In medical visualization, multiple isosurfaces are usually extracted from medical volume image and used to represent (approximate) the boundary surfaces of different structures in the image. In this paper, we will discuss the approximating problem of the boundary surface (contained within volume image) by isosurface. It is quite common that a medical volume image can contain multiple interesting structures; we present a novel approach for the selection of multiple isosurfaces to approximate the boundary surfaces of these multiple structures. With this approach, the discrete sampling points of the gray values of the boundary surfaces within volume image are computed first. Then by identifying appropriate clusters from the discrete sampling points and computing the mean of each cluster, we can determine the corresponding isosurfaces for approximating these multiple boundary surfaces.

Date Published: 9 May 2002
PDF: 9 pages
Proc. SPIE 4684, Medical Imaging 2002: Image Processing, (9 May 2002); doi: 10.1117/12.467078

Published in SPIE Proceedings Vol. 4684:
Medical Imaging 2002: Image Processing
Milan Sonka; J. Michael Fitzpatrick, Editor(s)