SIMULTANEOUS EMISSION AND TRANSMISSION SCANNING IN PET ONCOLOGY - THEEFFECT ON PARAMETER-ESTIMATION

We investigated potential sources of bias due to simultaneous emission and transmission (SET) scanning and their effect on parameter estimat ion in dynamic positron emission tomography (PET) oncology studies. Th e sources of bias considered include: i) variation in transmission spi llover (into the emission window) throughout the field of view, ii) in creased scatter arising from rod sources, and iii) inaccurate deadtime correction, Net bias was calculated as a function of the emission cou nt rate and used to predict distortion in [F-18]2-fluoro-2-deoxy-D-glu cose (FDG) and [C-11]thymidine tissue curves simulating the normal liv er and metastatic involvement of the liver. The effect on parameter es timates was assessed by spectral analysis and compartmental modeling. The various sources of bias approximately cancel during the early part of the study when count rate is maximal. Scatter dominates in the lat ter part of the study, causing apparently decreased tracer clearance w hich is more marked for thymidine than for FDG. The irreversible dispo sal rate constant, K-i, was overestimated by <10% for FDG and >30% for thymidine, We conclude that SET has a potential role in dynamic FDG P ET but is not suitable for C-11-labeled compounds.