Effects of motion estimation methods on 4D gated cardiac SPECT reconstruction

Cardiac gated SPECT imaging suffers from several degrading factors including limited data counts, depth-dependent blur, attenuation, scatter, and motion blur. In recent work we demonstrated that use of temporal processing could lead to the most improvement in 4D reconstruction. In this work we investigate how motion estimation models can affect the reconstruction in 4D, and in particular, we examine how much room is still left in optical flow estimation toward improving the image quality of reconstruction. In our experiments we conducted a thorough, quantitative evaluation using simulated imaging study with multiple noise realizations from the NCAT phantom. As an upper bound, the known motion from NCAT was used for reconstruction. The results show that 4D reconstruction with a periodic optical flow model can achieve results almost matching that with the known motion.