultaneously identical regions interest (ROIs, 50 pixels) liver, urinary bladder, tumor both sets. SUV max (maximal counts per pixel ROI) avg (average pixel) measured. Results: from underestimated true greater degree than those (20% vs. 5% underestimation). patient derived consistently lower normal tissue: Tumor IR+SAC 9.6 ± 4.5, IR+MAC it 7.7 3.5, FBP+MAC 6.9 3.0, FBP+SAC 8.6 4.1 (all P