PDA

View Full Version : motion analysis of hands



N.shaw-hamilton
11-22-1994, 11:32 PM
Dear all,

Many thanks to everyone who wrote in answer to my question on motion analysis
of hands/fingers. Here's a list of replies for anyone that's interested.

************************************************** *****************

Hello,

> I'm trying to set up a motion analysis system for looking at finger/hand
> motion. Has anyone tried this? I'm having problems trying to establish a
> marker system for such a small scale.

One of my colleagues at Queen's University in Ontario had good results using an
Opti-trak system to measure motion of the hand for a study of wrist kinematics
(I was one of his guinea pigs). His name is Ian Wright. He is still there
until the end of this term, I believe, but I'm not sure how often he checks his
e-mail address. You can try to reach him at wright@me.queensu.ca, or you might
want to get in touch with his past supervisor, Dr. Carolyn Small, whose e-mail
address should be in the biomch-l list (probably small@me.queensu.ca).

Hope this helps.
Stephen Ferguson ferguson@ccgate.ari.ch
ASIF Research Institute 41 81 44 22 88
Davos, Switzerland 41 81 44 22 88 (fax)


************************************************** ****************

Nicky -

I seem to recall seeing a setup to look at hand motion at the Mayo Clinic
a couple of years ago - I was there for a Gait Conference, but when we
toured the lab, someone was using a Polhemous (sp?) system to look at
hand motion. I don't recall who, but Dwight MEglan was showing us
around, so he might know.

Good luck

Marc

-------------------------------------------------------------------------------
Marcus P. Besser, PhD []
Assistant Director, Human Performance Laboratory _ /_/
Assistant Professsor, Department of Physical Therapy / /
Thomas Jefferson University / \
Philadelphia, Pennsylvania, USA 19107-5233 /_ \_
Phone: 1-215-955-1645 Fax: 1-215-923-2475 email: besserm@jeflin.tju.edu
------------------------------------------------------------------------------

************************************************** ********************

I've used a couple of motion analysis systems to measure
finger span between the thumb and index finger -- real simple
stuff, but it taught me a lot. Mainly, forget SelSpot and use
OptoTrac (I don't know if you have a choice). The latter gave us
an order of magnitude smaller error (variance of spans measured
for a known length) than the former. We mounted the LEDs on the
fingertips using double-sided foam tape (from 3M) and then slipping
the tips of surgical gloves over the LEDs. The thin latex did not
bother the IR image, and greatly improved stability of the mount.

Good luck ...

From: clv2@Edu.CWRU.po (Clayton L. Van Doren)

************************************************** ************************

Hi, Nicky,
Dr. FongChin Su and Dr. Sakai did a study on hand motion in playing piano while
they were visitiing Mayo Clinic. They used Motion Analysis system. Dr. Su is
working on a technical note on the detailed techniques of data collection. Here
is Dr. Su's e-mail address: "fcsu@biomech.bme.ncku.edu.tw". Good luck.

Bing Yu, Ph.D.
Orthopedic Biomechanics Laboratory
Mayo Clinic
Rochester, MN 55905


************************************************** **************************

Dear Nicky:
We are constntly using 3 video cameras to analyze movement of hand
motion. You need to zoom close to the hand and use a calibration points
around the hand. The accrucy is .1 mm.
Gideon Ariel, Ph.D.


From: ariel1@com.netcom.ix (Gideon Ariel)


************************************************** **************************

Nicky,
Please let me know the results of the query concerning a
marker system for motion analysis of the hands. I may want to use
a similar system to validate the CyberGlove.

FYI: If you are not familiar with the CyberGlove, it is glove that
contains wire sensors that is used to measure position and movment of the
fingers and hand. The cost for one glove ranges from $10,000 to
$20,000, but you may want to contact the company for an accurate
quote. The address is:

Virtual Technologies
2175 Park Blvd.
Palo Alto, CA 94306
phone (415) 321-4900

************* Raymond McKenna, P.T.
** Physical Therapist/Research Assistant
** ** ** ** Texas Woman's University
** ** ** ** School of Physical Therapy
** ** ** ** ** 1130 M.D. Anderson Blvd.
********** ** ** Houston, Texas 77030
** ** (713) 794-2075
******** e-mail: HG_MCKENNA@TWU.EDU



************************************************** ************************

Nicky:
I have done 3-d kinematic analysis of ant locomotion using the DLT
technique, so a human hand seems HUGE by comparison. For a calibration
device, I found that small plastic LEGO blocks worked very well. I just
measured them with a vernier calipers. The biggest problem I have had is
the lack of depth of field. That is, at high magnification and high
framing rates light is in short supply. Thus the volume of space that is
in good focus is small. Solutions: add more light, close down lens
aperture.

Rodger Kram Ph.D.
Assistant Professor
Human Biodynamics Dept.
Univ. of California, Berkeley
rkram@garnet.berkeley.edu


************************************************** ************************

We have used the data glove, and an exo-skeleton for such finger tracking
stuff and also an Ascension Flock of birds or a Pholhemus magnetic position/
orientation sensor for tracking hand position.

Keith Karn
keith@cvs.rochester.edu



************************************************** ************************

I have done some work investigatin the hand motion during the luge
start. I have found that sewing small reflective markers on various
sizes of tight fitting cross-country ski gloves has provided adequate
tracking of the hand in motion. I use markers 6-7 mm square. I have
also used fast drying glue to stick the markers on the actual joint
locations ofthe subjects hand (they again were wearing tight fitting
gloves) the only problem with using the reflective markers is that the
filming bachground area must be realitively dark. I then use a halogen
utility lamp to reflect the markers. If I can be of more help please
E-mail me at shumphre@nmu.edu Good luck!, Sean Humphreys.



************************************************** ************************

Hi Nicky:
It sounds like we're doing a lot of similar things. Anyway, we presented
a paper at last years American Society of Biomechanics (17th Meeting in
Iowa, City 1993) about it. I was the first author. We had same small
markers from Selspot that we used (i think it was their no. 2 LED). LEt
me know if you have more questions.

////
Jack Tigh Dennerlein ~0-0~
http://mote.berkeley.edu/~jax/jax.html \- /


************************************************** *************************

Dear Nicky,
We have an ExpertVision system (Motion Analysis Corp) here (The University
of Sydney, Biomechanics Division)but I think our methods would apply equally
to your Vicon (is that what you are still using at Strathclyde?). In a study
of upper limb function (Jebsen) we measured upper trunk, arm, forearm, hand,
index finger and thumb. For the hand and digits we used small (4mm)
spherical markers attached with double-sided hypoallergenic adhesive tape to
the joints. Six cameras were necessary to cope with the rotations and two of
them were close up on the hand.
Good Luck, Richard

From: R.Smith@au.edu.su.cchs (Richard Smith)


************************************************** **************************

Dear Nicky,

Reading your question at BIOMCH-L I thought you might be interested to know
that in our laboratory we have been recording drawing and writing movements
with our Optotrak (3D movement registration) system. Our recordings involve
arm, hand, and index finger movements. Optotrak uses three infrared sensors to
register locations of infrared diodes in space. The diameter of these markers
is about 6 mm. We apply 4 of these to the index finger, two on both the first
and second phalanx, in order to estimate extension/flexion in their
interphalangeal joint. Extension/flexion and adduction/abduction in the first
metacarpophalangeal joint can be estimated by attaching another three markers
at the back of the hand. These are also used to estimate ext/flex and abd/add
of the wrist by relating them to another two markers at the forarm. We even
estimated pro/supination of the forearm with another markers. So we used a
total of 14 markers. In order to prevent the markers on the second phalanx of
the index finger from being obscured by the hand (due to the writing posture)
we had to mount them on a small strip that was attached to the phalanx under a
90 deg angle. So when the hand was flat on the table the strip with the markers
was pointing upward. Our main interest is joint angle *changes*, rather than
exact joint angles. Applying 3D geometry to the marker positions we can quite
well derive these changes. However, marker positions were chosen such that
marker positions were influenced as little as possible by soft tissue
deformations in *writing and drawing* tasks. For other tasks (slight) changes
of marker positions may be needed.
I realize that I am not being very explicit, but this message it really just
meant to let you know that we were faced with the same problems that I think you
have and that we have worked out a solution that fits our specific needs. If
you think our experience may be of any help to you, please inform me about your
goals and problems, so that maybe I can help.


Joost Schillings,
schillings@nici.kun.nl



************************************************** ***********************

Nicky,

You may want to try asking a former colleuge of mine who did his
masters thesis on the hand and pinching and grasping. I recall
helping him on 3D motion analysis using retroreflective markers:

Brock Horsley
kinetic@GVSU.EDU

Good Luck!


_/_/_/
Jim Patton
_/ _/ _/ _/_/_/_/_/_/ _/_/_/ _/_/ _/ Graduate Student
_/_/_/ _/_/ _/ _/ _/ _/ _/ _/ _/ Biomedical Engineering
_/ _/ _/ _/ _/ _/ _/ _/ _/ _/ Northwestern U
_/ _/_/_/ _/ _/ _/ _/ _/ _/_/ j-patton@nwu.edu
_/ _/ _/ _/ _/ _/_/_/ _/ _/ (312) 549-3263




************************************************** *************

Contact Karin Hollerbach (hollerbach1@LLNL.gov). Karin is using a MacReflex
System (Qualisys) to look at hands. She is located at the Lawrence Livermore
National Laboratory in California. Good Luck!


From: BillH4740@com.aol


************************************************** ******************

Dear Nicky,

I suggest that you contact Mr. Peter Meddings at Oxford Metrics, Ltd.,
manufacturer of the VICON system. Tel: 44 865 244 656

-Lee
From: LEEBARNES@com.delphi


************************************************** *******************

Nicky,

I have used the Vicon to track finger motion (both the
old VX and the new 370). I used their smallest
marker balls (8mm) on each phalanx without much
trouble. It used a small sized calibration object and
got residuals below 1mm. The problem is which motion
analysis system do you have?

Peter Kyberd
Oxford Orthopaedic Engineering Centre


************************************************** ************************

Dear Nicky Shaw-Hamilton,

Regarding your search for a motion analysis system for finger and
hand motion, have you considered the OPTOTRAK? Some of our customers
are using the OPTOTRAK for various grasping and hand motion studies.
It is an active marker system with high accuracy, realtime 3d
position tracking. There are options that allow such things as
realtime rigid body tracking and collecting analog data along with
the positional data. Our standard markers are infrared light
emitting diodes, mounted on disks of diameter 4, 8, or 16mm. One
would typically use the 4 or 8mm size for hand or finger studies.

If you have a copy of the proceedings from the 1994 IEEE
International Conference on Robotics and Automation, you may wish to
look at the first paper in section A1.10 (Rohling and Hollerbach).
It shows the use of the system by the Biorobotics Lab at McGill
University here in Canada to study the human index finger. Please
note, however, that the rather imposing apparatus in figure 1 is not
our system. An example of one of our markers is shown in figure 5 of
the paper.

If you would like further information about our system, please let me
know and I can mail or fax you a complete description. I could also
provide contact information should you wish to speak to current users
of the OPTOTRAK.

Good luck with your search.

Yours sincerely,

Margaret Fraser
Northern Digital Inc.

************************************************** *********************

Hi Nicky-
We're using a set of fiberoptics to measure joint angles in the hand.
The fiberoptic cable is run along the joint of interest, and the power loss as
the joint (and cable) bends is proportional to the joint angle.
My colleague has built a prototype system using low-grade fiberoptics.
The outer coating of the cable is stripped away in the area over the joint;
this increases the power loss when that area is bent. Spurious signal loss in
other areas of the cable between the site of interest and the
transmitter/receiver are diminished because those parts of the cable are still
clad. The associated electronics attempt to keep the power level of the
transmitted light constant; the measured quantity is the power the transmitter
requires to maintain the desired output (ie received) level.
This is still a prototype system, so I don't have a ready-made device (or
much information, for that matter) to give you. You would have to design/build
a significant amount of the system from scratch. You can get high-loss
fiberoptic cable and very small emitters/receivers quite cheaply.
good luck!
-Claudia Ranniger


From: "Claudia Ranniger"