Parameter study on Marker error

Dear all,

I am currently analysing some data taken from a recent study. I would
like to run a parameter study on a few key marker positions. Has
anyone managed to do a parameter study that takes into account a 2mm
marker movement radius (trying to take into account skin movement
artifact and data to model marker postions estimation).

I have gone through the tutorials, but i am struggling a little on
creating a radius as a parameter.

best wishes

Peter

Hi Peter

I assume you are using the GaitApplication…

I have not tried this so there might be other ways of doing this, here is my
suggestion:

1 Run gaitapplication2 with the Scale option on, to get the best possible
solution

2 Update the LocalmakerCoordinatesAndSize files with the found solution in
the file outputpar.any so that all segments and markers will be at the
optimal position once the model is loaded.

3 Manually introduce the 2 mm error on the marker in question in the
localmarkerandcoordinateanssixe.any file

4 Rerun the model using the POSONLY option , if you did not introduce the
error it should give the same solution as before but the error should
introduce a slight difference in the found motion

You can not do this directly as a parameter study in AnyBody if you are
using the GaitApplication2, because AnyBody will not execute the
gaitapplication2. The only way to do this in an automatic way would be
control the gaitapplication2 and the console application AnyBodyCon from
another program such as Matlab or similar.

Best regards

Søren, AnyBody Support


From: anyscript@yahoogroups.com [mailto:anyscript@yahoogroups.com] On Behalf
Of peter.worsley
Sent: 03 November 2008 12:03
To: anyscript@yahoogroups.com
Subject: [AnyScript] Parameter study on Marker error

Dear all,

I am currently analysing some data taken from a recent study. I would
like to run a parameter study on a few key marker positions. Has
anyone managed to do a parameter study that takes into account a 2mm
marker movement radius (trying to take into account skin movement
artifact and data to model marker postions estimation).

I have gone through the tutorials, but i am struggling a little on
creating a radius as a parameter.

best wishes

Peter

[Non-text portions of this message have been removed]

Hi Soren,

I have now written a Matlab program utilising the AnyBodyCom function, to
analyse variance in marker position estimation pre GaitApplication2 execution.

The only problem I have now is automating the saving of StudyGait string after
setting the inital pos in MarkerPlacementStudy. I have experiemented using
AnyOutputFile’s but with no luck.

your thoughts would be greatly appreciated as always

best wishes

Peter

— On Mon, 3/11/08, AnyScript Support <anyscriptsup@anybodytech.com> wrote:
From: AnyScript Support <anyscriptsup@anybodytech.com>
Subject: RE: [AnyScript] Parameter study on Marker error
To: anyscript@yahoogroups.com
Date: Monday, 3 November, 2008, 9:16 PM

Hi Peter

I assume you are using the GaitApplication…

I have not tried this so there might be other ways of doing this, here is my
suggestion:

1 Run gaitapplication2 with the Scale option on, to get the best possible
solution

2 Update the LocalmakerCoordinatesAndSize files with the found solution in
the file outputpar.any so that all segments and markers will be at the
optimal position once the model is loaded.

3 Manually introduce the 2 mm error on the marker in question in the
localmarkerandcoordinateanssixe.any file

4 Rerun the model using the POSONLY option , if you did not introduce the
error it should give the same solution as before but the error should
introduce a slight difference in the found motion

You can not do this directly as a parameter study in AnyBody if you are
using the GaitApplication2, because AnyBody will not execute the
gaitapplication2. The only way to do this in an automatic way would be
control the gaitapplication2 and the console application AnyBodyCon from
another program such as Matlab or similar.

Best regards

Søren, AnyBody Support


From: anyscript@yahoogroups.com [mailto:anyscript@yahoogroups.com] On Behalf
Of peter.worsley
Sent: 03 November 2008 12:03
To: anyscript@yahoogroups.com
Subject: [AnyScript] Parameter study on Marker error

Dear all,

I am currently analysing some data taken from a recent study. I would
like to run a parameter study on a few key marker positions. Has
anyone managed to do a parameter study that takes into account a 2mm
marker movement radius (trying to take into account skin movement
artifact and data to model marker postions estimation).

I have gone through the tutorials, but i am struggling a little on
creating a radius as a parameter.

best wishes

Peter

[Non-text portions of this message have been removed]


Users of the AnyBody Modeling System help each other create biomechanical
models in the AnyScript language.Yahoo! Groups Links

[Non-text portions of this message have been removed]

Hi Soren,

sorry, I meant the ModelGait string. The study gait will not change from each
run of the trial.

many thanks

Peter

— On Mon, 3/11/08, AnyScript Support <anyscriptsup@anybodytech.com> wrote:
From: AnyScript Support <anyscriptsup@anybodytech.com>
Subject: RE: [AnyScript] Parameter study on Marker error
To: anyscript@yahoogroups.com
Date: Monday, 3 November, 2008, 9:16 PM

Hi Peter

I assume you are using the GaitApplication…

I have not tried this so there might be other ways of doing this, here is my
suggestion:

1 Run gaitapplication2 with the Scale option on, to get the best possible
solution

2 Update the LocalmakerCoordinatesAndSize files with the found solution in
the file outputpar.any so that all segments and markers will be at the
optimal position once the model is loaded.

3 Manually introduce the 2 mm error on the marker in question in the
localmarkerandcoordinateanssixe.any file

4 Rerun the model using the POSONLY option , if you did not introduce the
error it should give the same solution as before but the error should
introduce a slight difference in the found motion

You can not do this directly as a parameter study in AnyBody if you are
using the GaitApplication2, because AnyBody will not execute the
gaitapplication2. The only way to do this in an automatic way would be
control the gaitapplication2 and the console application AnyBodyCon from
another program such as Matlab or similar.

Best regards

Søren, AnyBody Support


From: anyscript@yahoogroups.com [mailto:anyscript@yahoogroups.com] On Behalf
Of peter.worsley
Sent: 03 November 2008 12:03
To: anyscript@yahoogroups.com
Subject: [AnyScript] Parameter study on Marker error

Dear all,

I am currently analysing some data taken from a recent study. I would
like to run a parameter study on a few key marker positions. Has
anyone managed to do a parameter study that takes into account a 2mm
marker movement radius (trying to take into account skin movement
artifact and data to model marker postions estimation).

I have gone through the tutorials, but i am struggling a little on
creating a radius as a parameter.

best wishes

Peter

[Non-text portions of this message have been removed]


Users of the AnyBody Modeling System help each other create biomechanical
models in the AnyScript language.Yahoo! Groups Links

[Non-text portions of this message have been removed]

Hey Peter,

I have a slightly different suggestion for how to study how changes in the
marker positions alter the motion.

Use step 1-3 as Søren suggested to obtain the best possible marker positions
and segment lengths, update the LocalMarkerCoordinatesAndSize.any file with the
optimal solution, load the model, run SetInitialConditions in the
MarkerPlacementStudy and save a new ModelGait.txt file. This ModelGait.txt file
is the one where you wish to introduce small changes in the coordinates of the
markers.

4). Create a matlab script that can write out the ModelGait.txt with the numbers
that are already in it except all the MARKER specifications. - But use a
different name for the file, e.g. ModelGait-<MarkerChangedName>.txt

  1. Create a variable in matlab for each marker and initialize them with the
    optimized local marker coordinates.

Then run the following loop in matlab:

  1. Introduce a small change to the marker you wish to change.
  2. Write all the MARKER specifications in the ModelGait-<MarkerChangedName>.txt
  3. Execute GaitApplication2 with the following syntax: >> GaitApplication2
    ModelGait-<MarkerChangedName>.txt StudyGait.txt (remember to change to POSONLY
    analysis)
  4. When the analysis has completed, copy the file output-euler.txt to
    output-euler–<MarkerChangedName>.txt and the output-par.any to
    output-par–<MarkerCangedName>.any
  5. Go to 1 and repeat with a new marker change (remember to put the change you
    just made back to the initial value).

When you have run this for all the markers you are interested in, you can
visualize each of the results in new AnyBody models by making the two following
small changes:
In the JntDriver in JointAndDriversOptimized.any change the filename
output-euler.txt to the output-euler–<MarkerChangedName>.txt and in the
HumanModel.any change the output-par.any include statement to
output-par–<MarkerChangedName>.any.

I hope this helps you else write again.

Best regards
Michael Skipper Andersen
The AnyBody Research Project

----- Original Message -----
From: Peter Worsley
To: anyscript@yahoogroups.com
Sent: Wednesday, November 12, 2008 10:57 AM
Subject: RE: [AnyScript] Parameter study on Marker error

Hi Soren,

sorry, I meant the ModelGait string. The study gait will not change from each
run of the trial.

many thanks

Peter

— On Mon, 3/11/08, AnyScript Support <anyscriptsup@anybodytech.com> wrote:
From: AnyScript Support <anyscriptsup@anybodytech.com>
Subject: RE: [AnyScript] Parameter study on Marker error
To: anyscript@yahoogroups.com
Date: Monday, 3 November, 2008, 9:16 PM

Hi Peter

I assume you are using the GaitApplication…

I have not tried this so there might be other ways of doing this, here is my
suggestion:

1 Run gaitapplication2 with the Scale option on, to get the best possible
solution

2 Update the LocalmakerCoordinatesAndSize files with the found solution in
the file outputpar.any so that all segments and markers will be at the
optimal position once the model is loaded.

3 Manually introduce the 2 mm error on the marker in question in the
localmarkerandcoordinateanssixe.any file

4 Rerun the model using the POSONLY option , if you did not introduce the
error it should give the same solution as before but the error should
introduce a slight difference in the found motion

You can not do this directly as a parameter study in AnyBody if you are
using the GaitApplication2, because AnyBody will not execute the
gaitapplication2. The only way to do this in an automatic way would be
control the gaitapplication2 and the console application AnyBodyCon from
another program such as Matlab or similar.

Best regards

Søren, AnyBody Support


From: anyscript@yahoogroups.com [mailto:anyscript@yahoogroups.com] On Behalf
Of peter.worsley
Sent: 03 November 2008 12:03
To: anyscript@yahoogroups.com
Subject: [AnyScript] Parameter study on Marker error

Dear all,

I am currently analysing some data taken from a recent study. I would
like to run a parameter study on a few key marker positions. Has
anyone managed to do a parameter study that takes into account a 2mm
marker movement radius (trying to take into account skin movement
artifact and data to model marker postions estimation).

I have gone through the tutorials, but i am struggling a little on
creating a radius as a parameter.

best wishes

Peter

[Non-text portions of this message have been removed]


Users of the AnyBody Modeling System help each other create biomechanical
models in the AnyScript language.Yahoo! Groups Links

[Non-text portions of this message have been removed]

[Non-text portions of this message have been removed]