Surface area estimation in practice

Abstract

Consider a complex, convoluted three dimensional object that has been digitized and is available as a set of voxels. We describe a fast, practical scheme for delineating a region of interest on the surface of the object and estimating its original area. The voxel representation is maintained and no triangulation is carried out. The methods presented rely on a theoretical result of Mullikin and Verbeek, and bridge the gap between their idealized setting and the harsh reality of 3D medical data. Performance evaluation results are provided, and operation on segmented white matter MR brain data is demonstrated. © Springer-Verlag Berlin Heidelberg 2003.