Segmentation and analysis of 3D volume images
Juha Ylä-Jääski, O. Kübler · 2003
An algorithm is presented for the segmentation of volume data acquired in medicine by computer tomography (CT) or magnetic resonance (MR) imaging. The algorithm is based on edge detection by a 3-D implementation of the difference of Gaussians or the Laplace of Gaussians operator followed by an object refinement procedure. This consists of matched filters and/or a decision strategy which is included in a 3-D connected-component labelling procedure. Results from the analysis of segmented CT and MR data are presented utilizing an advanced display facility and quantitative 3-D analysis techniques.>