DEVELOPMENT OF THREE-DIMENSIONAL BIOMECHANICAL MODEL OF THE GLENOHUMERAL JOINT

Participants: S. LaScalza, R.E. Hughes, J.E. Kuhn, J.E. Carpenter, L.J. Huston

Keywords: shoulder, EMG, muscle model, muscle forces, glenoid force

Introduction

This project is the start of a long-term effort to investigate the biological response of the rotator cuff to repetitive loading. The ultimate goal is to determine what levels and frequency of rotator cuff loading produces injury, training effect, or maintenance of the tissue. This will involve investigating the effect of repetitive mechanical stresses on the extracellular matrix of tendons. The first step, however, is to accurately estimate the magnitude of forces generated by the rotator cuff during realistic sports and occupational tasks, as this information will be required to determine test conditions for in vitro experiments. Muscle forces are commonly estimated by either using a numerical optimization scheme to partition net intersegmental moment to muscles or by using electromyographic (EMG) recordings. Although optimization-based models exist for estimating shoulder muscle forces (Karlsson and Peterson, 1992; van der Helm, 1994), they do not predict co-contraction of antagonist muscles well. Co-contraction is particularly important at the shoulder due to the high demands for joint stability necessitated by the shallow geometry of the glenohumeral articular surface. Models based on EMG predict co-contraction, so an EMG-based shoulder model should be more appropriate for analyzing the rotator cuff musculature (de Luca and Forrest, 1973; Laursen et al., 1998). The purpose of the proposed project is to estimate rotator cuff forces during sporting, occupational, and rehabilitation activities using a three-dimensional EMG-driven biomechanical model of the glenohumeral joint. The rotator cuff forces will be used to estimate the joint reaction force acting at the glenohumeral joint.

Materials and Methods

The model will be formulated using engineering mechanics, where moment and force equilibrium conditions will be constructed for the joint. A mathematical model of the skeletal system will be used to determine the moments produced by the muscles at the glenohumeral joint. The force generated by each muscle will be a multiplicative model involving (1) muscle specific tension, (2) EMG (normalized to maximum voluntary contraction level), (3) a length-tension modulation factor, and (4) a force-velocity modulation factor. The parameters of the length-tension relationship will be estimated by fitting maximum predicted joint torque to maximum measured joint torque through the shoulder range of motion. The effect of tendon compliance on muscle length during active muscular contraction will be incorporated. The supraspinatus, infraspinatus, subscapularis, teres minor, teres major, latissimus dorsi, pectoralis major, biceps, triceps, coracobrachialis, and deltoid muscles will be included in the model. Upper extremity kinematics will be measured using an electromagnetic tracking system, and a sixteen channel electromyographic recording system will be used to measure muscle activation. Since electromyographic recordings at the shoulder are highly correlated, principle components regression will be used for parameter estimation.

Two experiments will be conducted. One experiment will be used to estimate the optimal muscle lengths. Based on the musculoskeletal geometry and estimated parameters, arm strength will be predicted for abduction, internal rotation, and external rotation. Predictions will be made for isometric and isokinetic exertions. Isokinetic measurements will be made using a Biodex dynamometer (Biodex Corp., Shirley, NY). Predicted and measured isokinetic strengths will be compared to evaluate the structure of the musculoskeletal model. A second experiment will be conducted to estimate muscle and joint reaction forces from electromyographic and kinematic data. Muscle and joint reaction forces will be estimated for isometric and dynamic abduction, adduction, internal rotation, and external rotation. Rehabilitation exercises involving these same motions will also be studied.

Progress

The software has been developed for collecting arm kinematic data.

References

1. de Luca, C. J. and Forrest, W.J. (1973) Force analysis of individual muscles acting simultaneously on the shoulder during isometric abduction. J. Biomechanics 6: 385-393.

2. Karlsson, D., and Peterson, B. (1992) Towards a model for force predictions in the human shoulder. J. Biomechanics 25: 189-199.

3. Laursen, B., Jensen, B.R., Nemeth, G., and Sjogaard, G. (1998) A model predicting individual shoulder muscle forces based on relationship between electromyographic and 3D external forces in static position. J. Biomechanics 31: 731-739.

4. van der Helm, F.C.T. (1994) A finite element musculoskeletal model of the shoulder mechanism. J. Biomechanics 27: 551-569.