Title: Lesion detectability and quantitative accuracy of PET/CT for different sized lesions at various lesion/background activity ratio in different acquisition and attenuation modes. A phantom study
Abstract:Objectives: The aim of this study is to characterize the effect of lesion size and lesion/background activity ratio (L/B ratio) on detectability by PET/CT and compare quantitative accuracy in differen...Objectives: The aim of this study is to characterize the effect of lesion size and lesion/background activity ratio (L/B ratio) on detectability by PET/CT and compare quantitative accuracy in different acquisition and attenuation modes. Methods: Plastic syringes with diameter of 3.0, 1.5, 1.3, 1.0 and 0.5 cm were fixed in a 20 cm diameter and 30 cm long cylinder phantom at the place about 5 cm off center. Another two 1.0 cm syringes were fixed at the center and verge of the phantom to detect off-center effect. The phantom was filled with 0.25 {mu}Ci/ml 18F-FDG as background. The syringes were filled with 18F-FDG, which is 4.0, 2.0, 1.5, and 1.2 times of background concentration respectively. The phantoms were imaged in a GE Discovery LS PET/CT scanner in 2D CT attenuation, 2D 68Ge attenuation and 3D CT attenuation mode. We use OSEM for 2D data reconstruction, measured method for CT attenuation correction and segmented method for 68Ge attenuation correction. Using CT images to help define the ROI, the tracer concentrations in the syringes were calculated and compared with actual concentrations. Results: All lesions were detected when the L/B ratio was 4 no matter what attenuation mode was used. When the L/B ratio was 2, only syringes of 1.0 cm or larger can be detected. When the L/B ratio is 1.5, 1.5 cm syringe can be detected by 2D CT and 2D 68Ge attenuation method while 3D method can detect 1.3 cm syringe. When the L/B ratio was 1.2, none of the syringes are detected by 2D 68Ge attenuation method. However, 3.0 cm syringe can be detected by 2D CT attenuation and 3D method when L/B ratio was 1.2. No off-center effect can be found in PET/CT. All Methods tend to under-evaluate concentrations in small hot regions, which can reach up to 60% of actual concentration in 0.5 cm region and about 40% in 1.0 cm region. In 2D method, the maximum value in ROI is tend to near the actual concentration at relatively large regions (>1.5 cm). But in 3D method, it tends to over-evaluate the concentrations in large region and background, but under-evaluate the small region concentration. It's better to use the average value in ROI to evaluate the concentrations in 3D mode. Conclusions: PET/CT detectability deteriorates when L/B ratio is less than 1.5. Quantitative errors exist in small regions due to partial volume effect. However, the error patterns are different in 2D and 3D mode. (authors)Read More
Publication Year: 2004
Publication Date: 2004-07-01
Language: en
Type: article
Access and Citation
AI Researcher Chatbot
Get quick answers to your questions about the article from our AI researcher chatbot