My talk will focus on two important aspects of Positron Emission Tomography (PET):  (i) Motion-compensation, and (ii) Pharmacokinetic analysis of dynamic PET images.

Motion-compensation in Dynamic PET Images: Dynamic PET images are degraded by inter-frame and intra-frame motion artifacts that can affect the quantitative and qualitative analysis of acquired PET data. A novel approach called Generalized Inter-frame and Intra-frame Motion Correction (GIIMC) algorithm is presented that unifies in one framework the inter-frame motion correction capability of Multiple Acquisition Frames and the intra-frame motion correction feature of (MLEM)-type Deconvolution methods. GIIMC employs a fairly simple but new approach of using time-weighted average of attenuation sinograms to reconstruct dynamic frames. A mean-motion threshold is also provided for individual frames to construct a framing sequence. Extensive simulation and human studies show that GIIMC algorithm outperforms conventional techniques producing images with superior quality and quantitative accuracy. 

Noise-reduction in Quantitative Myocardial Perfusion PET Imaging: A novel framework is presented for robust kinetic parameter estimation at the individual voxel level applied to absolute flow quantification in dynamic myocardial perfusion (MP) PET. Kinetic parameter estimation is formulated as nonlinear least squares with spatial constraints where the constraints are computed from physiologically driven clustering of dynamic images. The proposed framework improves quantitative accuracy, and has long-term potential to enhance capabilities of MP PET in the detection, staging and management of coronary artery disease.