First published online September 15, 2004
Journal of Experimental Biology 207, 3639-3648 (2004)
Published by The Company of Biologists 2004
doi: 10.1242/jeb.01182
Determination of peak vertical ground reaction force from duty factor in the horse (Equus caballus)
T. H. Witte1,
K. Knill1 and
A. M. Wilson1,2,*
1 Structure and Motion Lab, The Royal Veterinary College, Hawkshead Lane,
Hatfield, Hertfordshire, AL9 7TA, UK
2 Structure and Motion Lab, University College London, Royal National
Orthopaedic Hospital, Brockley Hill, Stanmore, Middlesex, HA7 4LP,
UK
View larger version (14K):
[in a new window]
|
Fig. 1. The orientation of the equine digit during foot on and foot off. The
accelerometer is mounted axially on the dorsal hoof wall with the sensitive
axis orientated disto-proximally. Critically, the foot acceleration vector
(indicated by the red arrows) is orientated along the sensitive axis of the
accelerometer at foot on (d) and at foot off (g). However, during roll over
(f), when the heel of the foot (the most rearward point of the ground-bearing
surface) has left the ground and the foot is rotating around the toe (the most
forward point of the ground-bearing surface), the vector is orientated
orthogonal to the sensitive axis.
|
|
View larger version (64K):
[in a new window]
|
Fig. 2. An accelerometer and telemetry unit in place on the distal limb of a horse.
The telemetry unit and battery are contained within the exercise bandage and
mounted on the lateral aspect of the third metacarpal bone. The accelerometer
is encased in epoxy and Kevlar fibres, mounted on the dorsal surface of the
hoof and protected from abrasion by the exercise surface with electrical
insulation tape.
|
|
View larger version (15K):
[in a new window]
|
Fig. 3. Representative graphs of simultaneous peak vertical ground reaction force
(GRFz) and accelerometer data collected during walk (A), trot (B) and canter
(C) locomotion. For walk and trot trials, force plate data were collected for
both front limbs (red broken line) and hind limbs (green broken line) during
the same trial. Beneath the force outputs, the corresponding accelerometer
traces are shown. Vertical broken lines indicate the timing of foot on (a) and
foot off (b). The red solid line indicates the output from the forelimb
accelerometer, and the green solid line the hind limb accelerometer data. Note
the precipitous rise in force at foot on and the rapid drop in force at foot
off, which allow the accurate determination of foot on and foot off from
vertical force data alone. Also note the absence of features in the
accelerometer trace during the stance phases, except a minor undulation
shortly prior to foot off, corresponding to heel off and foot rotation. The
automatic gain control in the telemetry link means that whilst the
accelerometer signal amplitude is the same in all three plots, the actual
acceleration will differ.
|
|
View larger version (29K):
[in a new window]
|
Fig. 4. Comparison of different accelerometer mounting positions by simultaneous
collection of data from several accelerometer locations. The output for the
hoof-mounted accelerometer is always shown in green, and the comparison
location in red. The output during locomotion on a hard surface of an
accelerometer mounted on the lateral aspect of the third metacarpal bone (MCP)
at walk (A) and trot (B), and the output of an accelerometer mounted on the
proximal phalanx (prox. phalanx) during walk (C) and trot (D) are shown. The
output during soft surface locomotion of an accelerometer mounted on the
proximal phalanx during walk (E), trot (F) and canter (G) is also shown.
Vertical broken lines indicate the time of foot on (a) and foot off (b).
Accelerometer output in volts cannot be converted to m s–2
due to automatic gain control within the analogue telemetry system used.
|
|
View larger version (27K):
[in a new window]
|
Fig. 5. Mean (± S.D.; broken lines) vertical ground
reaction force (blue) relative to percentage of stance duration for the
forelimb and hind limb at walk (A,B) and trot (C,D), and lead and non-lead
forelimbs at canter (E,F). A sine wave of equal base and area is superimposed
in red over the ground reaction force. Values in the upper left corner of each
graph indicate the percentage of the stance time for which the sine wave lies
within 1 S.D. of the mean force.
|
|
View larger version (22K):
[in a new window]
|
Fig. 6. Histograms showing the distribution of errors in peak limb force prediction
for individual stance phases using a sine wave of same base and area at walk
(A) and trot (B) and for the non-lead (C) and lead limbs (D) at canter.
|
|
View larger version (24K):
[in a new window]
|
Fig. 7. (A) Plot of front:hind ratio of peak vertical ground reaction force (GRFz)
versus peak forelimb GRFz, expressed as percentages. Data are shown
for walk (red diamonds), trot (green triangles) and canter non-lead leg
(filled blue circles). The linear regression line is fitted to all data
(y=1.6198–0.0358x, r2=0.2722,
P<0.0000). (B) Plot of front:hind vertical impulse ratio
versus peak GRFz, expressed as percentages. Symbols are as for A. The
horizontal red solid line shows the mean ratio across all speeds (1.33). The
red broken lines indicate ± S.D., and the blue
broken lines indicate the inter-quartile range.
|
|
View larger version (22K):
[in a new window]
|
Fig. 8. Scatter plot of predicted peak vertical ground reaction force (GRFz)
versus actual peak GRFz. Both are normalised to body mass. Data are
shown for walk (red diamonds), trot (green triangles), non-lead leg at canter
(blue circles) and lead leg at canter (black squares). The red line indicates
the function y=x. The red star shows the position of the
mean of the average lead and non-lead limb data.
|
|
View larger version (10K):
[in a new window]
|
Fig. 9. Difference in metacarpophalangeal (MCP) joint extension angle between left
and right forelimbs during treadmill locomotion at a range of speeds and gaits
for four horses. Colours indicate individual horses.
|
|
© The Company of Biologists Ltd 2004
