Quantitative characterization of mitral annulus and leaflets from transesophageal 3D echocardiography

We propose and validate a novel algorithm for the segmentation and quantification of the mitral annulus (MA) and mitral leaflets (ML) from transesophageal real-time 3D echocardiography volumes in a closed mitral valve (MV) configuration. Following initialization of 8 MA points and of the coaptation line, the MA and the anterior and posterior ML are automatically obtained in 3D. Once segmented, several morphological parameters, including local ML thickness and tenting, are obtained. Validation versus manual tracings was performed in 33 patients: 9 controls, with normal MA dimensions, 12 patients with dilative cardiomyopathy and 12 after MV repair with insertion of annuloplasty ring on the posterior MA region. MA and ML segmentations showed a high level of accuracy when compared with manual tracings, with errors of the order of the voxel size. Computed parameters were comparable with those found in literature for healthy MV. The potential clinical applicability to different MV pathologies, as well as repaired valves with implanted annular rings, was favorably tested.