The ability to determine muscle and joint forces in vivo is important for a variety of reasons. The two reasonable ways to do this at the present time are to make direct measurements with implanted transducers and to calculate the forces from mathematical models. Each of the two techniques has different advantages and disadvantages. Thus, the two would be complementary when combined. The mathematical approach is noninvasive and allows force determination in many situations and subjects, while the direct measurements allow validation of the mathematical calculations, but cannot be used in many situations. We propose to develop an animal model in which we can make simultaneous direct measurements and mathematical calculations of muscle and joint forces. Dogs will have multiple liquid metal strain gauges (LMSGs) attached to hind limb tendons and foil gauges to the femoral neck. The dogs will be trained to walk in the gait laboratory and all measurements necessary to calculate muscle and joint forces (displacement histories and foot-floor reactions) will be made while recording output from the strain gauges. Simultaneous EMG recordings will also be made. The animals will then be sacrificed for post-calibration of the strain gauges so that we may directly determine muscle and hip joint forces, to determine the inertial properties of the body segments, and to create a muscle model. The muscle and joint forces will then be calculated and compared to the EMG recordings and the measured muscle and joint forces. We will also refine our mathematical models by: (1) determining the contributions of passive elements to the hip moments and determining their effect on muscle and joint force predictions, and (2) developing a scheme that guarantees continuous muscle function by accounting for the contraction history. The first of these two refinements will be accomplished by experimentally measuring the passive moments at the hip in human subjects and then incorporating those moments into the gait analyses of the same subjects. The second refinement will be accomplished by using Hill's model to establish upper and lower bounds on the muscle force predictions at each subsequent discrete time during the gait cycle for which a muscle force is predicted.
| Cheng, R C; Brown, T D; Andrews, J G (1990) Non-uniqueness of the bicompartmental contact force solution in a lumped-parameter mathematical model of the knee. J Biomech 23:353-5 |
