Sergio Fonseca

10-07-1996, 02:29 AM

I want to thank all of you that answered my question. Below follows my

original question and a summary of responses I received.

Sergio Fonseca (fonserte@bu.edu)

Dept. of Physical Therapy

Boston University

Original Posting:

----------------------------------------------------------------------------

--------------

I am a physical therapist and a doctoral candidate doing my research work

in locomotion in children with cerebral palsy. I would like to know about

the availability of a software for 3D calculation of body center of mass

during walking that does not assume side symmetry. We are using a two

position sensor Optotrak system to generate 3D kinematic data from both

sides. Program codes (basic or C) in which the anthropometric variables can

be changed to accommodate different age groups (ex: Dampster or Jensen) are

preferable.

Replies:

----------------------------------------------------------------------------

-------------------------

Sergio,

I am working on a project for the Air Force to validate some software

which calculates the whole body center of mass and moments of

inertia. It allows the user to input anthropometric data and also

has an option for children where all that you input is age, weight,

and stature. There is one piece of software called GEBOD (GEnerator

of BODy data) which generates segment information and another called

ATB (Articulated Total Body model). And, the other piece of good

news is that this software is in the public domain. I don't really

know of any other software that determines all of this information in

3D.

BUT, one of the problems that we're finding with the software is that

it assumes symmetry. However, as I write this, I'm thinking that

GEBOD could be run twice--once for each side of the body, and you

would have to use individual anthropometric measurements for this to

work--then ATB could be modified so that it could take input from

both sides of the body to calculate 3D CG or center of mass. The

code is written in FORTRAN, but the listing would be available to

you.

If no one else suggests software that does not have a symmetry

assumption, I would say that this would be painful and tedious but

doable.

Beth Todd

Dr. Beth A. Todd

Assistant Professor

Mechanical Engineering

Box 870276

University of Alabama

Tuscaloosa, AL 35487-0276

btodd@coe.eng.ua.edu

(205)348-1623

fax: (205)348-7240

----------------------------------------------------------------------------

---------------------------

From: Gideon Ariel

Check the web site at: http://www.arielnet.com

----------------------------------------------------------------------------

--------------------------

I had some thoughts on your problem, which you posted on BiomechL. If you

use the method of Yeadon and Morlock for estimating limb segment inertial

properties, (J Biomech, vol 22, p 683-9, 1989) and also the method of

Dempster for estimating CoM of each limb, given by Winter, in

Biomechanics and Motor Control of Human Movement 1990, it is possible to

easily derive a kinetic model which is not dependent upon symmetry

assumptions, however you need to make anthroprometric measurements on each

subject. If you are interested I have some Matlab code for implementing

the method of Yeadon and Morlock, but this problem is fairly simple. The

real problem lies in a good algorithms to determine key positions of

joint centres which allow for pathology, age, sex, deformity, etc. It

really depends upon which part of the gait cycle you are interested in;

if it is the stance phase then the joint centre positions are of great

interest, else for swing the movement of CoM and the limb inertial

properties are of greater interest. Although some work has been done to

determine estimates of the instant joint centres of normal joints I know

of little work that has attempted to determine the same thing for

pathologic joints, perhaps for good reason. It is difficult to establish

an absolute pattern of movement for a so-called "normal" joint. The

problem becomes so much more intractable for any joint displaying a

degree of pathology or deformity.

Richie (H.S.) Gill

Oxford Orthopaedic Engineering Centre

University of Oxford

Nuffield Orthopaedic Centre

Headington

Oxford

OX3 7LD

email: richie.gill@ox.ac.uk

tel : +44 1865 227541

fax : +44 1865 742348

----------------------------------------------------------------------------

----------------------------

Hello Sergio,

Sorry, but I can not help you in the way you want, but I am sure

that it is not difficult to write a program to calculate it.

In engineering terms, it is a simple task to calculate the CM if

you have the spatial position of the limbs.

Anyway, I am PhD student also.

Carlo Pece Phone: +44 (0)1784 431341 ext:244

PhD Researcher Fax: +44 (0)1784 472879

Brunel University

Department of Design E-mail: Carlo.Pece@brunel.ac.uk

Egham

Surrey TW20 0JZ

UK

----------------------------------------------------------------------------

----------------------------

Hi there!

U may want to get in touch with Prof. C.L Vaughan at the University of Cape

Town. His Email Address is KVAUGHAN@ANAT.UCT.AC.ZA

hope this helps,

Cheers,

Noddy

\@/

@@@ \\|//

@@@@@ \\\|///

( @ @ ) ( O O )

____oOOo______^_____oOOo_____oOOo______O______oOOo ____

| |

| Nivedita Subramanian |

| Sackler Institute of N |

| Biomedical Sciences O |

| 550, 1st Ave, New York, D |

| NY 10016, USA D |

| Tel. +91 (212) 696 4693 Y |

| SUBRAN01@popmail.med.nyu.edu |

|_________________________________________________ ____|

original question and a summary of responses I received.

Sergio Fonseca (fonserte@bu.edu)

Dept. of Physical Therapy

Boston University

Original Posting:

----------------------------------------------------------------------------

--------------

I am a physical therapist and a doctoral candidate doing my research work

in locomotion in children with cerebral palsy. I would like to know about

the availability of a software for 3D calculation of body center of mass

during walking that does not assume side symmetry. We are using a two

position sensor Optotrak system to generate 3D kinematic data from both

sides. Program codes (basic or C) in which the anthropometric variables can

be changed to accommodate different age groups (ex: Dampster or Jensen) are

preferable.

Replies:

----------------------------------------------------------------------------

-------------------------

Sergio,

I am working on a project for the Air Force to validate some software

which calculates the whole body center of mass and moments of

inertia. It allows the user to input anthropometric data and also

has an option for children where all that you input is age, weight,

and stature. There is one piece of software called GEBOD (GEnerator

of BODy data) which generates segment information and another called

ATB (Articulated Total Body model). And, the other piece of good

news is that this software is in the public domain. I don't really

know of any other software that determines all of this information in

3D.

BUT, one of the problems that we're finding with the software is that

it assumes symmetry. However, as I write this, I'm thinking that

GEBOD could be run twice--once for each side of the body, and you

would have to use individual anthropometric measurements for this to

work--then ATB could be modified so that it could take input from

both sides of the body to calculate 3D CG or center of mass. The

code is written in FORTRAN, but the listing would be available to

you.

If no one else suggests software that does not have a symmetry

assumption, I would say that this would be painful and tedious but

doable.

Beth Todd

Dr. Beth A. Todd

Assistant Professor

Mechanical Engineering

Box 870276

University of Alabama

Tuscaloosa, AL 35487-0276

btodd@coe.eng.ua.edu

(205)348-1623

fax: (205)348-7240

----------------------------------------------------------------------------

---------------------------

From: Gideon Ariel

Check the web site at: http://www.arielnet.com

----------------------------------------------------------------------------

--------------------------

I had some thoughts on your problem, which you posted on BiomechL. If you

use the method of Yeadon and Morlock for estimating limb segment inertial

properties, (J Biomech, vol 22, p 683-9, 1989) and also the method of

Dempster for estimating CoM of each limb, given by Winter, in

Biomechanics and Motor Control of Human Movement 1990, it is possible to

easily derive a kinetic model which is not dependent upon symmetry

assumptions, however you need to make anthroprometric measurements on each

subject. If you are interested I have some Matlab code for implementing

the method of Yeadon and Morlock, but this problem is fairly simple. The

real problem lies in a good algorithms to determine key positions of

joint centres which allow for pathology, age, sex, deformity, etc. It

really depends upon which part of the gait cycle you are interested in;

if it is the stance phase then the joint centre positions are of great

interest, else for swing the movement of CoM and the limb inertial

properties are of greater interest. Although some work has been done to

determine estimates of the instant joint centres of normal joints I know

of little work that has attempted to determine the same thing for

pathologic joints, perhaps for good reason. It is difficult to establish

an absolute pattern of movement for a so-called "normal" joint. The

problem becomes so much more intractable for any joint displaying a

degree of pathology or deformity.

Richie (H.S.) Gill

Oxford Orthopaedic Engineering Centre

University of Oxford

Nuffield Orthopaedic Centre

Headington

Oxford

OX3 7LD

email: richie.gill@ox.ac.uk

tel : +44 1865 227541

fax : +44 1865 742348

----------------------------------------------------------------------------

----------------------------

Hello Sergio,

Sorry, but I can not help you in the way you want, but I am sure

that it is not difficult to write a program to calculate it.

In engineering terms, it is a simple task to calculate the CM if

you have the spatial position of the limbs.

Anyway, I am PhD student also.

Carlo Pece Phone: +44 (0)1784 431341 ext:244

PhD Researcher Fax: +44 (0)1784 472879

Brunel University

Department of Design E-mail: Carlo.Pece@brunel.ac.uk

Egham

Surrey TW20 0JZ

UK

----------------------------------------------------------------------------

----------------------------

Hi there!

U may want to get in touch with Prof. C.L Vaughan at the University of Cape

Town. His Email Address is KVAUGHAN@ANAT.UCT.AC.ZA

hope this helps,

Cheers,

Noddy

\@/

@@@ \\|//

@@@@@ \\\|///

( @ @ ) ( O O )

____oOOo______^_____oOOo_____oOOo______O______oOOo ____

| |

| Nivedita Subramanian |

| Sackler Institute of N |

| Biomedical Sciences O |

| 550, 1st Ave, New York, D |

| NY 10016, USA D |

| Tel. +91 (212) 696 4693 Y |

| SUBRAN01@popmail.med.nyu.edu |

|_________________________________________________ ____|