Parametric Imaging by Mixture Analysis in 3D Validation for Dual-Tracer Glucose Studies

The quantitative analysis of dynamic positron emission tomography (PET) data to obtain estimates of tissue characteristics, such as blood flow, energy consumption, or receptor density, usually relies on fitting an appropriate kinetic model to the radiotracer time course data. This chapter presents a technique for constructing parametric images from three-dimensional dynamic PET studies. The approach is based on a mixture analysis model in which the time activity curve (TAC) at a given volume element (voxel) is expressed as a weighted sum of sub-TACs corresponding to homogeneous tissues represented there. Estimates of metabolic parameters at a voxel are defined as a weighted sum of the parameters associated with the individual sub-TACs. Segmentation plays a key role in the methodology. This chapter also presents an overview of the implementation of the approach illustrated by the application to a dual-tracer study designed to measure the local cerebral glucose lumped constant.