Induction of Hypotheses Concerning Hip Arthroplasty: A Modified Methodology for Medical Research

Summary

Objectives: The objective of this study is to advocate a methodology for medical research that, in contrast to traditional medical methodology, exploits the flexibility of machine learning and retains the kind of statistical tests that are generally accepted in the medical field for the confirmation of hypotheses.

Methods: First, the medical problem is defined and data for an observed population are collected; then a machine learning tool is used to generate hypotheses regarding the problem; finally, statistical methods are used to determine the validity of the generated hypotheses.

Results: To illustrate this approach, the problem of defining indications for hip arthroplasty after an acute medial femoral neck fracture is investigated as a case study.

Conclusions: The methodology is similar to the usual style of applying machine learning, but insists on a link to the techniques of statistical tests that are normally used in medicine. It aims at a more flexible and economical use of experimental data than in the usual medical research, which is enabled by techniques of machine learning. At the same time, by reference to traditional statistical tests, it is hoped that this approach will lead to improved acceptance of machine learning in the medical field.

Keywords

Research Methodology - Medical Data Analysis - Machine Learning - Regression Trees - Hip Arthroplasty

• References

• 1 Hand DJ. Data mining: statistics and more. The American Statistician 1988; 52: 112-8.
• 2 Salzberg SL. On comparing classifiers: pitfalls to avoid and a recommended approach. Data Mining and Knowledge Discovery 1997; 1: 317-28.
• 3 Shrager J. et al. Computational models of scientific discovery and theory formation. San Mateo, California: Morgan Kaufmann; 1990: 1-24.
• 4 Srinivasan A. et al. Mutagenesis: ILP experiments in a non-determinate biological domain. In: Proceedings Fourth International Workshop ILP-94. Bonn: Bad Honnef; 1994: 1-22.
• 5 James SE. et al. Bi-articular hemiarthroplasty of the hip: a 7-year follow-up. Injury 1991; 22 (Suppl. 05) 391.
• 6 Kitamura S. et al. Functional outcome after hip fracture in Japan. Clin Orthop 1998; 348: 29.
• 7 Koval KJ. et al. Predictors of functional recovery after hip fracture in the elderly. Clin Orthop 1998; 348: 22.
• 8 Malhotra R. et al. Bipolar hemiarthroplasty in femoral neck fractures. Arch Orthop Trauma Surg 1995; 114: 79.
• 9 Evarts CM. Endoprosthesis as the primary treatment of femoral neck fractures. Clin Orthop 1973; 92: 69-76.
• 10 Goldhill VB. et al. Bipolar hemiarthroplasty for fracture of the femoral neck. J Orthop Trauma 1991; 5 (Suppl. 03) 318.
• 11 Lestrange NR. Bipolar arthroplasty for 496 hip fractures. Clin Orthop 1990; 251: 7.
• 12 Hudson JI. et al. Eight-year outcome associated with clinical options in the management of fe-moral neck fractures. Clin Orthop 1998; 348: 59.
• 13 Bateman JE. et al. Long-term results of bipolar arthroplasty in osteoarthritis of the hip. Clin Orthop 1990; 251: 54.
• 14 Torisu T, Utsunomiya K, Maekawa M, Ueda J. Use of bipolar hip arthroplasty in states of ace-tabular deficiency. Clin Orthop 1990; 251: 119.
• 15 Vazquez-Vela E, Vazquez-Vela G. Acetabular reaction to the Bateman bipolar prosthesis. Clin Orthop 1990; 251: 88.
• 16 Verberne GHM. A femoral head prosthesis with a built-in joint. A radiological study of the movements of the two components. J Bone Joint Surg 1983; 65B: 544.
• 17 Kenzora JE. et al. Outcome after hemiarthroplasty for femoral neck fractures in the elderly. Clin Orthop 1998; 348: 51.
• 18 Notage WM, McMaster WC. Comparison of bipolar implants with fixed-neck prosthesis in fe-moral neck fractures. Clin Orthop 1990; 251: 38.
• 19 Cornell CN. et al. Unipolar versus bipolar hemiarthroplasty for the treatment of femoral neck fractures in the elderly. Clin Orthop 1998; 348: 67.
• 20 Paton RW, Hirst P. Hemiarthroplasty of the hip and dislocation. Injury 1989; 20: 167-9.
• 21 Eiskjaer S. et al. Suvivorship analysis of hemiarthroplasties. Clin Orthop 1993; 286: 206-11.
• 22 Gebhard JS. et al. A comparison of total hip arthroplasty and hemiarthroplasty for treatment of acute fracture of the femoral neck. Clin Orthop 1992; 282: 123-31.
• 23 Franklin A, Gallannaugh SC. The bi-articular hip prosthesis for fractures of the femoral neck – a preliminary report. Injury 1983; 15: 159-62.
• 24 Gallinaro P. et al. Experience with bipolar pros-thesis in femoral neck fractures in the elderly and debilitated. Clin Orthop 1990; 251: 26.
• 25 Garrahan WF, Madden EJ. The long-stem bipolar prosthesis in surgery of the hip. Clin Orthop 1990; 251: 31.
• 26 Harris WH. Traumatic arthritis of the hip after dislocation and acetabular fractures:Treatment by mold arthroplasty. An end-result study using a new method of result evaluation. J Bone Joint Surg 1969; 51A: 737-55.
• 27 Karalič A. Employing linear regression in regression tree leaves. In: Neumann B. ed. Proceedings of Tenth European Conference on Artificial Intelligence. Vienna: Wiley; 1992: 410-1.
• 28 Quinlan R. Top down induction of decision trees. Mach Learn 1986; 1: 81-106.
• 29 Cestnik B, Bratko I. On estimating probabilities in tree pruning. In: Brazdil P. ed. Proceedings ECML91. Porto: Springer; 1991: 138-50.
• 30 Smrke D. et al. An analysis of implantations of hip endoprostheses with regression trees. Inform Med Slov [Print Ed] 1998 5. (1-2) 111-6.
• 31 Muggleton S. Inductive Logic Programming. Kluwer: 1992