Cardiac Motion Recovery: Continuous Dynamics, Discrete Measurements, and Optimal Estimation

A sampled-data filtering framework is presented for cardiac motion recovery from periodic medical image sequences. Cardiac dynamics is a continuously evolving physiological process, whereas the imaging data can provide only sampled measurements at discrete time instants. Stochastic multi-frame filtering frameworks are constructed to couple the continuous cardiac dynamics with the discrete measurements, and to deal with the parameter uncertainty of the biomechanical constraining model and the noisy nature of the imaging data in a coordinated fashion. The state estimates are predicted according to the continuous-time state equation between observation time points, and then updated with the new measurements obtained at discrete time instants, yielding physically more meaningful and more accurate estimation results. Both continuous-discrete Kalman filter and sampled-data Hinfinity filter are applied, and the Hinfinity scheme can give robust estimation results when the noise statistics is not available a priori. The sampled-data estimation strategies are validated through synthetic data experiments to illustrate their advantages and on canine MR phase contrast images to show their clinical relevance.