Journal of Mathematical Imaging and Vision
Phase contrast MRI segmentation using velocity and intensity
Scale-Space'05 Proceedings of the 5th international conference on Scale Space and PDE Methods in Computer Vision
| Hi-index | 0.00 |
We present a velocity-constrained front propagation approachfor myocardium segmentation from magnetic resonanceintensity image (MRI) and its matching phase contrastvelocity (PCV) images. Our curve evolution criterionis dependent on the prior probability distribution of the myocardialboundary and the conditional boundary probabilitydistribution, which is constructed from the MRI intensitygradient, the PCV magnitude, and the local phase coherenceof the PCV direction. A two-step boundary findingstrategy is employed to facilitate the computation. For thefirst image frame, a gradient-only fast marching/level setstep is used to approach the boundary, and a narrowbandis formed around the curve. The initial boundary is then refinedusing the full information from priors and all threeimage sources. For the other frames, the resulting contoursfrom the previous frames are used as the initializationcontours, and only refinement step is needed. Experimentresults from canine MRI sequence are presented, and arecompared to results from gradient-only segmentation.