simtk.org

I am using Hamner SR and Delp SL. Muscle contributions to fore-aft and vertical body mass center accelerations over a range of running speeds. Journal of Biomechanics; 46(4), 780-7. (2013) Subjects to run my own simulations.
I ran several simulations and got lots of datasets; However, when I tried to get the data means and standard deviation, I realized that there is a phase delay between two different subjects in each cycle. I was wondering how can I fix this problem in OpenSim or if there is a way for fixing it on Matlab, I would appreciate your help.

Furthermore, I download an Ideal Assistive Device paper ( Uchida et al) dataset and when I was looking into their data, I realized that they have data from 0 to 100 through the gait cycle percentage.
How we can get such data from OpenSim or, again, is there any possible solution for it on Matlab?.

I decide to send this email to you because it is so urgent for me to solve this issue as soon as possible. I would be grateful if you help me with this issue.

I am looking forward to hearing from you as soon as possible.
I have attached one of the figures to give a better understanding of the problem if you found it necessary.

I was wondering how can I fix this problem in OpenSim or if there is a way for fixing it on Matlab

It is a pretty standard practice to normalize gait data since gait events can vary from trial to trial. You will have to write code in Matlab that normalizes each stride from heel-strike to heel-strike.

How we can get such data from OpenSim or, again, is there any possible solution for it on Matlab?.

You don't get normalized data from OpenSim, you write your own function that time normalizes results.

Dear Professor,
Actually, as you can see in the pdf that was attached to the first post. Data's timeline was eliminated and it is normalized to 0-100 as a gait percentage.
Furthermore, I checked the Hamner and Delp model (Opensource available data on OpenSim website) they exactly divide each subject running into three toe-off to toe-off cycles.
Could you please explain what is your meaning by normalization? I still could not figure out how should I normalize each gait cycle such that their data match together.

I am looking forward to hearing from you as soon as your convenience time.

Ali.

alik wrote: ↑

Thu Nov 22, 2018 12:53 am

Dear Professor,
Actually, as you can see in the pdf that was attached to the first post. Data's timeline was eliminated and it is normalized to 0-100 as a gait percentage.
Furthermore, I checked the Hamner and Delp model (Opensource available data on OpenSim website) they exactly divide each subject running into three toe-off to toe-off cycles.
Could you please explain what is your meaning by normalization? I still could not figure out how should I normalize each gait cycle such that their data match together.

I am looking forward to hearing from you as soon as your convenience time.

Ali.

I am looking forward to getting an answer about this topic.

Could you please explain about normalization to a gait cycle more explicitly?
It would be great if someone can share a reference code or document for normalization.

It is really urgent for me to make sure that I am getting correct data.
Ali.

To Whom it may concern,

This post is a clarification of what I am doing to get an adequate and proper answer


1) I am using Hamner SR and Delp SL. Muscle contributions to fore-aft and vertical body mass center accelerations over a range of running speeds. Journal of Biomechanics; 46(4), 780-7. (2013) Model and datasets to generate some simulation.

2) I did simulations for each subject in three different Cycles and These simulations have been done for several subjects.

3) I am deleting those data's time and substituting it with :
4) Then, I am modifying each matrix size such that all of the matrices have the same size.


However, When I am plotting all of the subjects simultaneously, peaks of the subjects, As you can see attached PDF, are happening at different times.
It should be noted that Green plots are three gait cycle for a subject and Blue is another subject in three gait cycle.

I do not know whether these phenomena are correct or not and I would appreciate if you can confirm my normalization procedure.

Secondly, You introduce me "pretty standard practice to normalize gait data" which is writing a code that "code in Matlab that normalizes each stride from heel-strike to heel-strike."
I could not find any documentation or any relevant code and It would be great if you introduce a document or example of a code to do this procedure.

Sincerely,
Ali.

I am deleting those data's time and substituting it with: Time = linspace(0,100,length(Subject_CMC_Force))

Please check whether the time points in your Subject_CMC_Force data are equally spaced; if not, the substitution shown above will distort the timing of your data.

tkuchida wrote: ↑

Sat Dec 01, 2018 8:04 pm

Please check whether the time points in your Subject_CMC_Force data are equally spaced; if not, the substitution shown above will distort the timing of your data.

.

I really appreciate your proper and exact answer. As you said, time intervals are unequal and it is distorting data.

I would be really grateful if you let me know how we can normalize our data in this condition. I am so curious about your own paper "Ideal Assistive Device" data normalization; Could you please share an example of data normalization or introduce a documentation for doing this normalization.

Hello Ali,

Have a look at the example provided in the following thread on the Mathworks forum:

https://uk.mathworks.com/matlabcentral/ ... tion-force

I hope that helps!

Kind regards,

Nicos Haralabidis