Abstract
2519
Objectives In order to diagnosis the malignancy of tumor, SUV is used in 18F-FDG PET/CT, but the accuracy decreases by the motion of patient's respiration. To correct this, SUV and image quality were analyzed after obtaining the amplitude based respiratory-gating images and comparing with reconstruction parameter combination.
Methods Infrared rays type respiration motion tracing devise was installed on PET/CT scanner and experiments was conducted with respiratory-gating mode. 18F-FDG injected point (capillary tube) and performance (hot-cold-BKG) source were installed on the motion phantom which enables verrtical movement and 3 min/bed scan was conducted. Images before correction are static type, images after correction are gating type and FBP reconstruction was conducted after setting the amplitude changes and intervals according to motion cycle with 9 combinations of wave forms (Min, Max, Rest In/Ex) and duty cycle (20, 35, 50%) parameters. And differences between SUVmax, volume, FWHM, SNR and CNR were compared and statistically analyzed.
Results When comparing pre-images of respiration motion correction with post-images, SUVmax increased 89.9% from 1.38 to 2.62±0.48 and volume decreased 27.9% from 2.94 cm3 to 2.12±0.28 cm3. Spatial resolution evaluated by FWHM decreased 53.9% from 13.3 ㎜ to 6.12 ㎜ and SNR increased 11% from 5.36 to 5.95, and CNR increased 16.3% from 57.28 to 66.64. Therefore, when applying respiration correction technique, SUVmax, volume and image quality could be improved (P<0.01). Additionally, it could be identified SUVmax, volume and image quality were optimal in the parameter combination of rest-wave form and 35% duty cycle.
Conclusions When comparing with the images in existing scans, respiration corrected images that selectively reconstruct the signals in the intervals with little amplitude change rates and long lasting time can correct SUV changes and image distortion caused by respiration motion more precisely and image quality is excellent.