Segmentation and Analysis of 3D Cardiac Motion from Tagged MRI Images

This paper gives a n overview of o u r f ramework for the automated segmentation and motion analysis of cardiac motion from MRI tagging lines. I t consists of a series of novel methods which utilize theory from image processing, deformable models a n d finite elements. O u r f ramework consists of several steps. I n the first step we use G a b o r Filter banks a n d deformable models for the automatic segmentation of tagging lines a n d cardiac boundaries. T h e extracted tagging lines and boundaries a r e then used as input t o a volumetric deformable model for the heart 's motion estimation analysis. I n this step we first extract parameters that can determine the difference between a normal a n d a pathologic hear t motion. Second, using an Expectation-Maximization methodology (EM) we a r e able to de te rmine a given heart 's stress-strain relationship a n d fiber orientation. O u r hypothesis is that the 3D shape and motion analysis of the hear t will allow the faster a n d timely diagnosis of hear t disease compared to tradit ional 2D methods. W e present a series of segmentation, shape, motion a n d tissue proper ty analysis results. Keywords-3D H e a r t motion analysis, deformable models, tagging line segmentation, tagged M R I images