View Full Version : Summary: Visible Human data as a standard for biomechanicalmodels?

07-19-2002, 07:58 AM
Thanks to everyone who responded to my posting about the suitability of
the Visible Human data as a standard model for biomechanical studies.
Many interesting points were raised, such as the source of the data
(the ethics of using data from an executed convict), the need for 50th
percentile data in safety studies, and the possibility of scaling
Visible Human data to fit a 50th percentile model.

The original posting is below, followed by the responses.

Original posting:


Although many of us involved in modeling of human biomechanics often
develop models of the 50th percentile male or female, each research
group's model is slightly different. The Visible Human Project has made
it possible to develop musculoskeletal models (e.g., [1]) from a single
set of image data. If models are created from these data, it seems to me
that it would facilitate the creation of standard human biomechanical

However, one potential problem with the Visible Male dataset is its
size. ( I haven't evaluated the Visible Female dataset yet). A
comparison with the 1988 US Army Anthropometric Survey [2], showed that
the Visible Male is 3% taller and 15% heavier than the 50th percentile
male. I personally am interested in the neck musculoskeletal system, and
found that certain external head and neck dimensions were 6-30% greater
than the 50th percentile male. Garner and Pandy measured arm muscle
volumes from Visible Male images that were more than 3 times greater
than in anatomical studies (which were primarily done in cadavers). Even
considering the difference between live human and cadaver muscles, the
Visible Male seems to be an extremely big, muscular guy. This would
likely mean larger estimated magnitudes of muscle forces and moment arms
compared to a 50th percentile male.

Despite these size issues, I still see value in developing models from
Visible Human data, for elucidating general principles of biomechanics
and motor control. While I'm not advocating use of the Visible Human
data as the *only* standard for biomechanical models, I would like to
hear people's opinions on its use. Is it reasonable to use the data as a
representation of the average male or female, or is size a problem? Do
people anticipate that its use will facilitate sharing of data? Do
people have any other comments or concerns about using the Visible Human
dataset for biomechanical studies? As always, I will post a summary of

Anita Vasavada

[1] Garner, B, and Pandy, M (2001). Computer Methods in Biomechanics and
Biomedical Engineering 4: 93-126.
[2] Gordon, CC, et al. (1989). 1988 anthropometric survey of U.S. army
personnel: methods and summary statistics.


From Arno Grunendahl (arno@baustatik.rwth-aachen.de)

Hi Anita,

I'm doing FE-modelling of human lumbar spine.
As it was not possible for me to derive a FE-model from own CT data, I was
very interested in the Visible Human Project. To my disappointment I found
that data (pictures) can only be exported in standart picture formats (tif,
jpeg, eps etc.) but these formats are not supported by FE programs.
So I gave up Visible Human Project.
Maybe there is a way to transform Visible Human Data, that I don't know.
Silly me.....



>From Mark Thompson (mark.thompson@ort.lu.se)

Hi Anita

Not a biomechanically related observation at all, but I feel that some
people may be unhappy with working with data obtained from executed
criminals. I suppose that practically, since the data exist, they ought to
be used, but the question of the ethics of how one obtains detailed
anatomical measurements ought to be part of the general discussion of which
sources of those data one ought to use.

best wishes


__________________________________________________ ____

Mark Thompson, PhD
Biomechanics Laboratory
Department of Orthopaedics
Lund University Hospital
S-22185 Lund

>From Gary Christopher (garyalan59@mstar2.net)


I recently (last Fall) posed this question to my major professor, and while
our discussion didn't really go anywhere, I think it would be beneficial to
develop models, including mass-inertial characteristics, from the Visible
Human Project data set. Any differences between the Visible Human
characteristics could be scaled to (more closely) "match" the 50th
percentile male and/or female. I believe this to be of great potential for
the biomechanics community, and development of models based on these data
sets should be pursued diligently (perhaps I have stumbled across a
potential dissertation topic!).

Gary Christopher

>From A.J. ChandraSekhar (chandra@axiomconsult.com)

Dear Anita,

I come from engineering background, we work with 50thPercentile dummies

for occupant safety studies. Its a standard practice to do so , as validated

results are also available for these. Does safety study fall into what you are

terming as biomechanical study ? If yes it should get acceptance in the

automotive safety world before it becomes a everyday practice , with a

good reason to do so.

I don't know whether i made relevant sense,as I come from a different

background, but subscribe to this group as it helps understand the

full perspective .



Axiom Consulting.

>From Dave Hawkins (dahawkins@ucdavis.edu)

I believethat the Visible Human project has provided us with valuable information
that may be used to develop musculoskeletal models. Based on data
from this project we developed a digital lower limb musculoskeletal model
(1), computational techniques to alter the size of the musculature within
this model (to address similar concerns to the size concerns that you
expressed) (2), and are currently using this model to address a variety
of questions about the interactions between limb morphology and movement
dynamics. Hope this helps

1. Barr, A., and Hawkins, D. An anatomical database providing
three-dimensional geometric representation of lower limb
structures. Journal of Applied Biomechanics , 2000, 16:301-308.
2. Hawkins, D , and Barr, A. A computational approach for
simulating muscle morphologic changes in musculoskeletal modeling.
Computer Methods in Biomechanics and Biomedical Engineering, 2001,

Anita Vasavada, Ph.D.
Assistant Professor
Departments of Biological Systems Engineering and VCAPP
Programs in Bioengineering and Neuroscience
Washington State University
Pullman, WA 99164-6120
voice: (509) 335-7533
fax: (509) 335-4650

To unsubscribe send SIGNOFF BIOMCH-L to LISTSERV@nic.surfnet.nl
For information and archives: http://isb.ri.ccf.org/biomch-l