Fully Automatic, Retrospective Enhancement of Real- Time Acquired Cardiac Cine MR Images Using Image- Based Navigators and Respiratory Motion-Corrected Averaging Peter Kellman, 1 * Christophe Chefd’hotel, 2 Christine H. Lorenz, 3 Christine Mancini, 1 Andrew E. Arai, 1 and Elliot R. McVeigh 4 Real-time imaging may be clinically important in patients with congestive heart failure, arrhythmias, or in pediatric cases. However, real-time imaging typically has compromised spatial and temporal resolution compared with gated, segmented studies. To combine the best features of both types of imaging, a new method is proposed that uses parallel imaging to improve temporal resolution of real-time acquired images at the ex- pense of signal-to-noise ratio (SNR), but then produces an SNR-enhanced cine by means of respiratory motion-corrected averaging of images acquired in real-time over multiple heart- beats while free-breathing. The retrospective processing based on image-based navigators and nonrigid image registration is fully automated. The proposed method was compared with conventional cine images in 21 subjects. The resultant image quality for the proposed method (3.9  0.44) was comparable to the conventional cine (4.2  0.99) on a 5-point scale (P  not significant [n.s.]). The conventional method exhibited degraded image quality in cases of arrhythmias whereas the proposed method had uniformly good quality. Motion-corrected averag- ing of real-time acquired cardiac images provides a means of attaining high-quality cine images with many of the benefits of real-time imaging, such as free-breathing acquisition and tol- erance to arrhythmias. Magn Reson Med 59:771–778, 2008. © 2008 Wiley-Liss, Inc. Key words: MRI; heart; real-time; parallel MRI; SENSE; naviga- tor; motion correction; nonrigid; myocardial function; regional wall motion Real-time imaging may be used to image cardiac function and flow without breathholding or ECG triggering (1– 4), and might be important in patients with congestive heart failure or in pediatric cases. Real-time imaging is also beneficial in cases of arrhythmia. which is problematic for conventional gated, segmented cine imaging. However, real-time imaging typically has compromised spatial and temporal resolution compared with gated, segmented breathheld studies. Real-time imaging with improved spa- tial and/or temporal resolution maybe achieved using par- allel imaging (5– 8) at the expense of trading signal-to- noise ratio (SNR) for increased acquisition speed, thereby reducing image quality. To combine the best features of both types of imaging, a new method is proposed that produces an SNR-enhanced cine from images acquired in real-time by means of respi- ratory motion-corrected averaging. Fully automated, retro- spective processing of real-time acquired images is based on image-based navigators and nonrigid image registra- tion. A similar approach has been validated for ECG-trig- gered, free-breathing delayed enhancement imaging at a single cardiac phase (9,10). Free-breathing cardiac imaging with respiratory gating may be implemented using navigator echoes (11,12), which track the position of the diaphragm. Data are ac- cepted or rejected based on an acceptance window, which is typically several millimeters. Prospective slice-follow- ing techniques (13) improve the image quality by adjusting for heart motion within the acceptance window, thereby allowing a somewhat increased window with higher ac- quisition efficiency. However, the motion of the heart does not strictly track the diaphragm during free-breathing (14), which limits the dependence on simple models and linear motion-correction factors. Furthermore, use of conven- tional navigator echoes is problematic for continuous cine imaging due to the overhead time required to generate the navigator signal and due to interruption of the steady state. Self-navigator approaches have been demonstrated for car- diac cine imaging using radial acquisition (15,16), as well as three-dimensional (3D) Cartesian approaches (17,18), which rely on profiles derived by repeated sampling through the center of k-space. Nevertheless, dependence on a simple motion model still limits the image quality. Self-navigated techniques have not yet been reported for real-time imaging. The proposed approach is based on real-time image acquisition and strictly retrospective processing. In order to combine images from multiple heart beats during free- breathing, a navigator signal derived from the images is used to define an acceptance window. In-plane respirato- ry-induced distortion of the heart within the acceptance window is corrected using a nonrigid deformation. Pro- spective slice-following is not incorporated, therefore 1 Laboratory of Cardiac Energetics, National Heart, Lung and Blood Institute, National Institutes of Health, Department of Health and Human Services, Bethesda, Maryland, USA. 2 Siemens Corporate Research, Princeton, New Jersey, USA. 3 Siemens Corporate Research, Baltimore, Maryland, USA. 4 Johns Hopkins University School of Medicine, Department of Biomedical Engineering, Baltimore, Maryland, USA. Grant sponsor: Intramural Research Program of the NIH, National Heart, Lung and Blood Institute, and a Cooperative Research and Development Agree- ment between the National Heart, Lung and Blood Institute and Siemens Medical Solutions. *Correspondence to: Peter Kellman, Laboratory of Cardiac Energetics, Na- tional Institutes of Health, National Heart, Lung and Blood Institute, 10 Center Drive, MSC-1061, Building 10, Room B1D416, Bethesda, MD. E-mail: kellman@nih.gov Received 2 August 2007; revised 10 October 2007; accepted 20 November 2007. DOI 10.1002/mrm.21509 Published online in Wiley InterScience (www.interscience.wiley.com). Magnetic Resonance in Medicine 59:771–778 (2008) © 2008 Wiley-Liss, Inc. 771