Walking Direction and Muscle Forces


I’m modelling gait using a modified version of the GaitUniMiami application (AnyBody Version 3) but am getting unusual results.

The way our gait lab is set up, subjects walk either in the + or – y direction with the x direction across the walkway (medial/lateral) and the z direction up into the ceiling (right hand coordinate system). At the moment, if we have the same subject walk in the +y direction in one trial but –y direction in another trial and then look at the calculated muscle forces for those trials, they are different. (The trends are different with the muscle forces being “on” at different points of the gait cycle.) To try to figure out why this is occurring, I have looked at the joint angles (AnyVar flexion, abduction, rotation, etc) AnyBody returns from those trials, and the joint angles are the same regardless of the direction of walking so this is reading as expected.

Therefore, I think there may be an issue with the force plate data being read into the Environment file. Is there a change that needs to be made within the Environment file to account for walking direction? Or is this perhaps tied to the way c3d2any splits out the c3d Analog data? (Earlier we were having problems with how the force plates were assigned in the .analog file – we had data assigned to force plates 2 and 3 – without a force plate 1 – and so had to manually change the force plate assignments after running c3d2any…)

Thank you,

Hi Jennifer,

There is a mistake that i have seen a couple of times in similar cases, probably it is the same one. There is a reaction force between each foot and the plate it steps on (right foot and plate1 for example). When running a trial on the opposite direction (-y instead of +y) it is likely that each foot doesn’t step on the same plate as before (left foot on plate1 this time for example).

In this case if you still have the same reaction force as in the first trial it will definitly gives you strange activation patterns, because the reaction force goes to the wrong foot. So you have to take care to modify those reactions in the environment file if necesary.

Best regards, Sylvain.

Thank you. That sorted out the problem.