An improved periodic optical flow model for cardiac gated image reconstruction

Recently we developed a periodic modeling approach for determining the optical flow in a periodic image sequence, which is demonstrated to be beneficial for noise reduction in motion-compensated 4D reconstruction of cardiac gated images. In this approach, a Fourier harmonic model is used to exploit the temporal continuity and periodicity of the motion field in the sequence. In this work, we further develop this approach by exploring the use of a piecewise spatial smoothness constraint (in the form of total variation) on the motion field, which can better accommodate the discontinuity of the motion field at an object boundary. In the experiments, we demonstrated this approach for 4D reconstruction of gated cardiac SPECT images. Our results show that it could lead to further improved reconstruction of the myocardium in spite of strong imaging noise.