Region of Interest Motion Compensation in PET Image Reconstruction

Recent literature has seen a number of efforts in respiratory motion correction of PET imaging by considering motion as part of the system model. This type of method however relies on a motion map derived from images (4D PET or 4D CT) of the entire field of view (FOV). In this study we present a region of interest (ROI) based extension to this approach, whereby only the motion map of a user-defined ROI is required and the incorporation of motion into the system model is solely performed within the ROI. A physical phantom study and an NCAT phantom based computer simulation study were performed to test the feasibility of this method. The physical phantom study showed that the proposed ROI based approach had comparable performance within the ROI when comparing with the full image based approach using the same, accurate motion information. The NCAT phantom based study further testified that motion of image features of interest can be estimated more efficiently and potentially more accurately using the ROI based approach.