Spatially Realistic Human Heart Finite Element Models from Medical Imaging

Computer Tomography (CT) and in particular super fast, 64 and 256 detector CT has rapidly advanced over recent years, such that excellent spatial resolution cardiac imaging, with highly reduced motion artifacts, has become a reality. In this paper, we introduce and apply our comprehensive framework to construct spatially realistic three dimensional (3D) finiteelement and boundary element mesh models of the human heart from such high resolution CT medical imaging. The overall IMAGING2MODELING framework consists of several image processing, geometric processing and quality meshing modules. While relying on this framework, the main highlights of this paper with its focus on human heart modeling, are three key geometry processing algorithms, namely, geometric filtering or GEOMETRIC CURATION, anatomy guided decomposition and annotation or GEOMETRIC SEGMENTATION, and weighted distance field blending or GENERALIZED OFFSETS. These three algorithms, due to the very nature of the computation involved, espouse a common paradigm of geometric modeling with weighted distance functions. The spatially realistic human heart modeling results presented in the paper, demonstrate both the robustness (to imaging noise) and necessary applicability of these algorithms to deal with the challenges posed by real-world patient specific CT data collected from our medical collaborators.