simtk.org

Dear all

I want to manually scale an osim model to obtain different models with full range of heights and weights (20 models including male and female). Later, I aim to use them in my simulation to check whether my approach robustness. To do this, I wonder if it's true to reduce/increase the height and weight of the original model in various percentage. For example, I consider +/-5%,10%,15%,20%,25%,30%. Do you think the scaling factors of the height and weight are the same? By doing this I have acceptable models?

Or do you have any idea to get a group of subjects with reliable mean and standard deviation?

Best regards

Taylor Dick presented an abstract at NACOB last summer (page 159 of this PDF) that scaled body segment dimensions proportional to the 0.33 power of mass. If you're looking for height/weight means and SDs, the Fukuchi et al. datasets for walking and running are great resources for healthy adult data (there's a general demographics file in the dataset with height, weight, and sex).

Either way you may want to scale muscle strength as well - I have a demo on GitHub of using the Handsfield et al. regression equations for the lower body; the abstract from above just scales muscle strength with the 0.67 power of mass, which is probably the way to go if you are scaling muscles that aren't in the Handsfield et al. data.