Research Papers

Design and Calibration of Load Sensing Orthopaedic Implants

[+] Author and Article Information
G. Bergmann, F. Graichen, A. Rohlmann, P. Westerhoff, B. Heinlein

Julius Wolff Institut, Charité—Universitätsmedizin Berlin, Augustenburger Platz 1, 13353 Berlin, Germany

A. Bender

 Berlin-Brandenburg Center for Regenerative Therapies, Berlin, Germany

R. Ehrig

 Konrad-Zuse-Zentrum für Informationstechnik, Berlin, Germany

J Biomech Eng 130(2), 021009 (Mar 28, 2008) (9 pages) doi:10.1115/1.2898831 History: Received September 27, 2006; Revised June 21, 2007; Published March 28, 2008

Contact forces and moments act on orthopaedic implants such as joint replacements. The three forces and three moment components can be measured by six internal strain gauges and wireless telemetric data transmission. The accuracy of instrumented implants is restricted by their small size, varying modes of load transfer, and the accuracy of calibration. Aims of this study were to test with finite element studies design features to improve the accuracy, to develop simple but accurate calibration arrangements, and to select the best mathematical method for calculating the calibration constants. Several instrumented implants, and commercial and test transducers were calibrated using different loading setups and mathematical methods. It was found that the arrangement of flexible elements such as bellows or notches between the areas of load transfer and the central sensor locations is most effective to improve the accuracy. Increasing the rigidity of the implant areas, which are fixed in bones or articulate against joint surfaces, is less effective. Simple but accurate calibration of the six force and moment components can be achieved by applying eccentric forces instead of central forces and pure moments. Three different methods for calculating the measuring constants proved to be equally well suited. Employing these improvements makes it possible to keep the average measuring errors of many instrumented implants below 1–2% of the calibration ranges, including cross talk. Additional errors caused by noise of the transmitted signals can be reduced by filtering if this is permitted by the sampling rate and the required frequency content of the loads.

Copyright © 2008 by American Society of Mechanical Engineers
Your Session has timed out. Please sign back in to continue.



Grahic Jump Location
Figure 7

Accuracy test of shoulder implant. Test load −Fz applied after calibration at Point P0 (Fig. 4), Calculation Method A. The external force Fz decreases to −1750N and goes back to zero (left scale). The maximum error of Fz (right scale) stays below 32N (1.8% of calibration range) while the error of the other two force components (cross talk) is less than 10N. Higher errors during the first half of the loading cycle. Error peaks can be reduced by software filtering.

Grahic Jump Location
Figure 6

Strain error along inner circumference of the tube. Percent difference (Eq. 8) between vertical strain εconst caused by the central force 2F (Fig. 3, top middle) and the strains ε(α), caused by the twin-force F+F (Fig. 3, top right and lower row). Four different implant designs (Fig. 3). Data from the middle of the inner tube (purple arrows in Fig. 3) around half of the tube circumference (angle α in Fig. 3).

Grahic Jump Location
Figure 5

Calibration Arrangement CA9. Calibration Forces A are applied successively at the loading points P0–P8. Force directions: P0=direction−z; P1, P2, P5, P6=angleα against plane yz; P3, P4, P7, P8=angleα against plane xz.

Grahic Jump Location
Figure 4

Calibration Arrangement CA21. The implant is clamped at the bottom and a loading plate is mounted on top. Calibration Forces A are applied successively at the loading points P0–P20.

Grahic Jump Location
Figure 3

Influence of design on measured strains. The strains acting in vertical direction are indicated in color. Top left: 3D model of Design A with section plane and loading areas (purple). Other pictures: view at section plane and inner side of tube for Designs A (basic design, dimensions of Test Transducer 1), Design B (thicker loading plate), Design C (outer notch), and Design D (inner notch). A central force 2F causes constant strains εconst (green areas) in the middle of the tube (top middle). The twin forces F+F at the edges of the loading plate lead to nonconstant strain distributions (top right and lower row). Each color change corresponds to a strain change of Δε≈10% of εconst. Purple arrows indicate the height at which the strains in Fig. 6 were evaluated.

Grahic Jump Location
Figure 2

Six-component load transducer. Six load components L1–L6 act relative to the systems x, y, and z: three force components Fx, Fy, and Fz and three moment components Mx, My, and Mz. The coordinate system can be defined at any location.

Grahic Jump Location
Figure 1

Schematics of instrumented implants (a) Shoulder joint, (b) Intratibial load in sheep, (c) prototype of vertebral body, and (d) prototype of knee joint. 1=tube with six SGs, 2=interface implant bone or implant cement, 3=joint contact area, 4=bone cement, 5=humerus, 6=scapula, 7=sheep tibia, 8=vertebral body, and 9=tibia. Load application areas are either joint contact areas or interfaces between implant and bone or cement. Six force and moment components are calculated from the signals of six SGs in an intermediate tube (see Fig. 2).




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