Abstract Motion effects (including respiratory motion, body motion, cardiac motion) and associated mismatched attenuation correction substantially degrade the cardiac PET image quality and quantitative accuracy. Although a number of motion correction methods have been implemented on clinical scanners, they 1) are largely respiratory motion corrections requiring external motion tracking hardware; and 2) does not take into account motion-induced PET-CT mismatch in attenuation correction. The correction of motion for 82Rb cardiac dynamic PET imaging is particularly challenging, as the rapid tracer kinetics of 82Rb leads to substantial tracer distribution change in the dynamic frame images over time. 82Rb’s ultra-short half-life of only 75 seconds also poses additional significant challenges of image noise. In this R01, we propose to develop a series of motion correction methods with aligned PET-CT for both static and dynamic cardiac applications. All the new developments proposed here are data-driven based approaches, meaning no external motion tracking hardware is required. We will focus on cardiac PET data with 82Rb as the tracer, while most of the developments can be directly applied to other tracers. The proposed motion correction technology developments and translations will have a significant impact on both currently used clinical protocols and emerging new applications, such as evaluation of endocardial/epicardial (EN/EP) flow ratios. Specifically, in Aim 1, we will develop and optimize respiratory and bulk body motion correction methods. In Aim 2, we will develop and optimize data-driven methods to correct cardiac (intra-beat) motion. In Aim 3, we will develop methods to correct PET/CT mis-alignment due to motion. All the developed technologies will be evaluated and validated using human and large animal data. In Aim 4, the industry partner of this grant will translate the developed motion correction methods to end-users.