Layered pentahedral mesh generation for biomechanical geometries with unclosed surfaces

Zhipeng Lei, Jingzhou Yang

Research output: Contribution to journalArticlepeer-review

1 Scopus citations

Abstract

Available methods for automatic volumetric mesh generation require the inputted surface to be watertight. However, biomechanical geometries obtained from a 3D scanner, computed tomography (CT), or magnetic resonance (MRI) may not provide a closed surface. This paper proposes a layered pentahedral mesh generation method for a biomechanical geometry that is defined by any unclosed surface. Based on the initial geometries, two basic meshing problems for biomechanical model were solved. One is that the biomechanical geometry is defined by two surfaces and the volume bounded by the two surfaces is meshed. The other one is that the geometry defined by one surface and a list of points is meshed where there are predefined depths on these points. A distance function was defined for determining relative distance between a surface and a plane. The volumetric mesh was created by adding layers of pentahedral elements, and these pendtahedral elements were smoothed using the signal processing method. The proposed method was demonstrated with examples of meshing the deadspace of headform/respirator and developing the finite element model of a human face with different soft tissue thicknesses. Further, two applications, the CFD simulation of air leakage of headform/respirator and finite element contact simulation between a headform and a respirator, used the proposed method to create volumetric mesh. Results show that the proposed mesh method was able to create biomechanical models with acceptable mesh qualities from highly curved geometries.

Original languageEnglish
Pages (from-to)231-245
Number of pages15
JournalComputer-Aided Design and Applications
Volume10
Issue number2
DOIs
StatePublished - 2013

Keywords

  • Biomechanical geometry
  • Layered pentahedral mesh
  • Respirator

Fingerprint

Dive into the research topics of 'Layered pentahedral mesh generation for biomechanical geometries with unclosed surfaces'. Together they form a unique fingerprint.

Cite this