Vicom markers and static balance

Archive AnyScript Yahoo archive
1

Hi to everyone! my name is Peter and I have recently started a PhD
looking into post knee arthroplasty biomechanics. I aim to use Vicom
motion analysis and force feedback data (C3D data) to build a
musculoskeltal model in Anybody. Is there any guidance on marker
positions for gait/stairs/sit-stand to get the best results when
converting the data to Anybody?
I am also going over the idea of testing static and dynamic balance,
with this there will be sensorimotor information, is it possible to
add in this neural input element into model?

I look forward to hearing back from you

Many thanks

Peter Worsley

2

Hi Peter

Welcome to the group

It sounds as an interesting problem, which should be possible to do.

In principle there is no strict requirements on how to locate the
markers. When you do the AnyBody model you will have to create the
same markers in the model, so it is ofcourse a good idea to make use
of bonylandmarks, this makes it easier to place the markers in the
model. From experience we have seen though that it is a good idea to
have multiple markers on the foot. So for example markers placed
both medially and laterally on the forefoot together with a heel
maker. Sometimes it can be difficult to get the inversion eversion
controlled. You can also have a look at the Gait3D model for an idea
on a suitable markerset, this model is however missing the extra
markers on the forefoot.

I am not sure i fully understand your ideas about the balancing and
the sensorimotor input. What is it that you can measure?. AnyBody is
based on inverse dynamics so the motion is the input. You can
however also create models which has an extra optimization loop
around, this will let you optimize the motion to meet a certain
objective.

I hope this made things more clear, otherwise please ask again.

Best regards
Søren, AnyBody Support

— In anyscript@yahoogroups.com, “peter.worsley”
<peter.worsley@…> wrote:
>
> Hi to everyone! my name is Peter and I have recently started a PhD
> looking into post knee arthroplasty biomechanics. I aim to use
Vicom
> motion analysis and force feedback data (C3D data) to build a
> musculoskeltal model in Anybody. Is there any guidance on marker
> positions for gait/stairs/sit-stand to get the best results when
> converting the data to Anybody?
> I am also going over the idea of testing static and dynamic
balance,
> with this there will be sensorimotor information, is it possible to
> add in this neural input element into model?
>
> I look forward to hearing back from you
>
> Many thanks
>
> Peter Worsley
>

3

— In anyscript@yahoogroups.com, “AnyBody Support” <support@…> wrote:
>Thanks for the advice on marker positioning soren,

Don’t worry about the sensorimotor input, i have talked with my PhD
supervisor and this is not going to be included in the model. I do
however have a few other questions for you;

As my project is looking into the specifically knee joint, i would
like to input some data into joint laxity. I am right in thinking that
the knee is currently modelled as a striaght hinge joint? If so could
i put some ant/post translation and valgus/varus parameters into the
dynamics of the knee during gait for instance?

Also regarding the foot and ankle; is the foot a rigid body or have
you modelled the joints within the foot? to what extent is the ankle
modelled within the joint with regards to inversion, eversion and
pronation?

I am recieving data from the software guided surgery for the exact
location of the prostheses. Is it possible for this to be inputted
into the anybody models?

Finally, alot of the current literature identifies co-contractions
within quads/hams during gait for knee arthroplasty patients. I
realise the model is restricted by the maths (6 equations)and that the
model is based apon an optimisation of muscle action for a given
movement. What happens when i start inputting asymetrical non-optimal
data into the model, can it still be processed?

Sorry for the bombardment

Many thanks

Peter Worsley

> Hi Peter
>
> Welcome to the group
>
> It sounds as an interesting problem, which should be possible to do.
>
> In principle there is no strict requirements on how to locate the
> markers. When you do the AnyBody model you will have to create the
> same markers in the model, so it is ofcourse a good idea to make use
> of bonylandmarks, this makes it easier to place the markers in the
> model. From experience we have seen though that it is a good idea to
> have multiple markers on the foot. So for example markers placed
> both medially and laterally on the forefoot together with a heel
> maker. Sometimes it can be difficult to get the inversion eversion
> controlled. You can also have a look at the Gait3D model for an idea
> on a suitable markerset, this model is however missing the extra
> markers on the forefoot.
>
> I am not sure i fully understand your ideas about the balancing and
> the sensorimotor input. What is it that you can measure?. AnyBody is
> based on inverse dynamics so the motion is the input. You can
> however also create models which has an extra optimization loop
> around, this will let you optimize the motion to meet a certain
> objective.
>
> I hope this made things more clear, otherwise please ask again.
>
> Best regards
> Søren, AnyBody Support
>
>
>
>
> — In anyscript@yahoogroups.com, “peter.worsley”
> <peter.worsley@> wrote:
> >
> > Hi to everyone! my name is Peter and I have recently started a PhD
> > looking into post knee arthroplasty biomechanics. I aim to use
> Vicom
> > motion analysis and force feedback data (C3D data) to build a
> > musculoskeltal model in Anybody. Is there any guidance on marker
> > positions for gait/stairs/sit-stand to get the best results when
> > converting the data to Anybody?
> > I am also going over the idea of testing static and dynamic
> balance,
> > with this there will be sensorimotor information, is it possible to
> > add in this neural input element into model?
> >
> > I look forward to hearing back from you
> >
> > Many thanks
> >
> > Peter Worsley
> >
>

4

Hi Peter

Yes you are right about the knee joint it is currently made as a hinge
joint.

If you would like to add ant/post translation and valgus/varus parameters to
the model the knee joint will obviously need to be changed from being a
hinge joint

That said it is possible to implement more complex knee model, please see
this link which is about a knee model developed in AnyScript

http://www.anybody.aau.dk/pdf/EshuisKnee.pdf

The basic kinematic measures in AnyBody AnyKinLinear and AnyKinRotational
can be combined in any way you like to represent different joint types. So
if you have specific ideas on how the joint should work it might be possible
to create your own tailor made joint.

The foot is created as a stiff segment and the ankle joint is created as a
universal joint.

If you have geometric data from prostheses this can be include as a .stl CAD
file. This will allow you to visually see the prostheses it will however not
automatically give you a joint behaviour according to the prostheses
geometry, this is to be controlled by a combination of AnyScript measures.

I am not sure I fully understand the last question, if you are running a
model with different knee joints, the differences in the constraints will be
reflected in the muscle recruitment. Most likely the muscle recruitment
algorithm will be able to balance the model, but it will naturally depend on
the type of constraints used in the joints and the muscle configuration.

I hope this answered your questions otherwise please ask again

Best regards

Søren, AnyBody support

From: anyscript@yahoogroups.com [mailto:anyscript@yahoogroups.com] On Behalf
Of peter.worsley
Sent: 26 October 2007 14:58
To: anyscript@yahoogroups.com
Subject: [AnyScript] Re: Vicom markers and static balance

— In anyscript@yahoogrou <mailto:anyscript%40yahoogroups.com> ps.com,
“AnyBody Support” <support@…> wrote:
>Thanks for the advice on marker positioning soren,

Don’t worry about the sensorimotor input, i have talked with my PhD
supervisor and this is not going to be included in the model. I do
however have a few other questions for you;

As my project is looking into the specifically knee joint, i would
like to input some data into joint laxity. I am right in thinking that
the knee is currently modelled as a striaght hinge joint? If so could
i put some ant/post translation and valgus/varus parameters into the
dynamics of the knee during gait for instance?

Also regarding the foot and ankle; is the foot a rigid body or have
you modelled the joints within the foot? to what extent is the ankle
modelled within the joint with regards to inversion, eversion and
pronation?

I am recieving data from the software guided surgery for the exact
location of the prostheses. Is it possible for this to be inputted
into the anybody models?

Finally, alot of the current literature identifies co-contractions
within quads/hams during gait for knee arthroplasty patients. I
realise the model is restricted by the maths (6 equations)and that the
model is based apon an optimisation of muscle action for a given
movement. What happens when i start inputting asymetrical non-optimal
data into the model, can it still be processed?

Sorry for the bombardment

Many thanks

Peter Worsley

> Hi Peter
>
> Welcome to the group
>
> It sounds as an interesting problem, which should be possible to do.
>
> In principle there is no strict requirements on how to locate the
> markers. When you do the AnyBody model you will have to create the
> same markers in the model, so it is ofcourse a good idea to make use
> of bonylandmarks, this makes it easier to place the markers in the
> model. From experience we have seen though that it is a good idea to
> have multiple markers on the foot. So for example markers placed
> both medially and laterally on the forefoot together with a heel
> maker. Sometimes it can be difficult to get the inversion eversion
> controlled. You can also have a look at the Gait3D model for an idea
> on a suitable markerset, this model is however missing the extra
> markers on the forefoot.
>
> I am not sure i fully understand your ideas about the balancing and
> the sensorimotor input. What is it that you can measure?. AnyBody is
> based on inverse dynamics so the motion is the input. You can
> however also create models which has an extra optimization loop
> around, this will let you optimize the motion to meet a certain
> objective.
>
> I hope this made things more clear, otherwise please ask again.
>
> Best regards
> Søren, AnyBody Support
>
>
>
>
> — In anyscript@yahoogrou <mailto:anyscript%40yahoogroups.com> ps.com,
“peter.worsley”
> <peter.worsley@> wrote:
> >
> > Hi to everyone! my name is Peter and I have recently started a PhD
> > looking into post knee arthroplasty biomechanics. I aim to use
> Vicom
> > motion analysis and force feedback data (C3D data) to build a
> > musculoskeltal model in Anybody. Is there any guidance on marker
> > positions for gait/stairs/sit-stand to get the best results when
> > converting the data to Anybody?
> > I am also going over the idea of testing static and dynamic
> balance,
> > with this there will be sensorimotor information, is it possible to
> > add in this neural input element into model?
> >
> > I look forward to hearing back from you
> >
> > Many thanks
> >
> > Peter Worsley
> >
>

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

5

— In anyscript@yahoogroups.com, “AnyBody Support” <support@…> wrote:
Hi Soren, thank you for the knee model, it certainly is a good start
on upgrading the current hinge joint. If i was going to develop a
‘tailor made’ knee joint for each of my sample going through motion
analysis and musculoskeletal modelling, I would need accurate force
feedback from the foot. At the moment the foot is modelled as a stiff
segment, has there been any previous work done to segment the foot
into fore/mid/hind foot areas? I feel this would be very beneficial
when modelling stairs for instance.

Also I understand the current optimisation methods are based on your
min/max theory with the option of squaring the figures. Will there be
any other optimisation techniques available? for example energy
expenditure. (this might make it easier to input the EMG data i am
collecting)

Thank you for your help and support

Peter

> Hi Peter
>
>
>
> Yes you are right about the knee joint it is currently made as a hinge
> joint.
>
>
>
> If you would like to add ant/post translation and valgus/varus
parameters to
> the model the knee joint will obviously need to be changed from being a
> hinge joint
>
>
>
> That said it is possible to implement more complex knee model,
please see
> this link which is about a knee model developed in AnyScript
>
>
>
> http://www.anybody.aau.dk/pdf/EshuisKnee.pdf
>
>
>
> The basic kinematic measures in AnyBody AnyKinLinear and
AnyKinRotational
> can be combined in any way you like to represent different joint
types. So
> if you have specific ideas on how the joint should work it might be
possible
> to create your own tailor made joint.
>
>
>
> The foot is created as a stiff segment and the ankle joint is
created as a
> universal joint.
>
>
>
> If you have geometric data from prostheses this can be include as a
.stl CAD
> file. This will allow you to visually see the prostheses it will
however not
> automatically give you a joint behaviour according to the prostheses
> geometry, this is to be controlled by a combination of AnyScript
measures.
>
>
>
> I am not sure I fully understand the last question, if you are running a
> model with different knee joints, the differences in the constraints
will be
> reflected in the muscle recruitment. Most likely the muscle recruitment
> algorithm will be able to balance the model, but it will naturally
depend on
> the type of constraints used in the joints and the muscle
configuration.
>
>
>
> I hope this answered your questions otherwise please ask again
>
>
>
> Best regards
>
> Søren, AnyBody support
>
>
>
>
>
>
>
> _____
>
> From: anyscript@yahoogroups.com [mailto:anyscript@yahoogroups.com]
On Behalf
> Of peter.worsley
> Sent: 26 October 2007 14:58
> To: anyscript@yahoogroups.com
> Subject: [AnyScript] Re: Vicom markers and static balance
>
>
>
> — In anyscript@yahoogrou <mailto:anyscript%40yahoogroups.com> ps.com,
> “AnyBody Support” <support@> wrote:
> >Thanks for the advice on marker positioning soren,
>
> Don’t worry about the sensorimotor input, i have talked with my PhD
> supervisor and this is not going to be included in the model. I do
> however have a few other questions for you;
>
> As my project is looking into the specifically knee joint, i would
> like to input some data into joint laxity. I am right in thinking that
> the knee is currently modelled as a striaght hinge joint? If so could
> i put some ant/post translation and valgus/varus parameters into the
> dynamics of the knee during gait for instance?
>
> Also regarding the foot and ankle; is the foot a rigid body or have
> you modelled the joints within the foot? to what extent is the ankle
> modelled within the joint with regards to inversion, eversion and
> pronation?
>
> I am recieving data from the software guided surgery for the exact
> location of the prostheses. Is it possible for this to be inputted
> into the anybody models?
>
> Finally, alot of the current literature identifies co-contractions
> within quads/hams during gait for knee arthroplasty patients. I
> realise the model is restricted by the maths (6 equations)and that the
> model is based apon an optimisation of muscle action for a given
> movement. What happens when i start inputting asymetrical non-optimal
> data into the model, can it still be processed?
>
> Sorry for the bombardment
>
> Many thanks
>
> Peter Worsley
>
> > Hi Peter
> >
> > Welcome to the group
> >
> > It sounds as an interesting problem, which should be possible to do.
> >
> > In principle there is no strict requirements on how to locate the
> > markers. When you do the AnyBody model you will have to create the
> > same markers in the model, so it is ofcourse a good idea to make use
> > of bonylandmarks, this makes it easier to place the markers in the
> > model. From experience we have seen though that it is a good idea to
> > have multiple markers on the foot. So for example markers placed
> > both medially and laterally on the forefoot together with a heel
> > maker. Sometimes it can be difficult to get the inversion eversion
> > controlled. You can also have a look at the Gait3D model for an idea
> > on a suitable markerset, this model is however missing the extra
> > markers on the forefoot.
> >
> > I am not sure i fully understand your ideas about the balancing and
> > the sensorimotor input. What is it that you can measure?. AnyBody is
> > based on inverse dynamics so the motion is the input. You can
> > however also create models which has an extra optimization loop
> > around, this will let you optimize the motion to meet a certain
> > objective.
> >
> > I hope this made things more clear, otherwise please ask again.
> >
> > Best regards
> > Søren, AnyBody Support
> >
> >
> >
> >
> > — In anyscript@yahoogrou <mailto:anyscript%40yahoogroups.com>
ps.com,
> “peter.worsley”
> > <peter.worsley@> wrote:
> > >
> > > Hi to everyone! my name is Peter and I have recently started a PhD
> > > looking into post knee arthroplasty biomechanics. I aim to use
> > Vicom
> > > motion analysis and force feedback data (C3D data) to build a
> > > musculoskeltal model in Anybody. Is there any guidance on marker
> > > positions for gait/stairs/sit-stand to get the best results when
> > > converting the data to Anybody?
> > > I am also going over the idea of testing static and dynamic
> > balance,
> > > with this there will be sensorimotor information, is it possible to
> > > add in this neural input element into model?
> > >
> > > I look forward to hearing back from you
> > >
> > > Many thanks
> > >
> > > Peter Worsley
> > >
> >
>
>
>
>
>
> [Non-text portions of this message have been removed]
>

6

Hi Peter

You are right about the foot model; this is currently a stiff segment.

There have been some previous work on the foot please see

http://www.anybodytech.com/116.0.html?
<http://www.anybodytech.com/116.0.html?&tx_ttnews%5btt_news%5d=21&tx_ttnews%
5bbackPid%5d=37&cHash=a4a430913d>
&tx_ttnews[tt_news]=21&tx_ttnews[backPid]=37&cHash=a4a430913d

There is some ongoing work on foot models, but foot models tend to be very
complex, so it may take a while until it will be available. For now I think
you will have to try to do without it.

We hope to make more criteria’s available in the future; in principle we
would like to have a wide range of criteria’s, so that the user can pick the
one he thinks suits his needs. Please notice that you can do something
already with the current methods. By applying a high quadratic penalty this
will be equivalent to quadratic criteria, there is a tutorial about this,
please see

http://www.anybodytech.com/691.0.html

It shows the effect of the different parameters of the AnyBodyStudy

I do not fully understand the comment about using the EMG data as input.
AnyBody is inverse dynamics so it will not be possible to drive the model
with recorded EMG data, but maybe this is not what you meant?

Please ask again if you have further questions.

Best regards

Søren

From: anyscript@yahoogroups.com [mailto:anyscript@yahoogroups.com] On Behalf
Of peter.worsley
Sent: 05 November 2007 13:00
To: anyscript@yahoogroups.com
Subject: [AnyScript] Re: Vicom markers and static balance

— In anyscript@yahoogrou <mailto:anyscript%40yahoogroups.com> ps.com,
“AnyBody Support” <support@…> wrote:
Hi Soren, thank you for the knee model, it certainly is a good start
on upgrading the current hinge joint. If i was going to develop a
‘tailor made’ knee joint for each of my sample going through motion
analysis and musculoskeletal modelling, I would need accurate force
feedback from the foot. At the moment the foot is modelled as a stiff
segment, has there been any previous work done to segment the foot
into fore/mid/hind foot areas? I feel this would be very beneficial
when modelling stairs for instance.

Also I understand the current optimisation methods are based on your
min/max theory with the option of squaring the figures. Will there be
any other optimisation techniques available? for example energy
expenditure. (this might make it easier to input the EMG data i am
collecting)

Thank you for your help and support

Peter

> Hi Peter
>
>
>
> Yes you are right about the knee joint it is currently made as a hinge
> joint.
>
>
>
> If you would like to add ant/post translation and valgus/varus
parameters to
> the model the knee joint will obviously need to be changed from being a
> hinge joint
>
>
>
> That said it is possible to implement more complex knee model,
please see
> this link which is about a knee model developed in AnyScript
>
>
>
> http://www.anybody. <http://www.anybody.aau.dk/pdf/EshuisKnee.pdf>
aau.dk/pdf/EshuisKnee.pdf
>
>
>
> The basic kinematic measures in AnyBody AnyKinLinear and
AnyKinRotational
> can be combined in any way you like to represent different joint
types. So
> if you have specific ideas on how the joint should work it might be
possible
> to create your own tailor made joint.
>
>
>
> The foot is created as a stiff segment and the ankle joint is
created as a
> universal joint.
>
>
>
> If you have geometric data from prostheses this can be include as a
.stl CAD
> file. This will allow you to visually see the prostheses it will
however not
> automatically give you a joint behaviour according to the prostheses
> geometry, this is to be controlled by a combination of AnyScript
measures.
>
>
>
> I am not sure I fully understand the last question, if you are running a
> model with different knee joints, the differences in the constraints
will be
> reflected in the muscle recruitment. Most likely the muscle recruitment
> algorithm will be able to balance the model, but it will naturally
depend on
> the type of constraints used in the joints and the muscle
configuration.
>
>
>
> I hope this answered your questions otherwise please ask again
>
>
>
> Best regards
>
> Søren, AnyBody support
>
>
>
>
>
>
>
> _____
>
> From: anyscript@yahoogrou <mailto:anyscript%40yahoogroups.com> ps.com
[mailto:anyscript@yahoogrou <mailto:anyscript%40yahoogroups.com> ps.com]
On Behalf
> Of peter.worsley
> Sent: 26 October 2007 14:58
> To: anyscript@yahoogrou <mailto:anyscript%40yahoogroups.com> ps.com
> Subject: [AnyScript] Re: Vicom markers and static balance
>
>
>
> — In anyscript@yahoogrou <mailto:anyscript%40yahoogroups.com> ps.com,
> “AnyBody Support” <support@> wrote:
> >Thanks for the advice on marker positioning soren,
>
> Don’t worry about the sensorimotor input, i have talked with my PhD
> supervisor and this is not going to be included in the model. I do
> however have a few other questions for you;
>
> As my project is looking into the specifically knee joint, i would
> like to input some data into joint laxity. I am right in thinking that
> the knee is currently modelled as a striaght hinge joint? If so could
> i put some ant/post translation and valgus/varus parameters into the
> dynamics of the knee during gait for instance?
>
> Also regarding the foot and ankle; is the foot a rigid body or have
> you modelled the joints within the foot? to what extent is the ankle
> modelled within the joint with regards to inversion, eversion and
> pronation?
>
> I am recieving data from the software guided surgery for the exact
> location of the prostheses. Is it possible for this to be inputted
> into the anybody models?
>
> Finally, alot of the current literature identifies co-contractions
> within quads/hams during gait for knee arthroplasty patients. I
> realise the model is restricted by the maths (6 equations)and that the
> model is based apon an optimisation of muscle action for a given
> movement. What happens when i start inputting asymetrical non-optimal
> data into the model, can it still be processed?
>
> Sorry for the bombardment
>
> Many thanks
>
> Peter Worsley
>
> > Hi Peter
> >
> > Welcome to the group
> >
> > It sounds as an interesting problem, which should be possible to do.
> >
> > In principle there is no strict requirements on how to locate the
> > markers. When you do the AnyBody model you will have to create the
> > same markers in the model, so it is ofcourse a good idea to make use
> > of bonylandmarks, this makes it easier to place the markers in the
> > model. From experience we have seen though that it is a good idea to
> > have multiple markers on the foot. So for example markers placed
> > both medially and laterally on the forefoot together with a heel
> > maker. Sometimes it can be difficult to get the inversion eversion
> > controlled. You can also have a look at the Gait3D model for an idea
> > on a suitable markerset, this model is however missing the extra
> > markers on the forefoot.
> >
> > I am not sure i fully understand your ideas about the balancing and
> > the sensorimotor input. What is it that you can measure?. AnyBody is
> > based on inverse dynamics so the motion is the input. You can
> > however also create models which has an extra optimization loop
> > around, this will let you optimize the motion to meet a certain
> > objective.
> >
> > I hope this made things more clear, otherwise please ask again.
> >
> > Best regards
> > Søren, AnyBody Support
> >
> >
> >
> >
> > — In anyscript@yahoogrou <mailto:anyscript%40yahoogroups.com>
ps.com,
> “peter.worsley”
> > <peter.worsley@> wrote:
> > >
> > > Hi to everyone! my name is Peter and I have recently started a PhD
> > > looking into post knee arthroplasty biomechanics. I aim to use
> > Vicom
> > > motion analysis and force feedback data (C3D data) to build a
> > > musculoskeltal model in Anybody. Is there any guidance on marker
> > > positions for gait/stairs/sit-stand to get the best results when
> > > converting the data to Anybody?
> > > I am also going over the idea of testing static and dynamic
> > balance,
> > > with this there will be sensorimotor information, is it possible to
> > > add in this neural input element into model?
> > >
> > > I look forward to hearing back from you
> > >
> > > Many thanks
> > >
> > > Peter Worsley
> > >
> >
>
>
>
>
>
> [Non-text portions of this message have been removed]
>

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