Finite element models of the human tongue: a mixed-element mesh approach. Issue 6 (2nd November 2017)

• Record Type: Journal Article
• Authors: Rohan, P.-Y.; Lobos, C.; Nazari, M. A.; Perrier, P.; Payan, Y.
• Is Part Of: Computer methods in biomechanics and biomedical engineering. Volume 5:Issue 6(2017)
• Page Start: 390
• Page End: 400
• Publication Date: 2017-11-02
• Subjects: Finite element analysis -- biological soft tissue -- mixed-element mesh -- volumetric locking; Imaging systems in biology -- Periodicals; Imaging systems in medicine -- Periodicals; Biomechanics -- Data processing -- Periodicals; Biomedical engineering -- Periodicals; 616.0757
• DOI: 10.1080/21681163.2015.1105760
• Languages: English
• ISSNs: 2168-1163

Abstract

Abstract : One of the key factors for obtaining accurate and reliable results using the finite element method is the discretisation of the domain. Traditionally, two main types of elements are used for three-dimensional mesh generation: tetrahedral and hexahedral elements. Tetrahedral meshes are automatically generated but standard displacement-based tetrahedral elements generally suffer from performance issues in terms of convergence rate and accuracy of the solution associated with volumetric and shear locking. Because of these distinct disadvantages, hexahedral meshes have been used up until now for the design of biomechanical models of the orofacial system in particular for medical applications. However, hexahedral meshing is very costly and labour-intensive when the mesh must be hand-made. The aim of the present contribution is to evaluate the performance of mixed element meshes as an alternative to all-tetrahedral or all-hexahedral meshing for the analysis of problems involving nearly incompressible materials at large strains. The case study of a semi-confined compression experiment of an elastic cylindrical specimen was analysed. The theoretical expression of deformation was derived from the literature. We observed that linear mixed element meshes allowed results very close to those obtained using hexahedral ones. As a second experiment, we generated a mixed element mesh of the tongue and analyse its simulated response to activation of the posterior Genioglossus muscle. Overall, our results show that mixed element meshes can be used as computationally less demanding alternative to all-hexahedral meshes for the analysis of problems involving nearly-incompressible materials at large strains.