Related Experiment Videos
An improved method for finite element mesh generation of geometrically complex structures with application to the
D L Camacho1, R H Hopper, G M Lin
1Department of Biomedical Engineering, Duke University, Durham, NC 27708-0281, USA.
Journal of Biomechanics
|December 10, 1997
Summary
This study presents an automated method for creating finite element models from CT images using hexahedral elements. The technique smooths boundaries, reducing surface stress errors in complex structures like the skullbase.
Area of Science:
- Biomedical Engineering
- Computational Mechanics
- Medical Imaging
Background:
- Finite element analysis (FEA) is crucial for simulating biomechanical behavior.
- Generating accurate FEA models from medical imaging (CT) presents challenges, especially with complex geometries.
- Existing voxel-based methods often result in discontinuous surfaces and stress inaccuracies.
Purpose of the Study:
- To develop an automated method for generating hexahedral-only finite element meshes from CT images.
- To improve the accuracy of FEA models by smoothing irregular boundaries at surfaces and material interfaces.
- To reduce surface stress errors compared to traditional voxel-based approaches.
Main Methods:
- An automated algorithm was developed to reconstruct finite element meshes exclusively using hexahedral elements from CT scan data.
- The method incorporates boundary smoothing to address irregularities at model surfaces and internal material interfaces.
- Validation was performed using a circular ring model and a human skullbase model, comparing smoothed vs. unsmoothed meshes.
Main Results:
- Smoothed models exhibited lower Root Mean Square (RMS) error in surface Von Mises stress compared to unsmoothed models across various mesh densities (e.g., 0.080-0.15 MPa vs. 0.11-0.24 MPa for ring models).
- Element-to-element stress oscillation, a measure of surface stress discontinuity, was significantly reduced in smoothed models for both the ring (4.0-6.8 kPa mm-2 vs. 11.5-15.0 kPa mm-2) and skullbase (1.83 kPa mm-2 vs. 5.52 kPa mm-2) models.
- The automated technique enables rapid reconstruction of complex geometries with improved surface accuracy.
Conclusions:
- The developed automated method effectively generates accurate, hexahedral-only finite element meshes from CT images.
- Boundary smoothing is critical for reducing surface stress errors and oscillations in FEA models of complex anatomical structures.
- This approach offers a significant improvement over voxel-based methods for creating reliable biomechanical models from medical imaging data.