Dr. Roger GUNN, McConnell Brain Imaging Center

The development of Positron Emission Tomography (PET) over the last two decades has provided neuroscientists with a unique tool for investigating the neurochemistry of the human brain in vivo. Injection of an appropriate positron emitting radiotracer and measurement of the resulting 4D spatio-temporal distribution using a PET tomograph allows for the estimation of quantitative biological images e.g. blood flow, metabolism, and receptor-ligand interactions. The generation of these images requires the application of appropriate tracer kinetic modeling techniques which extract relevant biological parameters from the kinetic data. Traditionally, kinetic analysis in PET has been based on the a priori selection of a particular compartmental model from which parameter estimates are obtained using non-linear least squares methods. However, these methods are slow (which restricts parameter estimation to gross anatomical regions) and assume that the selected model is correct. This talk presents a new technique which is based on a general compartmental description of the tracer's fate in vivo and determines a parsimonious model consistent with the measured data. The technique involves the determination of a sparse selection of kinetic basis functions from an overcomplete dictionary using the method of basis pursuit denoising. The system parameters are readily calculated from the estimated impulse response function and are easily interpreted in a biological framework. The method is expedient which permits the generation of parametric images by application of the method to individual voxel time activity curves.