Hybrid whole-body dynamic TOF PET imaging for simultaneous estimation of compartmental and patlak parametric maps from continuous bed motion data

Abstract

A number of studies have highlighted the clinical role of kinetic modelling following dynamic PET acquisition, since it is able to provide quantitative parameters, more closely related to the underlying pathology than semi-quantitative indices, such as SUV. However, due to the limited axial FOV of current PET scanners, temporally continuous dynamic data are only available for a single bed position. Therefore, full compartmental modelling is axially limited as opposed to whole body imaging protocols which are used in oncology, limiting its applicability in this field. Furthermore, if the blood activity concentration is derived from the images, similar dynamic data of a large blood pool are needed. Lately a new approach to clinical dynamic whole-body imaging has been proposed, consisting of an initial blood pool (cardiac) scan, followed by a number of whole body passes, to estimate whole-body Ki Patlak parametric images. While such an approach improves upon previous whole-body protocols for kinetic parameter estimation, it doesn't fully exploit the information that could be extracted, as well as the benefits of continuous bed motion (CBM) acquisition in dynamic imaging. In this work, we propose a modified CBM whole-body dynamic protocol to simultaneously perform full compartmental modelling in the FOV covering the initial blood pool scan as well as whole-body Patlak analysis. Furthermore, the new protocol does not restrict the initial blood pool scan, for which full compartmental modelling can be performed, to be limited over the heart region but to be chosen based on the localization of the disease within the body. Initial clinical whole-body dynamic scans are reported, along with the estimated micro- and macro-parameter maps. Initial results suggest that not only Ki images from full modelling provide superior tumor to background contrast and improved variance to Patlak Ki but additional micro-parameters are made available to the referring clinician, enabling multi-parametric evaluation of the disease within a clinically feasible dynamic imaging protocol.