A Semiautomated Approach for the Quantification of the Left Ventricle Chamber Volumes From Cine Magnetic Resonance Images

Claudio Fabbri1, Keigo Kawaji2, Noreen Nazir2, Victor Mor Avi2, Amit Patel2, Cristiana Corsi3
1University of Bologna, 2University of Chicago, 3DEI- University of Bologna


Cardiac Magnetic Resonance is the golden standard for the quantification of left ventricular volumes (LV) from 3D Cine sequences. However manual segmentation of the endocardium and epicardium boundaries has proven to be a time-consuming task, and the tools used in in clinical practice depend on the cardiologist experience to assess the mitral valve plane, identifying papillary muscles and trabeculations. The goal of this work was to develop a new algorithm for the segmentation of the endocardium and epicardium of the left ventricle, to this end both a probabilistic and a Malladi-Sethian level-set models were used, followed by morphological operators and curvature flow to regularize the surfaces and exclude unconnected regions. The proposed algorithm and the ensuing workflow were applied on CMR data from the University of Chicago Cardiac Imaging Center in 10 patients. Bland-Altman analysis for end-diastolic and end-systolic volumes was conducted, and clinical indexes derived from the surfaces were in good agreement with the manual tracings. This work provides the basis for faster and still accurate quantification of the cardiac function from Cine MR, and the basis for further processing of MR data for the assessment of heart remodeling and diseases.