Henrik Pedersen1, Henrik B.W. Larsson1, Rasmus Larsen2
1Functional Imaging Unit, Glostrup Hospital, Glostrup, Denmark; 2Image Analysis DTU Informatics, Technical University of Denmark, Lyngby, Denmark
Prediction of the respiratory and beating motion of the heart has several useful applications in cardiovascular MRI. Despite its usefulness, there is currently no available technique that establishes a predictive model of both cardiac and respiratory 3D motion. The main bottleneck is the inherent slow nature of MRI that prevents obtaining real-time 3D images of the heart with both high spatial and temporal resolution. In this work, we present a novel technique that predicts 2D real-time images for any cardiac and respiratory motion state. By combining several 2D acquisitions, the method predicts 3D images of the heart, including through-plane motion.