Parallel Adaptive Method for Selecting Points of Interest in Structures: Cranial Deformation

Abstract

In this paper, we present a cranial deformation simulation by applying a stress model using the Finite Element Method (FEM). We determine the geometry of cranium by processing 134 Computed Tomography (CT) images. We implement an image segmentation algorithm to build a three-dimensional cranium structure. We simulate stress and propose the boundary conditions according to the final position of forces applied by a stereotactic frame fixed on the human head. We used probabilistic and adaptive methods to select points in the structure with the purpose of reducing the computational cost and computational error due to discretization problem. We implement the algorithms using parallel programming (PosixThread and MPI).