Technical Report

Accuracy of Digitization Using

Automated and Manual Methods
IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

Background and Purpose. Computerized 3-dimensional (3-D) motion

measurement systems are used by those interested in human motion. The
purposes of this study were (1) to determine the limits of accuracy in
determining intersegmental angles during pendular motion at varying
speeds and (2) to determine changes in accuracy introduced by autodigitization and digitization by experienced manual raters. Methods. Angular
speed of a T-shaped pendulum was systematically increased by releasing
the pendulum from 4 angles (0 @no movement#, 45, 90, and 120).
Twelve reference angles calculated from markers placed on the pendulum were estimated over 20 frames for 10 trials at each release position.
Results. Mean errors across trials and frames for intersegmental angles
reconstructed by a 3-D motion measurement system were within 61
degree across all release positions. An analysis of variance and a post hoc
Tukey test revealed that the mean error for the autodigitized trials was
larger than that for the manually digitized trials. For the autodigitized
trials, the static trials (release position50) produced less mean error than
the trials with movement produced. The ICCs showed a high degree of
consistency among all raters, ranging from .707 to .999. Conclusion and
Discussion. Our findings support the conclusion that under carefully
controlled conditions, a 3-D motion measurement system can produce
clinically acceptable measurements of accuracy across a range of angular
speeds. Furthermore, acceptable accuracy is possible regardless of the
digitization method. @Wilson DJ, Smith BK, Gibson JK, et al. Accuracy of
digitization using automated and manual methods. Phys Ther.
1999;79:558566.#

Daniel J Wilson
Bryan K Smith
J Kyle Gibson
Byung K Choe
Brenda C Gaba
John T Voelz

558

IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

Key Words: Accuracy, Digitization, Kinematics, Videotape analysis.

Physical Therapy . Volume 79 . Number 6 . June 1999

IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

he proliferation of 3-dimensional (3-D) motion

measurement systems has led to numerous studies addressing the accuracy and precision of
these systems.13 Estimates of the accuracy and
precision of various systems are usually provided by the
manufacturer. As Haggard and Wing stated, however,
behavioral testing @motion analysis# may take place
under less than ideal conditions.4(p316) Thus, assessing
the accuracy and reliability of data obtained with a
motion measurement system under varying conditions
becomes important.
One measure of a systems accuracy is the degree of
agreement between a reference standard and the estimates of that standard made by use of the system. There
has been general agreement among researchers that
system users should determine the limits of their systems accuracy in order to provide a basis for making
inferences based on the data. To provide system users
with suggestions and bases of comparison, several
authors13,5,6 have published accuracy limits for a variety

of kinematic systems (Tab. 1). The common method in

these studies is the use of automated digitization.
Measurement systems using special processing units to
digitize videotaped images online based on predefined
light-intensity thresholds (eg, Motion Analysis system*)
are in widespread use. Other systems, such as the Ariel
Performance Analysis System (APAS) and the Peak
Performance Technologies System, record the images
directly on videotape. A computer frame-grabbing board
is then used to convert the images to digital information
so that they may be scanned for light-intensity contrasts.
The light intensity of each pixel is determined, and
those light intensities above a user-defined level are used
to calculate coordinate positions. The use of the videotaped images makes the portability of such systems
possible, allowing the videotape data to be collected in a
*

Motion Analysis Corp, 3617 Westwind Blvd, Santa Rosa, CA 95403.

Ariel Dynamics Inc, 7887 Dunbrook Rd, Suite B, San Diego, CA 92126.

Peak Performance Technologies Inc, 7388 S Revere Pkwy, Suite 601, Englewood, CO 80112.

DJ Wilson, PhD, is Clinical Assistant Professor, Gait Laboratory, Department of Physical Medicine and RehabilitationDC 0046.00, Howard A Rusk
Rehabilitation Center, University of MissouriColumbia, One Hospital Dr, Columbia, MO 65212 (USA) (dan_wilson@pmr.missouri.edu). Address
all correspondence to Dr Wilson.
BK Smith, JK Gibson, PT, OCS, BK Choe, BC Gaba, and JT Voelz were graduate students, Department of Health and Exercise Sciences, University
of MissouriColumbia, at the time this study was conducted.
Concept, research design, and writing were provided by Wilson; data collection, by Smith, Gibson, Caba, and Voelz; data analysis, by Smith; and
project management, by Choe. Dr John Hewett and Jane Johnson of the Medical Infomatics Group at University of Missouri-Columbia assisted with
statistical planning, computing, and interpretation.
This study was funded, in part, by the National Institute of Child Health and Human Development, National Institutes of Health (grant 5 T32
HD07460 05).
The opinions contained in this article are those of the grantee and do not necessarily reflect those of the National Institutes of Health.
This article was submitted July 28, 1998, and was accepted February 24, 1999.
1999 by the American Physical Therapy Association Inc

Physical Therapy . Volume 79 . Number 6 . June 1999

Wilson et al . 559

Table 1.
Accuracy Limits Reported for Selected Computerized Kinematic Motion Measurement Systems

Author(s)

Measurement
System

Data
Acquisition
Region

Activity
Description

Reference
Standard

Accuracy
Limits

Wilson, et al6

Ariel Performance
Analysis System

1.83 m wide 3
0.61 m deep 3
1.52 m high

Dynamic angular
motion of a rigid
pendulum

12 angles at 4
initial angular
positions used to
vary angular
speed

Mean error of 0.182 across

all angular speeds

Klein and
DeHaven1

Ariel Performance
Analysis System

2 m wide 3
0.7 m deep 3
1.35 m high

Static measurement of
a meter stick and
goniometric angles

50-cm linear marks

on a meter stick
Goniometric
angles ranging
from 10 to
170.

49.8760.35 cma
(n527), 49.9560.78
cmb (n527)
Mean of the average
deviation (n517)
was 0.2660.21.

Scholz and
Millford3

Peak Performance
Technologies
Motion
Measurement
System

2 m wide 3 2 m
deep 3 1.31 m
high

Dynamic angular
motion of a rigid
pendulum

32 angles, each
estimated at 3
orientations

0.11.0; perpendicularc
0.01.2; 30
counterclockwise, and
30 clockwised

Vander Linden
et al5

Motion Analysis
System

1.63 m long 3
0.72 m deep 3
1.27 m high

Static goniometric
angles at varying
locations within the
field of view

Angles ranging
from 20 to
180 in 10
increments

Average within-trial variability

less than 0.4 for all
locations

178.5-mm rigid bar

attached to a
human leg moving
dynamically

Linear estimate
during walking
and sit-to-stand
trials

Walking: X# 5177.1 mm,

X# 5176.8 mm

Static goniometric
angles

Angles ranging
from 45 to
100 in 5
increments

95% confidence interval for

angle variability was within
0.5

Scholz2

WATSMART
(Waterloo Spatial
Motion Analysis
Recording
Technique)

1.2 m wide 3
1.2 m deep 3
1.8 m high

Movement trajectory
of a robot arm

Intraclass correlation
coefficients ranging from
.20 to .99 for varying
trajectories

Standard lens; camera pair positioned 3.8 m apart, 3.8 m from the front of data acquisition region (1.75 m high).
Wide-angle lens; cameras placed 3 m apart, 3 m from the front of data acquisition region (1.75 m high).
c
Frame to which the pendulum was attached was parallel with the field of view.
d
Frame to which the pendulum was attached was rotated 30 degrees counterclockwise and 30 degrees clockwise.
b

variety of settings. These settings, however, may not

always provide the conditions necessary for automated
digitization. For example, in our gait laboratory, we
often film patients outdoors on a variety of walking
surfaces to evaluate their gait.
The process of obtaining videotaped images and digitizing them manually off-line creates a number of potential
sources of error. Among these sources of error are
digitizer (observer) error, calibration error, marker
error, and skin movement or artifact error.7 Digitizer
error affects both the accuracy and the reliability of
measurements obtained with a motion measurement
system.4 Digitizer error often arises from improper man560 . Wilson et al

ual alignment of the superimposed crosshairs with the

landmark of interest. The crosshairs, provided by the
software to aid in locating the landmarks, must be
positioned exactly for accurate position determination.
Tracking errors occur when the object of interest cannot
be located accurately during motion. These errors most
often occur due to movement that is too rapid for the
hardware capabilities of the system. The Nyquist theorem states that the frame rate must be twice the motion
rate for accurate measurement. For example, a 60-Hz
camera records every 0.017 second. Movement that
occurs repeatedly during this time frame at a higher rate

Physical Therapy . Volume 79 . Number 6 . June 1999

IIIIIIIIIIIIIIIIIIIIIIII
Figure 1.
Details of the kinematic data collection setting. Note that the pendulums
axis of rotation is the top circle within the square (also a marker
location). The pendulums trajectory was centered within the calibration
space. C15camera 1, C25camera 2.

than 30 Hz cannot be tracked accurately. In this study,

we chose to investigate the effects of both digitizer error
and tracking error on the accuracy of a motion measurement system.

Figure 2.

The purposes of this study were (1) to determine the

limits of accuracy in determining intersegmental angles
during pendular motion at varying speeds and (2) to
determine changes in accuracy introduced by autodigitization and digitization by experienced manual raters.
To create the most realistic scenario possible for both
clinical gait laboratory and research, multiple raters
participated. Each digitizer performed a manual reduction of the same images to allow for interdigitizer
comparisons.

A Panasonic AG-7150 videocassette player was used for

the analysis of videotapes. Ariel software, version 6.92,
was used on an AST IBM 386 computer, with videotaped
images converted to digital information before digitizing
using a frame-grabbing board.

Method

Experimental Setting
Two Panasonic AG-455P video camcorders were placed
at 30-degree angles relative to the plane of activity for
filming (Fig. 1). The camcorders were equipped with
12:1 variable-speed control power zoom lenses with
digital focus. Film speed was 60 Hz, with a shutter speed
of 1/500 second. A single 300-W floodlight was positioned to the outside of each camera to illuminate the
retroreflective markers. Each camera was placed at a
height of 101.6 cm.

Panasonic Corp, One Panasonic Way, Secaucus, NJ 07094.

Physical Therapy . Volume 79 . Number 6 . June 1999

Dimensions and marker locations of the pendulum and background.

Trigonometrically determined angle measurements are given in Table 2.

A calibration structure was constructed of 5.08-cm polyvinyl chloride piping. Thirty calibration points (system
maximum) were inserted over 8 threaded 1.27-cm metal
calibration rods. Each calibration point consisted of a
2.54-cm-diameter spherical wooden ball wrapped in retroreflective tape. The calibration balls, which had holes
through their centers, were inserted over the calibration
rods at 30.48, 60.96, 91.44, 121.92, and 152.42 cm from
ground level. This placement of calibration balls formed
a rectangular calibration area 182.88 cm wide (x direction), 152.40 cm high (y direction), and 60.96 cm deep
(z direction). The position of each of the 30 balls was
verified prior to data collection by 3 independent observers using a metal tape measure until all observers
measurements agreed.
The testing equipment consisted of a rigid T-shaped
pendulum suspended by the bottom edge (Fig. 2). The
pendulum was fastened to a piece of plywood, which
served as the background, by a metal bolt. The bolt
Wilson et al . 561

Table 2.
Trigonometrically Calculated Reference Angles (Italicized) and Means of Reconstructed Angular Estimates (in Degrees) by Each Manual Digitizer
Collapsed Across the Four Angular Speeds
Anglea

Reference u
Digitizer 1 (auto)
Digitizer 2
Digitizer 3
Digitizer 4
Digitizer 5
Digitizer 6
Mean error by angleb

DAE

DAF

HAG

HAF

10.50
10.43
10.48
10.48
10.52
10.48
10.49
0.027

20.80
20.70
20.75
20.75
20.80
20.77
20.78
0.042

10.50
9.76
9.78
9.75
9.88
9.82
9.78
0.448

20.70
20.59
20.68
20.69
20.72
20.70
20.71
0.028

DBE

DBF

HBG

HBF

13.40
13.26
13.40
13.36
13.43
13.40
13.41
0.037

27.20
26.42
26.81
26.60
26.80
26.84
26.72
0.502

13.30
13.15
13.17
13.18
13.20
13.18
13.17
0.125

27.00
26.93
26.97
26.97
26.96
26.95
26.96
0.043

DCE

DCF

HCG

HCF

21.30
21.32
21.24
21.33
21.23
21.23
21.26
0.048

56.30
56.20
56.29
56.28
56.31
56.26
56.30
0.030

20.56
20.56
20.67
20.66
20.67
20.63
20.69
0.087

55.50
55.89
56.16
56.11
56.04
55.94
56.19
0.555

Anglea
Reference u
Digitizer 1 (auto)
Digitizer 2
Digitizer 3
Digitizer 4
Digitizer 5
Digitizer 6
Mean error by angle

Anglea
Reference u
Digitizer 1 (auto)
Digitizer 2
Digitizer 3
Digitizer 4
Digitizer 5
Digitizer 6
Mean error by angle
a
b

Mean Error
by Raterb
0.255
0.203
0.083
0.105
0.093
0.080

Mean Error
by Rater
0.285
0.138
0.198
0.143
0.133
0.165

Mean Error
by Rater
0.128
0.210
0.190
0.183
0.155
0.215

See Figure 1 for marker locations.

Calculated using the absolute value of the mean errors.

served as the axis of rotation, and nonplanar movement

about the bolt was minimized. Eight spherical 1.90-cmdiameter polystyrene balls wrapped in retroreflective
tape were secured to the pendulum with double-sided
tape. The background plywood and pendulum were
painted black in an effort to minimize reflection.

Data Collection
The calibration frame was filmed within the activity field
and removed. Camera settings were maintained thereafter, with the autofocus placed on manual (fixed). Twelve
reference angles were calculated trigonometrically from
the 8 reference balls positioned on the pendulum (Tab. 2).
Angular values were rounded to the nearest 10th of a
degree.
The pendulums trajectory was videotaped while the
pendulum moved at 4 angular speeds, including a
zero-movement condition. The angular speeds were
produced by rotating the pendulum about its axis (from
562 . Wilson et al

vertical) to 45, 90, or 120 degrees for release. These

positions were selected to provide a range of angular
speeds varying from slower (no movement) to faster
than those commonly encountered in human gait analysis. The angular position of the pendulum was verified
prior to each trial using an angle locator.i Ten trials at
each of the 4 angular speeds were videotaped.
To provide a common spatial orientation for comparison, 10 frames on either side of the vertical downward
position were digitized for each angular speed. The
angular speed changed across these 20 frames, except
for the zero-movement condition. The variability due to
this change in angular speed on a frame-to-frame basis is
hereafter referred to as frame. The data collection

Johnson Level and Tool Manufacturing Co Inc, 6333 W Donges Bay, Mequon,
WI 53092.

Physical Therapy . Volume 79 . Number 6 . June 1999

IIIIIIIIIIIIIIIIIIIIIIII
Table 3.
Analysis of Variance Effects for the Overall General Linear Model
Source

df

SS

MS

Position
Rater
Position 3 rater
Angle
Frame

3
5
15
11
18

48.84
76.47
17.06
4650.97
637.51

16.28
15.29
1.14
422.82
35.42

115.42
108.44
8.07
2997.95
251.13

.0001
.0001
.0001
.0001
.0001

procedure thus produced 120 reconstructed angular

estimates (12 angles 3 10 trials) for each of the 4
angular speeds.

angular speed relative to the differences among the

angular speeds.10 The ICCs were calculated as a percentage of variation among groups as follows:

Data Reduction
The 80 independent film clips (4 angular speeds 3 10
trials 3 2 views) were manually digitized across the 20
frames by 5 experienced raters. Each rater had a minimum of 16 weeks (an academic semester) of experience
in manual digitization. Data produced by each persons
digitization were compared with autodigitized data to
verify their accuracy. The data were transformed to 3-D
coordinates using direct linear transformation (DLT)
and smoothed with a Butterworth second-order, low-pass
recursive filter. The filters cutoff frequency (fc) was
chosen on the basis of the point of linearization of the
filtered and unfiltered displacement, using successive
increments in the fc.8 The fc chosen (6 Hz) agreed well
with values for fc in published human gait analysis
studies.9 Three-dimensional coordinates generated by
the DLT were used to compute reconstructed estimates
of the 12 reference angles across the 20 frames at each
angular speed.

sA2 /@s21sA2#

Data Analysis
Variability scores (error) for each angle were calculated
by subtracting the reconstructed 3-D angle from the
calculated reference angle. Independent variables
included the pendulums release position (angle), the
angles formed by the markers attached to the pendulum
(angle), the 20 frames digitized for each trial (frame),
and the 6 raters (including autodigitization) (rater). A
4-factor (position 3 angle 3 frame 3 rater) analysis of
variance (ANOVA) for repeated measures on each variable was used to evaluate accuracy. Angle and frame
were treated as blocking variables to eliminate any
interactions with position or rater. This statistical treatment does not include the variability due to these 2
variables in the main effects (rater and position)
ANOVA.
Intraclass correlation coefficients (ICCs) were used to
estimate the variability within release positions (across
frames). The ICCs were calculated for each rater within
each release position. The ICC is a measure of the
similarity among the angle estimates within a given
Physical Therapy . Volume 79 . Number 6 . June 1999

Where sA2 is the variance among groups and s2 is the

error of variance. Thus, the ICC has the advantage over
other correlation methods (eg, Pearson product-moment
correlation) of the ability to estimate the agreement for
repeated measures (angles).
Results
Mean errors (reconstructed angle subtracted from reference angle) for each rater and each angle are given in
Table 2. The mean errors for each of the raters were
within 61 degree. Similarly, the mean error of the
estimates of each of the 12 angles was within 61 degree.
The largest error associated with the angles was found in
estimating angle HCF (mean error50.56) and angle
DBF (mean error50.50). Although these were the 2
largest angles estimated, no clear pattern of error associated with angle size was evident.
Results of the overall ANOVA for the general linear
model are shown in Table 3. Each of the main effects
and the position 3 rater interaction were significant
(effects discussed individually in next section). We tested
for differences due to rater within each position and for
differences due to position for each rater.

Rater (Digitizer) Effects

To determine whether there were differences between
raters, a 3-factor ANOVA was used. Multiple comparisons were done in 2 ways: (1) all pairs of raters were
compared using the Tukey multiple-comparison procedure, and (2) each manual rater was compared with the
autodigitizer using the Dunnett test.
The overall ANOVA for each of the 4 release positions
was significant (F516.28; df53,37; P ,.0001) (Tab. 3).
The results of the ANOVA for the rater effects by
position are summarized in Table 4. A Tukey multiplecomparison test for each release position revealed the
mean error for the autodigitized trials (0.255) to be

Wilson et al . 563

Table 4.

Ratera Effects by Position

Release
Position ()

Tukey Multiple-Comparison
Test

Dunnett Test

.0001

1.2, 4, 5, 6
3.2, 4, 5

1.2, 4, 5, 6

45

.0001

1.2, 3, 4, 5, 6
2, 3, 6.4, 5, 4,5

1.2, 3, 4, 5, 6

90

.0001

1.2, 3, 4, 5, 6
4,5

1.2, 3, 4, 5, 6

120

.0001

1.2, 3, 4, 5, 6
3.2, 6

1.2, 3, 4, 5, 6

Rater 1 refers to autodigitization.

Table 5.

trials. There was not, however, a consistent trend of

decreasing reproducibility with increasing angular speed
(release position).

Position Effects by Rater

Rater

Tukey Multiple-Comparison Test

1 (auto)
2
3
4
5
6

.0001
.0001
.0001
.0001
.0001
.0001

0,45, 90,
0, 45,90,
45,0, 90,
0, 45,90,
0,45, 90,
45,0, 90,

120,
120
120,
120
120,
120,

45, 120,90
0,90, 120
45, 120,90
0, 120,90

The 90-degree release position produced the least consistency for each rater, except for rater 2. The range of
ICCs for this release position was .703 to .883. Despite
the greater angular speed (relative to the 90 release
position), the 120-degree release position produced
ICCs ranging from .884 to .939.
Discussion

larger (P ,.01) than the mean error for most of the

manual raters (range50.080 0.203). The results of
the Dunnett test confirmed that the mean error for the
autodigitized trials was larger (P ,.0001) than the mean
error for all of the manual raters (except rater 3 for the
zero-movement trials) for each of the 4 release positions
(Tab. 4).

Position (Angular Speed) Effects

A 3-factor ANOVA was used to determine whether there
was a position effect within each rater. Multiple comparisons were performed using the Tukey multiplecomparison procedure. For the autodigitized trials, the
zero-movement trials produced less (P ,.05) mean error
than the 3 conditions with movement (Tab. 5). Three of
the 5 manual raters also produced the least error while
digitizing the zero-movement trials. The remaining 2
manual raters had their least error associated with the
trials produced by releasing the pendulum from the
45-degree position. The greatest error was associated
with the 90-degree release position for each rater
(including autodigitization), either alone or in combination with the 120-degree release position.
Rater Consistency
The range of ICCs for all raters (including autodigitization) was .999 to .703 (Tab. 6). In all except one case
(rater 4), the highest ICC was for the zero-movement

564 . Wilson et al

In analyzing intersegmental angles during motion, at the

range of angular speeds used in this investigation, both
experienced manual raters and autodigitization produced what we would consider excellent accuracy (61).
The error associated with the autodigitized trials agreed
with the error ranges reported by other authors (Tab. 1).
No consistent increase in error increase was found to be
associated with increasing angle size or increasing angular speed due to release position.
The autodigitized trials contained greater error
(X50.223) than the manually digitized trials
(X50.153). The reason for the greater error within the
autodigitized trials was unclear; however, we believe that
lighting conditions may have produced digitizing error
specific to autodigitization. While filming an object in
motion, lighting conditions may produce errors in centroid identification.7 Centroid calculations are based on
the mathematical center of all pixels identified above a
threshold light level. In our laboratory, we have identified varying pixel configurations as lighting conditions
are altered due to the motion of the object being
videotaped. In contrast, manual raters have the advantage of a spherical presentation of markers even if
shading conditions of the marker are inconsistent. The
center of the marker may be identified regardless of
uneven lighting of the marker. A number of other

Physical Therapy . Volume 79 . Number 6 . June 1999

IIIIIIIIIIIIIIIIIIIIIIII

potential sources of error will be discussed later in this

section.
Despite the largest error values, the measurements
obtained in the autodigitized trials had the best reliability. This result seems to indicate that errors in the
autodigitized trials were consistent, perhaps indicating a
nonrandom (systematic) error component. Whether
this nonrandom error was specific to our data collection
setting or system is unclear. Similar studies performed in
other laboratories may help answer this question.
The results of this investigation may be deceiving, given
the various statistically significant differences reported.
The statistically significant differences in error reported
do not imply a lack of clinical utility of the kinematic
data. The range of errors for the 12 angles was 0.028 to
0.556 degree. We believe these values are clinically
acceptable, in agreement with error ranges reported by
other authors.1,2,5 Statistical interpretation of the results
is relevant only to the degree that it may help find ways
to reduce error by improving instrumentation or data
collection methods.
The generalizability of our results is limited with respect
to clinical applications. The idealized data collection
setting used to evaluate the limits of the systems accuracy does not reflect all of the sources of error that may
exist in clinical settings. One important source of error
in clinical videotape settings is the placement of external
markers on the body to estimate joint rotational centers
or the mechanical axis of a body segment. There is some
question as to whether these markers can be placed in
the correct locations and whether these placements can
be reproduced. It is generally difficult to locate a good
estimate of the instantaneous center of rotation.11 Due
to the idealized setting used here, instantaneous centers
of rotation were not estimated. In clinical practice,
however, at least 3 markers are necessary to calculate 3-D
angles between 2 rigid bodies: one on the instantaneous
center of rotation and one on each body (segment). The
use of 2 markers on each body segment allows for the
calculation of the angular change between the lines
(vectors) connecting the markers on the respective
segments. Error in the placement of markers at any of
the 3 points will produce error in the angle calculated
from these markers.
Schamhardt et al7 contended that the size of the markers
is the main determinant of the obtainable accuracy in a
kinematic system. For a high resolution (1,000 3 800
pixels) image system, covering a field of view of 5 3 4 m,
5 pixels or image lines in height and width was reported
to be needed for acceptable accuracy. Dividing the
screen resolution (1,000 pixels) by the field of view (500
cm) gives the conversion factor (2 pixels/cm) for find-

Physical Therapy . Volume 79 . Number 6 . June 1999

Table 6.
Interframe Digitization Consistency by Rater
Release
Position ()

ICC

1 (auto)

0
45
90
120

.999
.975
.850
.917

0
45
90
120

.979
.785
.852
.937

0
45
90
120

.988
.976
.848
.926

0
45
90
120

.703
.827
.703
.884

0
45
90
120

.968
.901
.836
.939

0
45
90
120

.978
.976
.883
.933

Rater

ing the needed marker size. Subsequently dividing the

required image lines (5 pixels) by the conversion factor
gives the marker size (2.5 cm diameter) needed to
obtain the required accuracy.
In our study, the APAS image resolution is governed by
the frame-grabbing board. The board has an image
resolution of 640 3 480 pixels. The 1.83 m (width) 3
1.52 m (height) field of view used in this study, therefore, would require 1.43-cm-diameter markers (5.0
pixels/3.50 pixels/cm) for acceptable accuracy. The
1.90-cm-diameter markers used should not have created
unacceptable errors in accuracy.
One source of error that was not present in our study was
the possibility of miscalibration. Calibration must be
performed in order to determine the positions of the
camera(s) in the laboratory space. The accuracy of the
3-D coordinates is determined mainly by the accuracy of
the calibration process. Estimates of the distance
between the D and H markers (59.8 cm) on the pendulum (Fig. 2) obtained by the APAS were exact.
Conclusion
This investigation demonstrated limits of accuracy for
the APAS for autodigitization and manual digitization.
Our findings support the conclusions (1) that the accuracy of determining intersegmental angles (61) is, in
Wilson et al . 565

our view, acceptable for most clinical uses and (2) that
the most likely error source may be changes of illumination of the reflective markers during motion and possibly the resolution of the frame-grabbing board or camera speed.

5 Vander Linden DW, Carlson SJ, Hubbard RL. Reproducibility and

accuracy of angle measurements obtained under static conditions with
the Motion Analysis video system. Phys Ther. 1992;72:300 305.

References

7 Schamhardt HC, van den Bogert AJ, Hartman W. Measurement techniques in animal locomotion analysis. Acta Anat (Basel). 1993;146:123129.

1 Klein PJ, DeHaven JJ. Accuracy of three-dimensional linear and

angular estimates obtained with the Ariel Performance Analysis System. Arch Phys Med Rehabil. 1995;76:183189.
2 Scholz JP. Reliability and validity of the WATSMART threedimensional optoelectric motion analysis system. Phys Ther. 1989;69:
679 689.
3 Scholz JP, Millford JP. Accuracy and precision of the Peak Performance Technologies motion measurement system. Journal of Motor
Behavior. 1993;25:27.
4 Haggard P, Wing AM. Assessing and reporting the accuracy of
position measurements made with optical tracking systems. Journal of
Motor Behavior. 1990;22:315326.

566 . Wilson et al

6 Wilson DJ, Smith BK, Gibson JK. Accuracy of reconstructed angular

estimates obtained with the Ariel Performance Analysis System. Phys
Ther. 1997;77:17411746.

8 Wood GA. Data smoothing and differentiation procedures in biomechanics. Exerc Sport Sci Rev. 1982;10:308 362.
9 Winter DA. Biomechanics and Motor Control of Human Movement. 2nd
ed. New York, NY: John Wiley & Sons Inc; 1990.
10 Sokal RR, Rohlf FJ. Biometry: The Principles and Practice of Statistics in
Biological Research. 2nd ed. New York, NY: WH Freeman and Co; 1981.
11 Leach DH, Dyson S. Instant centres of rotation of equine limb joints
and their relationship to standard skin marker locations. Equine Vet J
Suppl. 1988;49:113119.

Physical Therapy . Volume 79 . Number 6 . June 1999