Locating implants with respect to the bone in diagnostic X-ray images of the pelvis

This thesis is about the assessment of the migration of artificial hip sockets in clinical x-ray images. “Migration”means the displacement of the bony implant with respect to the bone in time. It is seen as the most objective indicator for an implant loosening, which is the main problem in hip replacement. The data source are the standard anterior-posterior pelvic radiographs which are usually acquired for the post-operative examination of the replacement. According to the general case, it is assumed that no computer tomograms of the pelvis are available. Due to the lack of 3D information of the bone, only the components of the migration parallel to the plane of the radiographic film can be observed. It is assumed that a migration of only 0.5 to 1.0 millimeter in the first two postoperative years implies an increased risk of later loosening. This correlation is especially important if a new type of cup or implantation technique is introduced. It means that a precise assessment of migration allows for an early judgement of the fixation properties and for the prevention of the use of implants with a high tendency to loose in many patients. Hence, a useful method for measuring migration should allow for the observation of implant displacements in the sub-millimeter range. The precision of the previously proposed methods for cup migration assessment was analysed theoretically considering the errors entering the measurement. Standard deviations in the range of 0.6mm to 3.0mm resulted. Based on the theoretical analysis, a computerized system for a more precise assessment of the migration was designed. In this system, which is named XIMIT, the sensitivity of the measurement towards the variable position and orientation of the pelvis at exposure is minimized. The most relevant of the remaining sources of error is the uncertainty in determining the position of the necessary bony landmarks and of the implant in the image. It is reduced by the use of image analysis tools. The precise localization of the bony landmarks is achieved by applying a state-of-the-art template matching algorithm. The user roughly determines the position of a landmark in all images of a pelvis and chooses the template, i.e. a region of interest containing the respective bony feature in the first post-operative radiograph. The algorithm finds the exact landmark coordinates in the later radiographs by geometrically fitting the bony structures in these images to the template. The implant is located based on the principle of Analysis-by-Synthesis. Using the CAD model of the artificial hip socket, a synthetic radiograph of the cup depending on its approximate 3D position and orientation in the x-ray system is generated. The exact 3D parameters are recovered by matching the synthetic image to the original one while virtually translating and rotating the cup.

[1]  Reinhard Koch,et al.  Automatic modelling of Natural Scenes for Generating Synthetic Movies , 1990, Eurographics.

[2]  M J Hubbard,et al.  The measurement of progression in protrusio acetabuli. , 1969, The American journal of roentgenology, radium therapy, and nuclear medicine.

[3]  D R Pedersen,et al.  How definition of "loosening" affects the incidence of loose total hip reconstructions. , 1986, Clinical orthopaedics and related research.

[4]  C. Engh,et al.  Evaluation of cementless acetabular component migration. An experimental study. , 1989, The Journal of arthroplasty.

[5]  R Poss,et al.  Clinical and radiographic evaluation of total hip replacement. A standard system of terminology for reporting results. , 1990, The Journal of bone and joint surgery. American volume.

[6]  D P Fyhrie,et al.  The acetabular teardrop and its relevance to acetabular migration. , 1988, Clinical orthopaedics and related research.

[7]  J. Charnley,et al.  Radiological demarcation of cemented sockets in total hip replacement. , 1976, Clinical orthopaedics and related research.

[8]  Reinhard Koch,et al.  Dynamic 3-D Scene Analysis Through Synthesis Feedback Control , 1993, IEEE Trans. Pattern Anal. Mach. Intell..

[9]  Takeo Kanade,et al.  Post-Operative Measurement of Acetabular Cup Position Using X-Ray/CT Registration , 2000, MICCAI.

[10]  R Bauer,et al.  Early migration predicts late aseptic failure of hip sockets. , 1996, The Journal of bone and joint surgery. British volume.

[11]  C J Taylor,et al.  An automated method for assessing routine radiographs of patients with total hip replacements , 1997, Proceedings of the Institution of Mechanical Engineers. Part H, Journal of engineering in medicine.

[12]  G Selvik,et al.  Mechanical loosening of total hip prostheses. A radiographic and roentgen stereophotogrammetric study. , 1986, The Journal of bone and joint surgery. British volume.

[13]  Martin Berger,et al.  Deformable area based template matching with application to low contrast imagery , 1999 .

[14]  G Selvik,et al.  Definition of endoprosthetic loosening. Comparison of arthrography, scintigraphy and roentgen stereophotogrammetry in prosthetic hips. , 1985, Acta orthopaedica Scandinavica.

[15]  B. Manaster,et al.  From the RSNA refresher courses. Total hip arthroplasty: radiographic evaluation. , 1996, Radiographics : a review publication of the Radiological Society of North America, Inc.

[16]  D W Hukins,et al.  Prosthetic hip failure: retrospective radiograph image analysis of the acetabular cup. , 1989, Journal of biomedical engineering.

[17]  G. Selvik,et al.  A Roentgen Stereophotogrammetric System , 1983, Acta radiologica: diagnosis.

[18]  J Kärrholm,et al.  Radiostereometry of hip prostheses. Review of methodology and clinical results. , 1997, Clinical orthopaedics and related research.

[19]  A A Amis,et al.  Measurement of acetabular erosion. The effect of pelvic rotation on common landmarks. , 1989, The Journal of bone and joint surgery. British volume.

[20]  M A Freeman,et al.  Accuracy of the Nunn method in measuring acetabular cup migration. , 1992, Upsala journal of medical sciences.

[21]  Gaudenz Danuser,et al.  Observing deformations of 20 nm with a low-numerical-aperture light microscope , 1996, Other Conferences.

[22]  F Baruffaldi,et al.  A computerized system for radiographical evaluation in total hip arthroplasty. , 1995, Computer methods and programs in biomedicine.

[23]  Karl-Rudolf Koch,et al.  Parameter estimation and hypothesis testing in linear models , 1988 .

[24]  A. Gruen ADAPTIVE LEAST SQUARES CORRELATION: A POWERFUL IMAGE MATCHING TECHNIQUE , 1985 .

[25]  Alan Watt,et al.  Advanced animation and rendering techniques - theory and practice , 1992 .

[26]  David W. Fanning,et al.  Idl Programming Techniques , 2000 .