Machine learning techniques for the optimization of joint replacements: Application to a short-stem hip implant

Today, different implant designs exist in the market; however, there is not a clear understanding of which are the best implant design parameters to achieve mechanical optimal conditions. Therefore, the aim of this project was to investigate if the geometry of a commercial short stem hip prosthesis can be further optimized to reduce stress shielding effects and achieve better short-stemmed implant performance. To reach this aim, the potential of machine learning techniques combined with parametric Finite Element analysis was used. The selected implant geometrical parameters were: total stem length (L), thickness in the lateral (R1) and medial (R2) and the distance between the implant neck and the central stem surface (D). The results show that the total stem length was not the only parameter playing a role in stress shielding. An optimized implant should aim for a decreased stem length and a reduced length of the surface in contact with the bone. The two radiuses that characterize the stem width at the distal cross-section in contact with the bone were less influential in the reduction of stress shielding compared with the other two parameters; but they also play a role where thinner stems present better results.

[1]  Thomas Roß,et al.  Feature selection for optimized skin tumor recognition using genetic algorithms , 1999, Artif. Intell. Medicine.

[2]  Omid Omidvar,et al.  Neural Networks and Pattern Recognition , 1997 .

[3]  R. Crowninshield,et al.  An analysis of femoral component stem design in total hip arthroplasty. , 1980, The Journal of bone and joint surgery. American volume.

[4]  J. Wolff.  Das Gesetz der Transformation der Knochen , 1893 .

[5]  João Folgado,et al.  Three-dimensional shape optimization of hip prostheses using a multicriteria formulation , 2007 .

[6]  Y. Yoon,et al.  Shape optimal design of the stem of a cemented hip prosthesis to minimize stress concentration in the cement layer. , 1989, Journal of biomechanics.

[7]  Vojislav Kecman,et al.  Gene extraction for cancer diagnosis by support vector machines - An improvement , 2005, Artif. Intell. Medicine.

[8]  W. Taylor,et al.  The difference between stretching and splitting muscle trauma during THA seems not to play a dominant role in influencing periprosthetic BMD changes. , 2012, Clinical Biomechanics.

[9]  José Manuel García-Aznar,et al.  Computational evaluation of different numerical tools for the prediction of proximal femur loads from bone morphology , 2014 .

[10]  Rui B Ruben,et al.  On the optimal shape of hip implants. , 2012, Journal of biomechanics.

[11]  Alexander J. Smola,et al.  Learning with Kernels: support vector machines, regularization, optimization, and beyond , 2001, Adaptive computation and machine learning series.

[12]  G. Bergmann,et al.  Patellofemoral joint contact forces during activities with high knee flexion , 2012, Journal of orthopaedic research : official publication of the Orthopaedic Research Society.

[13]  D T Davy,et al.  Effects of loading conditions and objective function on three-dimensional shape optimization of femoral components of hip endoprostheses. , 2000, Medical engineering & physics.

[14]  L. K. Hansen,et al.  Melanoma diagnosis by Raman spectroscopy and neural networks: structure alterations in proteins and lipids in intact cancer tissue. , 2004, The Journal of investigative dermatology.

[15]  Igor Kononenko,et al.  Machine learning for medical diagnosis: history, state of the art and perspective , 2001, Artif. Intell. Medicine.

[16]  M. Lerch,et al.  The Nanos Short Stem in Total Hip Arthroplasty: A Mid Term Follow-Up , 2011, Hip international : the journal of clinical and experimental research on hip pathology and therapy.

[17]  Vojislav Kecman,et al.  Gene Extraction for Cancer Diagnosis by Support Vector Machines , 2005, ICANN.

[18]  M. Viceconti,et al.  Mathematical relationships between bone density and mechanical properties: a literature review. , 2008, Clinical biomechanics.

[19]  Anthony Widjaja,et al.  Learning with Kernels: Support Vector Machines, Regularization, Optimization, and Beyond , 2003, IEEE Transactions on Neural Networks.

[20]  D R Sumner,et al.  Sensitivity of periprosthetic stress-shielding to load and the bone density-modulus relationship in subject-specific finite element models. , 2000, Journal of biomechanics.

[21]  T. Keaveny,et al.  Trabecular bone modulus-density relationships depend on anatomic site. , 2003, Journal of biomechanics.

[22]  J. F. Fitzgerald,et al.  The effect of hip fracture on mortality, hospitalization, and functional status: a prospective study. , 1997, American journal of public health.

[23]  Hans-Christian Hege,et al.  amira: A Highly Interactive System for Visual Data Analysis , 2005, The Visualization Handbook.

[24]  H Tanino,et al.  Three-dimensional computer-aided design based design sensitivity analysis and shape optimization of the stem using adaptive p-method. , 2006, Journal of biomechanics.

[25]  R Huiskes,et al.  Mathematical shape optimization of hip prosthesis design. , 1989, Journal of biomechanics.

[26]  P R Fernandes,et al.  Shape Optimization of a Cementless Hip Stem for a Minimum of Interface Stress and Displacement , 2004, Computer methods in biomechanics and biomedical engineering.

[27]  M. Gillies,et al.  Primary and long-term stability of a short-stem hip implant , 2010, Proceedings of the Institution of Mechanical Engineers. Part H, Journal of engineering in medicine.

[28]  Jean-Yves Audibert.  Optimization for Machine Learning , 1995 .

[29]  Celestino Ordóñez,et al.  Creating a quality map of a slate deposit using support vector machines , 2007 .

[30]  Fu Chang,et al.  Language identification of character images using machine learning techniques , 2005, Eighth International Conference on Document Analysis and Recognition (ICDAR'05).

[31]  Riccardo Poli,et al.  A neural network expert system for diagnosing and treating hypertension , 1991, Computer.

[32]  Antonius Rohlmann,et al.  Total hip joint prosthesis for in vivo measurement of forces and moments. , 2010, Medical engineering & physics.

[33]  T. Goswami,et al.  Effect of geometric parameters in the design of hip implants paper IV , 2004 .

[34]  Leonardo Franco,et al.  Generalization and Selection of Examples in Feedforward Neural Networks , 2000, Neural Computation.

[35]  H. Amstutz,et al.  "Modes of failure" of cemented stem-type femoral components: a radiographic analysis of loosening. , 1979, Clinical orthopaedics and related research.

[36]  William R Taylor,et al.  Influence of changes in stem positioning on femoral loading after THR using a short-stemmed hip implant. , 2007, Clinical biomechanics.

[37]  Bernard Widrow,et al.  Neural networks: applications in industry, business and science , 1994, CACM.

[38]  S. Stulberg,et al.  The short stem: promises and pitfalls. , 2013, The bone & joint journal.

[39]  Javier Martinez,et al.  Machine Learning Techniques as a Helpful Tool Toward Determination of Plaque Vulnerability , 2012, IEEE Transactions on Biomedical Engineering.

[40]  S. McMahon,et al.  Total Hip Arthroplasty Using a Short-Stem Prosthesis: Restoration of Hip Anatomy , 2015, Journal of orthopaedic surgery.

[41]  Christopher Boyle,et al.  Comparison of different hip prosthesis shapes considering micro-level bone remodeling and stress-shielding criteria using three-dimensional design space topology optimization. , 2011, Journal of biomechanics.

[42]  G. Bergmann,et al.  Musculo-skeletal loading conditions at the hip during walking and stair climbing. , 2001, Journal of biomechanics.

[43]  José M. Matías,et al.  Functional classification of ornamental stone using machine learning techniques , 2010, J. Comput. Appl. Math..

[44]  K. Jabbour,et al.  ALFA: automated load forecasting assistant , 1988 .

[45]  W. Taylor,et al.  Generic Rules of Mechano-Regulation Combined with Subject Specific Loading Conditions Can Explain Bone Adaptation after THA , 2012, PloS one.

[46]  Anil K. Jain,et al.  Neural networks and pattern recognition , 1994 .

[47]  Usama M. Fayyad,et al.  SKICAT: A Machine Learning System for Automated Cataloging of Large Scale Sky Surveys , 1993, ICML.

[48]  G. Duda,et al.  Strain shielding inspired re‐design of proximal femoral stems for total hip arthroplasty , 2017, Journal of orthopaedic research : official publication of the Orthopaedic Research Society.

[49]  P Kowalczyk,et al.  Design optimization of cementless femoral hip prostheses using finite element analysis. , 2001, Journal of biomechanical engineering.

[50]  M. Warner,et al.  Bone Remodelling of a Proximal Femur with the Thrust Plate Prosthesis: An In Vitro Case , 2004, Computer methods in biomechanics and biomedical engineering.

[51]  T J Santner,et al.  Design and analysis of robust total joint replacements: finite element model experiments with environmental variables. , 2001, Journal of biomechanical engineering.