|
Evaluation of Accuracy and Precision of two Radiographic Regimes for use with the Martell Computer Assisted Method of Measuring Acetabular Wear
Bragdon, C. R., Martell, J., Estok, D., Veldhoven, J, Malchau, H., Harris, W. H.
Introduction
We reported last year at this meeting on the development of a phantom hip wear model which we used to determine the accuracy and precision of RSA used to measure polyethylene wear from clinical radiograph films. The purpose of this study is to use our phantom to evaluate the accuracy and precision of wear measurements using the current form of the Martell semi-automated technique and a proposed new radiographic regimen, (Martell/Harris), having the potential of improving the quality of the clinical radiographic images.
Materials and Methods
The physical model and the techniques of the phantom used have been previously described (1). In addition to the A/P and cross-table lateral (standard Martell) projections, oblique A/P projections of the hip, (phantom), were obtained which were exactly 90° apart.
Results
The total average error of the 2-D analysis using the standard Martell analysis measured -48.6±131.56 microns. The total average error of the 3-D analysis measured 30.5±99.0 microns. This results in a calculated accuracy and precision at a 95% confidence interval of the three dimensional analysis of ±47 microns and ±37 microns respectively. The calculated accuracy and precision at a 95% confidence interval of the two dimensional analysis was ±71 microns and ±59 microns respectively. Preliminary results of the analysis of the Martell/ Harris method showed virtually identical results as that obtained using the phantom in the standard Martell analysis.
Conclusion
The results using the Martell/Harris technique are equal to that obtained with the standard Martell technique. By using two oblique A/P projections, the major disadvantages of using the cross-table laterals may be avoided.
Reference
1) Bragdon et al Journal of Orthopaedic Research, Accepted for publication.
|
