Generation of Patient-Specific Finite-Element Mesh from 3D Medical Images

Abstract

Finite element (FE) mesh generation remains an important issue for patient specific biomechanical modeling in clinical applications such as computer assisted planning and computer aided surgery. While some techniques make automatic mesh generation possible, in most existing procedures, patient medical image segmentation is required, which is tedious and time consuming. We present a novel patient-specific FE mesh morphing scheme which automatically generates patient-specific FE meshes from patient medical images without resorting to image segmentation. Within our scheme, a robust anisotropic filtering strategy with automatic parameter value selection is also proposed to effectively de-noise and enhance the magnetic resonance image (MRI). This de-noising algorithm can be apply to multi-modal images, which gives clinicians more freedom to choose the appropriate image modality based on each patient?s situation. The proposed FE mesh generation method is applied to three different clinical cases totaling 36 patient datasets. Results show that our method outperforms state-of-the-art methods in both generated mesh accuracy and mesh quality.