Stance phases are not properly separated with lightweight subjects

Question/Problem: When we have smaller, lightweight subjects (less than 50 pounds), stance phases are not properly separated.  One stance phase will get separated into two separate stances.  For example, a single right foot strike will get separated into the heel strike and the push-off portion of the stance phase.

Answer/Solution: The software decides when one footstep has ended and another has begun based on a percentage of the difference in the maximum pressure and the minimum pressure (above a certain threshold) in the movie.  This difference will be smaller when the subject does not weigh very much.  The resulting percentage of the difference will be smaller too.  So that is the reason that the problem is occurring.

Here is some information on the algorithm that our software uses to determine when one foot step ends and the next one begins.  The equation is:

F-separate = ((FMax – Fmin) * PhaseThreshold) + FMin

In software versions before 5.83, the value PhaseThreshold is set to 10% (0.10) and this cannot be changed.  With lighter subjects where FMax is lower, this can potentially separate a footstep at midstance if there is a very low valley in the bimodal force curve.  Starting with version 5.83 software, this PhaseThreshold value can be set in the *.ini file.  You are then able to lower the PhaseThreshold which may prevent the system from splitting footsteps at midstance.

However, if you are using the first contact and last contact trigger methods to start and stop the acquisition during a recording, lowering this PhaseThreshold has the potential to merge multiple steps together.  This recording method prevents the force values from ever reaching zero during a recording.  Sometimes the transition from one step to the next may exceed the step separation threshold.  We would recommend that you set your trigger start method to first contact and the stop method to frame count.  You can then set the frame count to a number of frames which will exceed the duration of an anticipated footstep.  Then, between each step there will be frames that have a force value of 0.

This procedure may also help to lower the force threshold for step separation in earlier software versions.  Since the formula incorporates the lowest force value during the recording (FMin), if there are frames with 0 force between every step, this FMin value will be 0.  When the first contact and last contact trigger settings are used, the FMin value is higher and this increases the step separation threshold.

Finally, this step separation algorithm is a display feature that is applied to the data when you open the file.  It is not a permanent change to the data.  So, if you were to upgrade to version 5.83 software and lower the PhaseThreshold value in the *.ini file, then this would correct the display problem for recordings made in earlier software versions.

Tekscan, Inc.
307 West First Street. South Boston, MA. 02127-1309. USA
tel: 800.248.3669 / 617.464.4500 fax: 617.464.4266
marketing@tekscan.com


Print Page