Significant socio-economic burden of colonic motility disorders necessitates in-depth analysis of this pathology. Current analysis techniques are based on diameter measurements of colonic lumen on cine-MR images. Interleaved multi-plane acquisition makes it difficult to perform simultaneous measurements on the line of plane intersections due to the out-of-plane motion (OPM) caused by respiration affecting the underlying anatomy. Low temporal acquisition rate and dark-banding artifact are the challenging factors for OPM compensation. In this paper, we propose the use of manifold learning in combination with in-plane motion tracking for estimating OPM. We evaluate the effectiveness of our approach on 8 MR patient data sets. Experimental results show the good performance of our approach. The proposed method is independent of the acquisition rate and is not limited to this specific application.
«