0
Technical Brief

A Validated Open-Source Multisolver Fourth-Generation Composite Femur Model

[+] Author and Article Information
Alisdair R. MacLeod

Centre for Biomechanics,
Department of Mechanical Engineering,
University of Bath,
Bath BA2 7AY, UK
e-mail: a.macleod@bath.ac.uk

Hannah Rose

Centre for Biomechanics,
Department of Mechanical Engineering,
University of Bath,
Bath BA2 7AY, UK
e-mail: hannahjrose@blueyonder.co.uk

Harinderjit S. Gill

Centre for Biomechanics,
Department of Mechanical Engineering,
University of Bath,
Bath BA2 7AY, UK
e-mail: r.gill@bath.ac.uk

1Corresponding author.

Manuscript received December 27, 2015; final manuscript received September 2, 2016; published online November 3, 2016. Assoc. Editor: Joel D. Stitzel.

J Biomech Eng 138(12), 124501 (Nov 03, 2016) (9 pages) Paper No: BIO-15-1668; doi: 10.1115/1.4034653 History: Received December 27, 2015; Revised September 02, 2016

Synthetic biomechanical test specimens are frequently used for preclinical evaluation of implant performance, often in combination with numerical modeling, such as finite-element (FE) analysis. Commercial and freely available FE packages are widely used with three FE packages in particular gaining popularity: abaqus (Dassault Systèmes, Johnston, RI), ansys (ANSYS, Inc., Canonsburg, PA), and febio (University of Utah, Salt Lake City, UT). To the best of our knowledge, no study has yet made a comparison of these three commonly used solvers. Additionally, despite the femur being the most extensively studied bone in the body, no freely available validated model exists. The primary aim of the study was primarily to conduct a comparison of mesh convergence and strain prediction between the three solvers (abaqus, ansys, and febio) and to provide validated open-source models of a fourth-generation composite femur for use with all the three FE packages. Second, we evaluated the geometric variability around the femoral neck region of the composite femurs. Experimental testing was conducted using fourth-generation Sawbones® composite femurs instrumented with strain gauges at four locations. A generic FE model and four specimen-specific FE models were created from CT scans. The study found that the three solvers produced excellent agreement, with strain predictions being within an average of 3.0% for all the solvers (r2 > 0.99) and 1.4% for the two commercial codes. The average of the root mean squared error against the experimental results was 134.5% (r2 = 0.29) for the generic model and 13.8% (r2 = 0.96) for the specimen-specific models. It was found that composite femurs had variations in cortical thickness around the neck of the femur of up to 48.4%. For the first time, an experimentally validated, finite-element model of the femur is presented for use in three solvers. This model is freely available online along with all the supporting validation data.

FIGURES IN THIS ARTICLE
<>
Copyright © 2016 by ASME
Your Session has timed out. Please sign back in to continue.

References

Figures

Grahic Jump Location
Fig. 1

(a) Anterior view of the experimental setup showing strain gauge locations around the femur. (b) Medial view of the FE model showing the loading and boundary conditions.

Grahic Jump Location
Fig. 2

(a) Definition of the sectional plane, (b) the exported cross section with nodes, and (c) the evaluation of the cortical thickness from the cross section

Grahic Jump Location
Fig. 3

(a) Experimental results versus generic finite-element model predictions of equivalent strain. (b) Experimental results versus specimen-specific finite-element predictions (specimens F9–F12) of equivalent strain. Values are shown for all the strain gauge locations at 50 N increments up to a maximum load of 500 N. N.B. least squares fit for y = mx + c, with values of the slope, m, given for each plot.

Grahic Jump Location
Fig. 4

Mesh convergence for the four strain gauge locations showing ±5% bounds of the equivalent strain predictions

Grahic Jump Location
Fig. 5

Linear regression of the experimentally measured strains versus FE predictions for specimen F10 and Bland–Altman plots for the three solvers: (a) abaqus, (b) ansys, and (c) FEBio

Grahic Jump Location
Fig. 6

Cross sections of four composite femur specimens (F9–F12), obtained as shown in Fig. 2(b), showing the eight locations on the inner and outer surfaces used to determine cortical thickness. Plot of cortical thickness at the eight locations around the femoral neck.

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
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