Medial-node models to identify and measure objects in real-time 3-D echocardiography

A method is proposed for the automatic, rapid, and stable identification an d measurement of objects in three-dimensional (3-D) images. It is based on local shape properties derived statistically from populations of medial pri mitives sought throughout the image space, These shape properties are measu red at medial locations within the object and include scale, orientation, e ndness, and medial dimensionality, Medial dimensionality is a local shape p roperty differentiating sphere-like, cylinder-like, and slab-like structure s, with intermediate dimensionality also possible. Endness is a property fo und at the cap of a cylinder or the edge of a slab. In terms of an applicat ion, the cardiac left ventricle (LV) during systole is modeled as a large d ark cylinder with an apical cap, terminated at the other end by a thin brig ht slab-like mitral valve (MV), Such a model, containing medial shape prope rties at just a few locations, along with the relative distances and orient ations between these locations, is intuitive and robust and permits automat ed detection of the LV axis in vivo, using real-time 3-D (RT3D) echocardiog raphy, The statistical nature of these shape properties allows their extrac tion, even in the presence of noise, and permits statistical geometric meas urements without exact delineation of boundaries, as demonstrated in determ ining the volume of balloons in RT3D scans, The inherent high speed of the method is appropriate for real-time clinical use.