0
Journal of Biomechanical Engineering | Volume 132 | Issue 11 | Technical Brief
Technical Briefs

Application of Neural Networks and Finite Element Computation for Multiscale Simulation of Bone Remodeling

[+] Author and Article Information
Ridha Hambli

 Institut Prisme/MMH, 8 Rue Léonard de Vinci, 45072 Orléans Cedex 2, Franceridha.hambli@univ-orleans.fr

J Biomech Eng 132(11), 114502 (Oct 12, 2010) (5 pages) doi:10.1115/1.4002536 History: Received January 10, 2010; Revised September 08, 2010; Posted September 10, 2010; Published October 12, 2010; Online October 12, 2010
Article
References
Figures
Tables
Citing Articles

In this paper, a novel multiscale hierarchical model based on finite element analysis and neural network computation was developed to link mesoscopic and macroscopic scales to simulate the bone remodeling process. The finite element calculation is performed at the macroscopic level, and trained neural networks are employed as numerical devices for substituting the finite element computation needed for the mesoscale prediction. Based on a set of mesoscale simulations of representative volume elements of bones taken from different bone sites, a neural network is trained to approximate the responses at the meso level and transferred at the macro level.
FIGURES IN THIS ARTICLE
<>
Copyright © 2010 by American Society of Mechanical Engineers
Your Session has timed out. Please sign back in to continue.

References

Figures

Grahic Jump Location
+Multiscale hierarchical approach for bone analysis. Two level analysis for the prediction of the remodeling process. Macroscale level: a whole bone computed using FE analysis. Mesoscale level: μ-CT FE model computed using trained NN. 2 mm3 RVE obtained using digital image-based modeling technique meshed using about 200,000 eight-node 3D brick elements.
View Large  |  Save Figure  |  Download Slide (.ppt)
Multiscale hierarchical approach for bone analysis. Two level analysis for the prediction of the remodeling process. Macroscale level: a whole bone computed using FE analysis. Mesoscale level: μ-CT FE model computed using trained NN. 2 mm3 RVE obtained using digital image-based modeling technique meshed using about 200,000 eight-node 3D brick elements.
Figure 1

Multiscale hierarchical approach for bone analysis. Two level analysis for the prediction of the remodeling process. Macroscale level: a whole bone computed using FE analysis. Mesoscale level: μ-CT FE model computed using trained NN. 2 mm3 RVE obtained using digital image-based modeling technique meshed using about 200,000 eight-node 3D brick elements.

Grahic Jump Location
+Meso-to-macro transition: NN is incorporated into the FE code ABAQUS via the routine UMAT
View Large  |  Save Figure  |  Download Slide (.ppt)
Meso-to-macro transition: NN is incorporated into the FE code ABAQUS via the routine UMAT
Figure 2

Meso-to-macro transition: NN is incorporated into the FE code ABAQUS via the routine UMAT

Grahic Jump Location
+Contour of trabecular bone damage predicted by the hybrid method. (a) Macroscopic contour predicted by NN computation. (b) RVE predicted results are averaged and passed to the macroscopic finite element level.
View Large  |  Save Figure  |  Download Slide (.ppt)
Contour of trabecular bone damage predicted by the hybrid method. (a) Macroscopic contour predicted by NN computation. (b) RVE predicted results are averaged and passed to the macroscopic finite element level.
Figure 3

Contour of trabecular bone damage predicted by the hybrid method. (a) Macroscopic contour predicted by NN computation. (b) RVE predicted results are averaged and passed to the macroscopic finite element level.

Grahic Jump Location
+Contour of trabecular bone density predicted by the hybrid method. (a) Macroscopic contour predicted by NN computation. (b) RVE predicted results are averaged and passed to the macroscopic finite element level.
View Large  |  Save Figure  |  Download Slide (.ppt)
Contour of trabecular bone density predicted by the hybrid method. (a) Macroscopic contour predicted by NN computation. (b) RVE predicted results are averaged and passed to the macroscopic finite element level.
Figure 4

Contour of trabecular bone density predicted by the hybrid method. (a) Macroscopic contour predicted by NN computation. (b) RVE predicted results are averaged and passed to the macroscopic finite element level.

Tables

Errata

Discussions

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Classification Model to Detect Failed HDD Components
Intelligent Engineering Systems through Artificial Neural Networks Volume 18
Computation of Gradient and Hessian in Feed-Forward Neural Networks: A Variational Approach
Intelligent Engineering Systems Through Artificial Neural Networks, Volume 17
Hybrid Cryptographic Scheme for Data Communication
International Conference on Advanced Computer Theory and Engineering (ICACTE 2009)
Modeling and Simulation of Coal Gas Concentration Prediction Based on the BP Neural Network
International Symposium on Information Engineering and Electronic Commerce, 3rd (IEEC 2011)
Parallel Computing Using Multilevel Hypergraph Partitioning for Three-Dimensional Scramjet
International Conference on Mechanical and Electrical Technology, 3rd, (ICMET-China 2011), Volumes 1–3
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In