Left ventricular shape analysis by transesophageal echocardiography

Cardiac shape is an important pathophysiologic adaptation, which has prognostic implications. The aim of this study is to present a novel approach using 3-dimensional (3D) volumetric data set which obtained from the ultrasound images, to perform surface curvature analysis that yields multi- parametric data on regional and global left ventricular (LV) shape. The steps to perform shape analysis consist of 3D acquisition of 2D images, edge extraction, 3D reconstruction, and shape parameter calculation. This yields the following shape parameters: (1) RSC, regional surface change of LV; (2) GSC, global surface curvature of LV; (3) SD, surface distances; (4) ND, normalized distances; (5) ER, effective radius; and (6) ER difference. First, we applied it to six mathematical models to validated the method. The analysis reproduced shape parameters identical to those predicted by mathematical modeling. Then, 3D images obtained (in vitro) of balloons with various shapes were examined. These yielded parameters in concordance with true shapes of the balloons, verified by mathematical modeling. Subsequently, we analyzed left ventricles from humans.