Purpose: We evaluated the impact of simultaneous emission and transmission (SET) measurements of quantification and noise in neurological PET studies.
Method: Bias in SET was measured as a function of emission count rate and used to predict distortion in simulated FDG tissue curves and its effect on model parameter estimates. Studies were performed on a brain phantom and a patient to verify predicted bias and examine the effect of SET on noise.
Results: In static imaging, SET underestimated tracer concentration by approximately 2%. In kinetic studies, tracer concentration was overestimated initially and underestimated during the mid to late part of the study, but bias in measurement of glucose metabolic rate was < 5% by simulation and < 10% in the patient study. SET imaging takes 10% longer than the emission part of a conventional scan to achieve comparable statistics.
Conclusion: Accurate neurological PET studies can be performed with SET. The relatively small bias can be predicted and potentially corrected.