Electronic Posters: Cardiovascular

Image Post-Processing

Hall B Monday 14:00-16:00 Computer 38

14:00 3712. Investigation of Myocardium Structure of Postinfarct Porcine Model Using Superquadric Glyphs

Yin Wu^1,2, Ed Xuekui Wu<sup>2,3

1</sup>Institute of Biomedical and Health Engineering, Shenzhen Institute of Advanced Technology, Shenzhen, Guangdong, China; ²Laboratory of Biomedical Imaging and Signal Processing, The University of Hong Kong, Pokfulam, Hong Kong; ³Dept. of Electrical and Electronic Engineering, The University of Hong Kong, Pokfulam, Hong Kong

Previous DTI studies on infarcted LV myocardium structure usually explored myocardial fiber orientation, diffusivity or diffusion anisotropies. In current study, superquadric glyphs were applied on infarcted porcine model. Diffusion tensor shape and laminar sheet structure were examined for the first time to describe infarcted myocardium structural alteration. Results show that significant change of diffusion tensor shape occurred in both infarct and adjacent regions. Apparent alteration of laminar sheet structure was observed in adjacent and remote regions. The current study demonstrates the ability of superquadric glyphs to detect myocardium structural degeneration and provides supplemental information for infarcted heart remodeling.

14:30 3713. Multiecho Dixon Fat and Water Separation Method for Diagnosing Pericardial Disease

Amir H. Davarpanah¹, Aya Kino¹, Kirsi Taimen¹, Philip Hodnet¹, Jeremy Collins¹, Cormac Farrelly¹, Saurabh Shah², Sven Zuehlsdorff², James Carr<sup>1

1</sup>Department of Radiology, Cardiovascular Imaging, Northwestern University, chicago, IL, United States; ²Siemens Medical Solutions, chicago, IL, United States

The VARPRO method (multi-echo gradient echo sequence with iterative fat/water decomposition reconstruction)for fat/water separation performs better than the standard fat saturation protocol currently used at our institution. The water image from this method presents with a more uniform fat suppression.

15:00 3714. Automated Evaluation of Left Ventricular Diastolic Function Using Velocity-Encoded Magnetic Resonance Imaging: Conventional and New Parameters

Emilie Bollache¹, Stéphanie Clément-Guinaudeau², Ludivine Perdrix³, Magalie Ladouceur^1,3, Muriel Lefort¹, Alain De Cesare¹, Alain Herment¹, Benoît Diebold^1,3, Elie Mousseaux^1,2, Nadjia Kachenoura<sup>1

1</sup>INSERM U678, UPMC, Paris, France; ²Radiology department, APHP, European Hospital Georges Pompidou, Paris, France; ³Echocardiography department, APHP, European Hospital Georges Pompidou, Paris, France

Phase-contrast (PC) Magnetic Resonance (MR) is not used in clinical routine to assess diastolic function, because of the lack of automated analyses. Thus, our aim was to develop a process to automatically analyze PC data. Automated segmentation of PC images and analysis of velocity and flow rate curves to derive diastolic parameters were developed and tested on 25 controls. Segmentation was successful in all subjects. Our conventional parameters were consistent with those previously presented in literature and our new parameters highly correlated with high prognosis value parameters. Our process may provide a valuable addition to the established cardiac MR tools.

15:30 3715. Unsupervised and Reproducible Image-Based Identification of Cardiac Phases in Cine SSFP MRI

Sotirios A. Tsaftaris^1,2, Xiangzhi Zhou², Richard Tang², Rohan Dharmakumar<sup>2

1</sup>Electrical Engineering and Computer Science, Northwestern University, Evanston, IL, United States; ²Radiology, Northwestern University, Chicago, IL, United States

It is particularly important for the evaluation of cardiac phase-resolved myocardial blood-oxygen-level-dependent (BOLD) MRI studies, to robustly and reproducibly identify end-systolic (ES) and end-diastolic (ED). Most automated methods rely on identifying the minimum and maximum of the blood pool area in the Left Ventricle chamber, but they are computationally intensive, susceptible to noise, and require prior localization and segmentation of the chamber. The purpose of this work is to develop automated methods to facilitate in the robust and reproducible evaluation of cardiac phase-resolved myocardial BOLD MRI through identification of ES and ED images.