Contribution of modelling soft tissue properties in simulations of dynamic movements

• Published in: Journal of sports sciences 2005-12, Vol.23 (11-12), p.1152-1152
• Main Authors: Gittoes, M, Brewin, M, Kerwin, D
• Format: Article
• Language: English
• ISSN: 0264-0414

Wobbling mass models, which accommodate soft tissue motion, have been suggested to replicate more realistically the kinetics of dynamic movements than rigid mass models. The greater complexity of wobbling mass models and the potential improvements in reproducing human motion are, however, challenged by the need for a greater number of model parameters. Insight into the benefits gained from increasing the sophistication of simulation models of dynamic movements is limited, since the "true" accuracies of wobbling and rigid mass models have not been thoroughly evaluated for a corresponding movement. A model's output may be quantitatively compared with an actual performance to assess the level of confidence that can be placed in the model and to establish the "true" model accuracy. The aim of this investigation was to examine the contribution of modelling soft tissue properties in simulations of dynamic movements by comparing the accuracy of a wobbling and a rigid mass model. A customized wobbling mass model of a female performing a landing was developed. The model comprised a rigid foot segment and shank, thigh and upper-body segments represented by wobbling and rigid masses, which were connected by spring-damper systems. Wobbling mass and ground-contact spring parameters were estimated for the wobbling mass model using an optimization procedure. A rigid mass model was created by setting all wobbling mass spring stiffness values to 1.0 x 10 super(10) N times m super(-1) to prevent relative motion between wobbling and rigid masses, while retaining the optimized ground-contact parameters of the wobbling mass model. Kinematic data from two actual drop-landings (height 0.46 m) performed by a female (age 20 years, mass 69.0 kg) were used to initiate and drive the models. Each model's accuracy in replicating the two landings was determined using ground reaction force data from the actual performances. The wobbling mass model replicated the measured ground reaction forces more successfully than the rigid mass model. The root mean squared differences between the measured and simulated profiles produced with the wobbling and the rigid mass model were 15.7% and 19.7% respectively, when expressed as a percentage of the measured force range. Similar mean differences of 7 ms (3.3%) and 6 ms (2.8%) were found between the measured and simulated times to peak vertical ground reaction force for the wobbling and the rigid mass model, respectively. The wobbling mass model repli