Computing shortest cycles on discrete surfaces for acurate topological modifications of medical image isosurfaces

Topological control over discrete isosurface is of primordial interest in medical applications, especially discrete model building for active contours. Previous attempts showed that the key point in acurately modifying topology was computation of shortest cycles on the surface of interest. This paper generalizes the shortest path algorithm to compute shortest cycles in a given homotopy class on a discrete surface with arbitrary topology. The algorithm is simple to implement and general to all kinds of discrete surfaces. The algorithm is validated against synthetic surfaces.