View Full Version : Summary of Responses: Foot pressure measurement

11-19-2007, 01:49 AM
Following is a summary of my original question and the responses received. Many thanks to everyone who replied.

Emma Pratt
Pre-Registration Clinical Scientist
Douglas Bader Rehabilitation Centre.
Queen Mary's Hospital,
Roehampton Lane,
London SW15 5PN

*Summary of Original question*

We are currently conducting research into the outcome measures used to analyse foot pressure measurement data, and methods of clinical reporting. We use the Tekscan F-scan system.
1. How do different centres report foot pressure measurements?
2. We would like to create a pressure image from a number of normals
with a COP trajectory overlaid, to compare clinically collected data to.
Does anyone have any idea of how we could do this using the Tekscan software/Excel etc?
The TekScan software allows you to export data is ASCII format (.txt file). You can read the files with Excel. Use Excel to normalize each of your individual trials to stance (100%). Then calculate an ensemble average of the trials

Kerry K. Rambarran, The Ottawa Hospital
Perhaps the attached paper can be useful in your research [Gait & Posture 22 (2005) 372-376. Assessment of sub-division of plantar pressure measurement
in children. Stebbins, J. et al.]. Dr. Michael Schwartz also developed a study
on the paths of the pressure centers (CoP)

Wagner de Godoy, Brazil
It seems that you use clinical software of Tekscan. If you use Matscan Research software, you can better manage your data. this enables you to easily export your data as ASCII and then you may open them in Matlab or excel. Still I am not sure whether averaging is a suitable way for representing the data because we don't have a so called "averaged walker" however I think you need to first normalize your data, for example use the length of the foot, or the weight of the subject, based on your aim. Tothe best of my knowledge, the best way for processing the data is to export them to Matlab.

Behdad Tahayori, School of Rehabilitation, Iran
Graphs of individual force time integrals for discreet areas of interest expressed as mean +/-sd for your sample population may look ok and give the comunication you desire. I have just finished a project where I looked at discreet plantar forces under the 1st mpj and hallux. I used Rs Scan 7, which can produce tabled summeries about time and force patterns. As you know the graphical output is not very useful for analysis of inter and intra subject comparisons. I made several runs of the pressure mat and statistically analysed the data in excel to produce one table of mean +/-sd parameters. These
are only statistics of point of interest tho iE peak force magnitude, time of peak force, etc. so they don't make very good graphs. I believe there is software that can resolve a graph into tabular x,y,data. You could digitize the graphs of each individual and process the data in excell to give means +/-sd of the data of interest and then produce a graph of the means with sd+/- that represents your population.The only problem is that it seems that force or pressure pattern integrals are extremely variable intra subject so that any mean and sd of many subject may not be very useful. Useful papers: Variability of the Centre of Pressure Pattern Integral During Walking. McPoil TG, Cornwall M, (JAPMA 88 6 259-267 1998). Generalizability of In-Shoe Peak Pressure Measures Using the F-Scan System. Mueller MJ, Strube MJ. (J Clin Bmech 11 3 159-164 1996)

David Smith
Gait and Posture 21 (2005) 255-262. Centre of pressure displacements in trans-femoral amputees during gait. Schmid, M. et al.
Chi-Wen, Lung, National Yang-Ming University
I am handling sort of the same problem in a research group in Denmark. We are not using Tekscan but RSScan. We have conducted a big experiment with pressure plates and done all the normal clinical measures on the level of pronation like x-ray, objective measures, goniometers and so on. I am starting to analyse the pressure data very soon, but I don't think there is a good way of getting a measure outfrom the pressure distribution. I am planing to code my own algorithm to get exactly that out what I need, that gives us the possibility to calculate a measure we think is intuitively right for the level ofpronation for example.I guess the way people do it now is they look at the navicular drop and angle of calcaneus, but how does that correlate to the pressure distribution?

Christian Gammelgaard, Denmark
__________________________________________________ _______________
Celeb spotting Play CelebMashup and win cool prizes