View Full Version : Summary: Marker placement for wrist analysis

04-25-1997, 05:47 AM
Dear all,

Fri, 11 Apr 1997 I posted a question regarding marker placement for
kinematic analysis of the wrist.
The following persons responded:
Joel M.C. Bach
Kris O'Connor
Anne Hollister
Your contributions are highly appreciated!

Below is my original posting followed by a summary of the responses,
both in short-form. I also included relevant parts of my
correspondence with Steve Work at Ascension Technical Support
(the manufacturer of MotionStar), as well as some
literature references with relevance to wrist analysis. I will mail the
complete responses to those who request it.

I also gladly receive further comments or questions regarding my
original posting or the solutions proposed.

Oyvind Stavdahl (Siv.ing., Dr.ing. student)
Department of Engineering Cybernetics
NTNU, The Norwegian University of
Science and Technology Direct line: +47 73 59 43 77
O. Bragstads plass 8 Switchboard: +47 73 59 43 76
N-7034 TRONDHEIM Fax: +47 73 59 43 99
NORWAY Email: Oyvind.Stavdahl@itk.ntnu.no

----------- Original posting (edited) ----------

Date: Fri, 11 Apr 1997 11:28:01 +0200
From: Oyvind Stavdahl
Subject: Marker placement for wrist analysis
To: BIOMCH-L@nic.surfnet.nl

We are about to conduct an experiment related to wrist
motion by means of a MotionStar magnetic tracking system.
We need to simultaneously measure both pro/supination,
radioulnar deviation and flexion/
extension, thus we need one MotionStar receiver ("marker")
distal to the wrist and one that has a constant orientation
with respect to the proximal radioulnar joint.

We have tried to attach the proximal receiver on a
plaster-of-paris "socket" that fits
firmly over the humerus condyles as well as
the olecranon. The socket is displaced with respect to
the bones when the underlying tissue moves, which
introduces errors in our wrist angle estimates.

Does anyone have experience with this type of
problem (I believe yes), and if so, where/how
should our MotionStar receivers be attached in
order to minimize the "crosstalk" from soft
tissue and irrelevant (elbow) joint movements?

Also, does anyone have experience with
linearization of magnetic tracking device
output in order to reduce errors induced by
metal objects etc.?

------------- Summary of responses -------------
My own comments are enclosed in [square brackets]. The term
"marker" denotes a MotionStar "receiver" i.e. the part of the
measurement system that is attached to the body segment which
we want to track.
The author accepts responsibility for any typos.

The crosstalk/skin motion problem:

- The only "easy" solution to the skin motion problem is anchoring
the markers directly to the bones. (Bach)
- The proximal marker should be attached to the ulna since
pronation/supination is a result of the radius moving about the
stationary ulna. (O'Connor)
- Mounting markers on thermoplastic plates which are then attached
to the forearm and hand helps to minimize the movement between our
markers and the skin. This may not be too helpful given the
relatively large size and weight of the MotionStar receivers. (Bach)
- The other solution would require the use of a third sensor,
placed on the upper arm. The forearm sensor could then be placed
proximally to the wrist. By adapting the joint coordinate system
suggested by Grood and Suntay (1985?) for the knee, pronation/supination
would be measured as the difference between the upper arm and
forearm orientations about the forearm long axis. (O'Connor)
- The following persons have all quantified the problems with skin
movements and markers etc.:
Karin Hollerbach
Bob Van Vorhis
Carolyn Small [ (lit. ref. below)]

The metal induced distortion problem:

- At the moment I don't know of any solution other than to remove
metallic objects from the testing area. One company, Skill Technologies,
Inc., is planning to implement a mapping algorithm in the future as
part of their motion analysis software, which would allow you to
correct for metallic distortions. (O'Connor)
- Metal in the field, fluorescent lights, powerlines etc. all can affect
the signals and introduce noise. We have used the polhemus device with
very little noise on the elbow and knee, the caveat being that there be
no metal in the vicinity and that all fluorescent lights be off. (Hollister)
- For electrically conductive metals, the measurement rate will play a
role in the amount of distortion due to eddy-current generation. Our
Pulse DC magnetic technology waits after applying the signal to the
transmitter before measuring the magnetic field generated. Waiting
allows time for the eddy-currents to die-down. The longer time waited,
the less eddy-currents exist to distort our generated field. The eddy-
current distortion will appear in the data to be constant between
measurement if the physical orientation is not changed. (Works)
[This applies to the MotionStar/Flock of Birds, but not to AC based trackers
like Polhemus.]
- Keep any large amounts of metal further from the transmitter than the
distance between the receivers and the transmitter. The distances
mentioned are in *all* directions. Large amounts of metal means metal
sheets, bars, or mesh. Not small numbers of metal screws, nails, or
other small hardware. (Works)

Data processing/miscellaneous comments:

- The three axes about which you wish to measure motions are non-
intersecting and non-orthogonal. (Bach)
[This is not a problem in our case because we will only study the
resultant orientation of the pro/sup-dev-ext/flex rotation triple;
none of the single rotation axes is interesting in itself.]
- Mounting the transmitter so close to the axis of rotation will make
the measurements extremely sensitive to artifactual errors (more so
than F/E and R/U). Mounting [the marker] on an outrigger may help but
it may also cause other problems with getting a secure and stable
attachment. (Bach)
[This does not apply to the MotionStar system, since each "marker"
yields its three translational and three rotational "global"
(transmitter-based) coordinates. The statement above applies to
systems that estimate rotations from marker positions only.]
- Andrei Lupichuk has written an algorithm
for the elbow data from the polhemus device which is very reproducible
between and within specimens. It was published in him master's thesis
and is written for a 486. (Hollister)
- Stuart Rothenberg (taroberts@aol.com) has a system for accurate
measurement of wrist motion which gives direct read-out of motion about
the wrist axes of rotation. The system is the easiest to use in the work
place. (Hollister)

Finally a few references for the benefit of those who want to "dig deeper"
into wrist kinematics and measurements:

R. H. Brumfield and J. A. Champoux, A Biomechanical Study of Normal Functional Wrist
Motion. Clinical Orthopaedics and Related Research (187):23-25, jul/aug 1984.

Erdman, A.G. et al. Kinematic analysis of the human wrist by stereoscopic
Instrumentation. J Biomech Engng 101: 124-133, 1979.

W. T. Jackson, M. S. Hefzy and H. Guo, Determination of Wrist Kinematics using a Magnetic
Tracking Device. Med. Eng. Phys. 16:123-33, mar 1994.

A. de Lange, J. M. G. Kauer and R. Huiskes, Kinematic Behavior of the Human Wrist Joint: A
Roentgen-Stereophotogrammetric Analysis, Journal of Orthopaedic Research 3:56-64, 1985.

Ronald L. Linscheid, Kinematic Considerations of the Wrist. Clinical Orthopaedics and Related
Research, (202):27-39, jan 1986.

K. A. Mann, F. W. Werner and A. K. Palmer, Frequency Spectrum Analysis of Wrist Motion for
Activities of Daily Living. Journal of Orthopaedic Research, 7(2):304-306, 1989.

J. A. Moore, R. E. Ellis, D. R. Pichora and A. M. Hollister, A Kinematic Technique for Describ
ing Wrist Joint Motion: Analysis of Configuration Space Plots. J of Engineering in Medicine,
Proc Instn Mech Engrs, Part H, 207:211-218,1993.

Oshima, J., Yamamoto, T. and Matsuoka, K., On an Estimation of the Location of the Axes of
Wrist Joint Movements. Proceedings of the 16th Annual International Conference of the IEEE
Engineering in Medicine and Biology Society, 16:333-4, nov 1994.

A. K. Palmer, F. W. Werner, D. Murphy and R. Glisson, Functional Wrist Motion: A Biome
chanical Study. The Journal of Hand Surgery, 10A(1):39-46, jan 1985.

Peterson, Steven W., and Erdman, Arthur G. Analysis and display of human
wrist motion. Biostereometrics `82, Proc of SPIE (International Soceity
for Optical Engineering), vol 361, p257-261.

J. Ruy, W. P. Cooney, L. J. Askew, K.-N. An and E. Y. S. Chao, Functional Ranges of Motion
of the Wrist Joint. The Journal of Hand Surgery, 16A(3):409-419, may 1991.

Small,C.F., Pichora,D.R., Bryant,J.T. and Griffiths,P.M. (1993) "Precision
and Accuracy of Bone Landmarks in Characterizing Hand Wrist Position", J.
Biomedical Eng., 15, 371-378

A paper by Sprigings (J of Biom 27-3) describes a study where small
reflective(?) spheres where attached to both sides of the elbow and to
both sides of a rod which was attached to the distal forearm - for stydying