MOVING PROPAGATION OF SUSPICIOUS MYOCARDIAL INFARCTION FROM DELAYED ENHANCED CARDIAC IMAGING TO CINE MRI USING HYBRID IMAGE REGISTRATION Yixun Liu 1,2,3 , Hui Xue 1 , Christoph Guetter 1 , Marie-Pierre Jolly 1 , Nikos Chrisochoides 3 and Jens Guehring 1 1 Siemens Corporate Research, Imaging and Visualization, Princeton, NJ, USA 2 Department of Computer Science, The College of William and Mary, Williamsburg, VA, USA 3 Department of Computer Science, Old Dominion University, Norfolk, VA, USA ABSTRACT Cardiac magnetic resonance imaging has proved its effec- tiveness to determine the patient-specific myocardial mo- tion/functional information via the cine imaging and to detect myocardial infarction in the delayed enhanced MRI (DE- MRI). Standard cardiac MR protocols usually acquire these two sets of images across multiple acquisitions with varying imaging slice geometry, pixel spacing and different breath- holdings, which could make the joint inspection of myocar- dial motion and infarction difficult. The purpose of this work is therefore to develop dedicated post-processing algorithms to register DE-MRI to corresponding cine image and propa- gate suspicious infarction to all cardiac phases. Suspicious infarction regions delineated in the DE-MRI can be used to define the region-of-interest for the quantification of regional wall motion abnormality. The proposed approaches are ap- plied to 6 patients and the evaluation shows the feasibility of a joint DE-MRI and cine assessment which can yield clinically valuable outcomes. Index Terms— Image registration, Cardiac magnetic res- onance imaging, Delayed Enhanced MRI, Moving propaga- tion 1. INTRODUCTION Cardiac magnetic resonance imaging has proved its effec- tiveness to determine the patient-specific myocardial mo- tion/functional information via the cine imaging as well as detection of myocardial infarction appearing hyperintense in the DE-MRI. Recent studies compared myocardial tissue viability revealed in the DE-MRI to the functional deficits measured with cine MRI [1], showing the so-called “peri- infarction zone” defined in DE-MRI is correlated well with the dysfunctional myocardial region defined in cine. This information is potentially valuable for reperfusion therapy, as regional motion of infarction zone defined before the therapy is assessed to evaluate the recovery of myocardium. THIS WORK IS SUPPORTED IN PART BY NSF GRANTS: CCF- 0916526, CCF-0833081, AND CSI-719929 AND BY THE JOHN SIMON GUGGENHEIM FOUNDATION. Although the clinical value of joint DE-MRI and cine im- age assessment is exhibited, standard clinical cardiac MR pro- tocols usually acquire two sets of images across multiple mea- surements with variant imaging plane prescription and multi- ple breath-holdings. Misalignment and local deformation of- ten appear between cine and DE-MRI, even imaging plane re- mains unchanged for two acquisitions by careful prescription, mainly due to inconsistent cardiac phases used for acquiring cine and DE-MRI, imperfect cardiac gating and respiratory motion. It is more problematic for patients with arrhythmias as unstable cardiac cycles make it unreliable to identify the matching cine frame acquired in the same cardiac cycle as the DE-MRI. Without an accurate mapping of infarction zone to cine images, regional myocardial changes in motion pattern caused by suspicious scars could only be visually assessed. The accurate alignment and deformation correction between cine and DE-MRI is thus a necessity for the successful joint assessment, where one aim is to propagate the infarction to all other cine frames throughout whole cardiac cycle and enable the quantitative regional motion pattern analysis. There are little research focusing on aligning cardiac MR images acquired across acquisitions with different pulse- sequences, compared to the large amount of studies con- ducted myocardial motion correction within one acquisition. One approach performing joint analysis is to manually seg- ment images from multiple acquisions and the resulting AHA model is matched [2]. Others rely on aligning epicardial sur- faces which are delineated before analysis [3]. More recently, a 2D-3D rigid image registration method was proposed to align DE-MRI and perfusion slices to the 3D whole heart coronary angiography volume [4], which enables the visual- ization of infarction in the 3D context. The contribution of this work is to develop dedicated post- processing algorithms for aligning DE-MRI with correspond- ing cine image and propagating suspicious infarction zone to all other cardiac phases. Infarction regions delineated in the DE-MRI can be used to define the region-of-interest (ROI) for the quantification of regional abnormality of myocardial mo- tion. To achieve these goals, we propose to align DE-MRI to