Meeting Banner
Abstract #1903

Unsupervised Fully Automated Inline Analysis of Global Left Ventricular Function in MR Imaging

Daniel Theisen1, Torleif Sandner1, Kerstin Bauner1, Carmel Hayes2, Maximilian F. Reiser1, Bernd J. Wintersperger1

1Department of Radiology, University of Munich, Munich, Germany; 2MR Division, Siemens Health Care, Erlangen, Germany

The purpose of this study was to implement a fully automated segmentation algorithm for the evaluation of the left ventricle within cine data acquisition and to compare the evaluated parameters of global left ventricular function to manual analysis as standard of reference in patients and healthy volunteers. There was a good agreement between automatic and manual results for EF, EDV, ESV and myocardial mass. Volumetric results of the automated inline analysis showed high correlation with results of manual segmentation. Unsupervised fully automated segmentation and contouring during image reconstruction enables accurate and instantaneous evaluation of global left ventricular performance.