Abstract

A standard way to segment medical imaging datasets is by tracing contours around regions of interest in parallel planar slices. Unfortunately, the standard methods for reconstructing three dimensional surfaces from those planar contours tend to be either complicated or not very robust. Furthermore, they fail to consistently mesh abutting structures which share portions of contours. In this paper we present a novel, straight-forward algorithm for accurately and automatically reconstructing surfaces from planar contours. Our algorithm is based on scanline rendering and separating surface extraction. By rendering the contours as distinctly colored polygons and reading back each rendered slice into a segmented volume, we reduce the complex problem of building a surface from planar contours to the much simpler problem of extracting separating surfaces from a classified volume. Our scanline surfacing algorithm robustly handles complex surface topologies such as bifurcations, embedded features, and abutting surfaces.

Citations