Traditionally biomechanical models of the lumbar spine have assumed that the forces exerted by the trunk muscles in response to external loads are deterministic. That is, given a set of lifting conditions (torque, position, velocity etc) the activation of the trunk muscles is precisely determined such that equilibrium exists. However, because of the indeterminate nature of the biomechanical system, variability in the individual muscle forces is not only possible, it is likely. Variability in the individual muscle forces alters the loading patterns on the spine because each muscle has its own vector line of action which has a component in compression, anterior shear and lateral shear. Therefore by omitting the stochastic nature of trunk muscle activation from the analysis, it is believed that deterministic biomechanical models are not capable of modelling the ranges of potential spine reaction forces and ultimately are underestimating risk of manual materials handling tasks.
The specific aim of this proposed research project is to develop a stochastic model of the lumbar spine during lifting. This model will use stochastic principles to predict the activation levels of ten trunk muscles under occupational lifting conditions including varied weights, postures, dynamic components and asymmetric lifting. This stochastic model will be developed in two phases. In the experimental phase human subjects will be asked to perform highly controlled simulated lifting motions repeatedly. The electromyographic activity of ten trunk muscles will be sampled and will be used to develop a database describing that particular lifting motion. In the modelling phase this database will be used in a simulation model which will generate muscle activities. Multiple runs of the simulation model will generate possible time dependent EMG traces suitable for input into an EMG assisted biomechanical model which will then generate stochastic spine reaction forces. The broad, long term objective of this research is to develop a model which can be used on the shop floor which will give an accurate representation of the internal trunk muscle forces, as well as spine reaction forces, as the worker performs his/her job. This could potentially allow ergonomists to make modifications in work station design and immediately assess the effectiveness of these interventions.
| Mirka, G A; Glasscock, N F; Stanfield, P M et al. (2000) An empirical approach to characterizing trunk muscle coactivation using simulation input modeling techniques. J Biomech 33:1701-4 |
