Exploiting an MRI derived arterial input function to improve the PET simultaneous estimation method: Validation of assumptions

Abstract

Kinetic analysis of dynamic PET data requires a correct knowledge of the arterial input function (AIF) which is often not straightforward to measure. An alternative method to invasive blood sampling is the simultaneous estimation (SIME) method, where the AIF and kinetic parameters are estimated at the same time using kinetic information from different regions. We have previously shown that if the total number of parameters estimated in the SIME optimization can be reduced in SIME method the bias of estimates can be reduced significantly. In this work, we compared four pairs of AIFs derived from 18F-Choline PET and DSC-MRI studies. For all patients, similar bolus shapes were observed once both PET-AIF and MRI-AIF were brought to same injection duration. Parameters representing the early part of the AIF were found to be convertible between modalities.