LOCOS Mission

The mission of the laboratory is to improve the value of musculoskeletal healthcare through collaborative research between quantitative researchers and physicians. We define value to be ratio of quality of healthcare services to their cost. Value can be increased at each one of the steps in the treatment process diagrammed below.

We use biomedical engineering and industrial engineering methods to increase value. Several past projects are summarized below to illustrate our philosophy. If you are a student looking for a research project, please contact me for specific projects related to the ideas outlined here. If you are interested in a full list of publications, click here.

Diagnosis of Rotator Cuff Tears

Proper diagnosis is critical to providing high value healthcare. Improper diagnosis can subject a patient to the underuse, overuse, or misuse of therapies. Any of these three events leads to a reduction in health care value. We sought to improve the diagnosis of a shoulder disorder known as a rotator cuff tear using a technique from machine learning known as a least-squares support vector machine (LS-SVM). We were able to classify shoulders with and without rotator cuff tears from simple isometric shoulder strength measurements about as well as ultrasound can, but without the expense of imaging.

Selected relevant publications:

Silver, A.E., Lungren, M.P., Johnson, M.E., ODriscoll, S.W., An, K-N, and Hughes, R.E. (2006) Using support vector machines to optimally classify rotator cuff strength data and quantify post-operative strength in rotator cuff tear patients. Journal of Biomechanics 39: 973-979. Abstract

Selected resources:

1. If you are not familiar with mathematical optimization, you may want to check out an online educational tool for linear programming.

2. We use an optimization-based method known as support vector machine (SVM) modeling. A good website to learn more about SVM modeling is LS-SVM Lab.

Pre-operative Planning of Distal Humerus Fracture Fixation

The success of treatment can depend on the quality of the pre-operative plan developed prior to surgery, and this is illustrated in the transition from Treatment Planning to Treatment in the above diagram. Consider the problem of placing screws in the distal humerus in an effort to provide stability to a comminuted distal humerus fracture. The surgeon faces may possible screw placement combinations. Some combinations of screws are not feasible because the screws would intersect. Moreover, not all plate holes must have screws placed through them. However, no hole may have more than one screw. We can formulate the problem of determining optimal screw placement by formulating and solving an integer programming model in which the decision variables are Boolean (one 1 if there is a screw; 0 otherwise). The result of one model simulation is illustrated in the figure to the right.

Selected relevant publications:

Maratt, J.D., Peaks, Y-S, Doro, L.C., Karunakar, M.A., and Hughes, R.E. (2008) An integer programming model for distal humerus fracture fixation planning. Computer Aided Surgery 13(3): 139-47. Abstract

Optimization of Conservative Management of Rotator Cuff Tears

Mathematical models can be used to optimize the delivery of treatment. We have formulated a mathematical model (integer program) for optimizing non-operative treatment of rotator cuff tears. Physical therapy is a critical element of non-operative management of these disorders. One fundamental fact we all live with is a 24-hour day. Each of us seeks to get as much as possible of out the available time. What exercises should a patient do to maximize the rehabilitation benefit if a person has a limited amount of time for physical therapy exercises? The approach is to formulate an integer programming model to optimally choose a set of exercises to perform. The model is based on a three-dimensional biomechanical model of the shoulder developed by the Delft Shoulder Group and implemented in AnyBody modeling software. This project is being conducted in collaboration with James Carpenter, M.D., who is an orthopaedic surgeon.

Selected relevant publications:

Hughes, R.E., Rock, M.G., and An, K-N (1999) Identification of optimal strategies for increasing whole arm strength using Karush-Kuhn-Tucker multipliers.  Clinical Biomechanics  14: 628-634. Abstract

Gatti, C.J., Scibek, J., Svintsitski, O., Carpenter, J.E., and Hughes, R.E. (2008) An integer programming models for optimizing shoulder rehabilitation. Annals of Biomedical Engineering 36(7):1242-53. Abstract

Quantification of the Cost of Variability in Total Knee Replacement Prosthetic Component Alignment

Although healthcare has adopted some techniques for improving quality of care from manufacturing (Total Quality Management and Six Sigma), medicine has not utilized all of the important developments of statistically-based quality improvement methods developed in industry. For example, the methods developed by Dr. Genichi Taguchi in the 1980s for incorporating quality considerations in the design phase have not been widely applied to biomedical device design, especially in orthopaedic applications. A central part of his theory was the development of a loss function that relates deviation from an ideal target value, as opposed to a range of acceptable values, to an increased monetary cost to society. This loss is a continuous function; the further the characteristic varies from the proper value, the greater the loss. The methods developed by Taguchi have proven useful in non-healthcare industries; biomedical engineers engaged in medical device design should also be able to benefit from these ideas. The long-term objective of this research is to develop tools that can be used by orthopaedic implant designers to improve the quality of patient care and improve outcomes for total knee arthroplasty (TKA) surgery. Figure 2 illustrates the Taguchi Loss function we developed for TKA as a function of the tibiofemoral angle of the femoral prosthetic component. We have performed some biomechanical studies to develop methods for reducing the variability of component alignment, which should (according to the Taguchi Loss Curve) reduce cost.

Selected relevant publications:

Kusuma, S., Urquhart, A.G., and Hughes, R.E. (2009) Taguchi loss function for varus/valgus alignment in total knee arthroplasty. The Open Biomedical Engineering Journal 3:39-42. Click here for full access to full paper.

Doro, L.C., Hughes, R.E., Miller, J.D., Schultz, K.F., Hallstrom, B.R., and Urquhart, A.G. (2008) The reproducibility of a kinematically-derived axis of the knee versus digitized anatomical landmarks using a knee navigation system. The Open Biomedical Engineering Journal 2: 52-56. Click here for access to full paper.

Doro, L.C., Hughes, R.E., and Urquhart, A.G. (2008) Enhancing a kinematically-derived axis of the knee using a knee navigation system. Journal of Musculoskeletal Research 11(3): 127-133.

[Return to LOCOS home page]

Last updated 10/2/08