Seediscussions,stats,andauthorprofilesforthispublicationat:https://www.researchgate.net/publication/230780639
ComparisonofStrain-GageandFiber-Optic
GoniometryforMeasuringKneeKinematics
DuringActivitiesofDailyLivingandExercise
ArticleinJournalofBiomechanicalEngineering·August2012
ImpactFactor:1.78·DOI:10.1115/1.4007094·Source:PubMed
CITATIONS
READS
6
150
6authors,including:
AbeerMohamed
AndrewSexton
UniversityofNewBrunswick
UniversityofNewBrunswick
2PUBLICATIONS6CITATIONS
11PUBLICATIONS11CITATIONS
SEEPROFILE
Allin-textreferencesunderlinedinbluearelinkedtopublicationsonResearchGate,
lettingyouaccessandreadthemimmediately.
SEEPROFILE
Availablefrom:AndrewSexton
Retrievedon:16May2016
Comparison of Strain-Gage
and Fiber-Optic Goniometry for
Measuring Knee Kinematics During
Activities of Daily Living and Exercise
The results indicate that both goniometers were within 2–5
degrees of the Vicon angles for gait and chair rise. For some deep
knee bend trials, disagreement with Vicon angles exceeded ten
degrees for both devices. We conclude that both goniometers can
record ADL knee movement faithfully and accurately, but should
be carefully considered when high (>120 deg) knee flexion angles
are required. [DOI: 10.1115/1.4007094]
Abeer A. Mohamed
Keywords: knee kinematics, flexible goniometer, fiberoptic
goniometry, gait, chair rise, deep knee bend
Jennifer Baba
1
Institute of Biomedical Engineering and
Department of Mechanical Engineering,
University of New Brunswick,
Fredericton, NB, E3B 5A3 Canada
Introduction
Rehabilitation following onset of knee joint disease, injury or
surgery often requires assessment of mobility during common activity of daily living, such as walking or rising from a chair [1,2],
or deep knee bends [3,4]. In these studies, highly sophisticated optical or radiographic systems were used to assess knee joint
motion; however, this technology is not available to the majority
of clinics, and costs of personnel, training and maintenance can be
prohibitive even when such facilities are readily available.
Wearable sensor technology offers a potential solution to this
barrier [5] by enabling functional outcomes to be measured outside the laboratory or clinic. Recent focus has centered on inertial
measurement devices for indirectly quantifying joint motion [6,7].
However, flexible goniometry can provide direct measurement of
joint angle and is easily deployed in the field [8]. Flexible straingauge goniometers are portable and reliable [9,10] and remain
popular in laboratory environments [8,11]. A common choice for
the knee is the SG150 Penny & GilesTM goniometer (Biometrics
Ltd, Gwent, UK; Figs. 1(a) and 1(c)) [8,12,13]. An alternative
flexible goniometer is the S700 ShapeSensorTM fiber-optic goniometer (Measurand Inc. Fredericton, Canada; Figs. 1(b) and 1(c)).
This device consists of a fiber-optic sensor embedded in a thin,
flat and flexible vinyl casing joining the two end blocks. Although
fiber-optic sensors have been use in rehabilitation and biomechanics applications [14], we know of no studies comparing these
two different flexible goniometer devices.
Therefore, the primary objective of this project was to evaluate
the measurement error of both a strain-gauge goniometer (Penny
& Giles Egon) and a fiber-optic goniometer (ShapeSensorTM,
James Beyea
Institute of Biomedical Engineering and
Faculty of Kinesiology,
University of New Brunswick,
Fredericton, NB, E3B 5A3 Canada
John Landry
Andrew Sexton
Institute of Biomedical Engineering,
University of New Brunswick,
Fredericton, NB, E3B 5A3 Canada
Chris A. McGibbon1
Institute of Biomedical Engineering and
Faculty of Kinesiology,
University of New Brunswick,
Fredericton, NB, E3B 5A3 Canada
e-mail: [email protected]
There is increasing interest in wearable sensor technology as a
tool for rehabilitation applications in community or home environments. Recent studies have focused on evaluating inertial
based sensing (accelerometers, gyroscopes, etc.) that provide only
indirect measures of joint motion. Measurement of joint kinematics using flexible goniometry is more direct, and still popular in
laboratory environments, but has received little attention as a
potential tool for wearable systems. The aim of this study was to
compare two goniometric devices: a traditional strain-gauge flexible goniometer, and a fiberoptic flexible goniometer, for measuring dynamic knee flexion/extension angles during activity of
daily living: chair rise, and gait; and exercise: deep knee bends,
against joint angles computed from a “gold standard” Vicon
motion tracking system. Six young adults were recruited to perform the above activities in the lab while wearing a goniometer
on each knee, and reflective markers for motion tracking.
Kinematic data were collected simultaneously from the goniometers (one on each leg) and the motion tracking system (both legs).
1
Corresponding author.
Contributed by the Bioengineering Division of ASME for publication in the
JOURNAL OF BIOMECHANICAL ENGINEERING. Manuscript received December 19, 2011;
final manuscript received June 1, 2012; accepted manuscript posted July 6, 2012;
published online August 6, 2012. Assoc. Editor: Richard Neptune.
Journal of Biomechanical Engineering
Fig. 1 The two flexible goniometers used in the study. (a) Penny
& Giles type flexible goniometer (Biometric’s Ltd) mounted on
the right knee. (b) ShapeSensorTM fiber-optic goniometer (Measurand Inc.) mounted on the left knee. (c) Form-factor comparison
of the two devices. (d) Plug-in Gait (lower extremity) marker
attachment sites and joint center locations.
C 2012 by ASME
Copyright V
AUGUST 2012, Vol. 134 / 084502-1
Downloaded From: http://biomechanical.asmedigitalcollection.asme.org/ on 01/18/2015 Terms of Use: http://asme.org/terms
Fgon), when used to capture sagittal plane knee motion, relative
to a “gold standard” optical tracking system (Vicon, Oxford Metrics, UK). The primary question being asked is: does the Fgon
meet the measurement standards set by the traditional Egon? This
information is required as a first step in evaluating the feasibility
of health professionals using flexible goniometers for rehabilitation applications in a wider range of environments.
2
Methods
2.1 Participants. Six volunteers (three males and three
females) participated in this study (mean 6 SD age: 23 6 2.3
years, weight: 66.9 6 11.43 kg, height: 167.8 6 11.58 cm). Participants had no musculoskeletal, neurological or sensory impairments or related disabilities. The study was approved by the
institutional research ethics board, and each volunteer provided
written informed consent.
2.2 Equipment and Procedures. Two goniometer devices
were evaluated for measuring dynamic knee flexion/extension during activity of daily living (ADL) and exercise: (1) Egon: 2-DOF2
Penny & GilesTM; and (2) Fgon; 1-DOF ShapeSensorTM. Data
from both devices were captured using a custom-built lightweight
multichannel A/D acquisition device (Biological Signals Interface,
or BioSITM, University of New Brunswick, Canada) carried in the
participant’s hand, and tethered to a laptop computer.
2.3 Device Characteristics. Linearity of goniometer devices
was determined from bench testing in the absolute range of knee
motion expected during experiments (0 6 150 deg) with the flexible
goniometer mounted on a large plastic reference goniometer fixed
to a flat table top. Two testers performed the trials twice. Both goniometers can measure in the positive and negative angle ranges
(depending on which side of the joint mounted), therefore data was
collected (and averaged over five seconds trials at 200 Hz) at ten
degree increments from 0 to 150 deg and 0 to 150 deg. Calibration coefficients were determined by fitting 1st order polynomials
to the goniometer and reference data for three different ranges: low
flexion: 0 to 670 deg (gait); moderate flexion: 0 to 6110 deg (chair
rise), and; high flexion: 0 to 6150 deg (deep knee bend).
2.4 Motion Analysis Testing. To mount the goniometers,
participant’s right and left legs were first fitted with Neoprene and
Velcro thigh and shin segment cuffs. These nonslip cuffs were
custom built with a medial pocket for mounting the goniometer
end blocks, as shown in Fig. 1(b) and 1(c). The Egon was worn on
the right knee and the Fgon was worn on the left knee (Fig. 1(d))
for all participants.
Motion analysis data was collected with an eight camera Vicon
MCam system (Oxford Metrics, UK) using the “Plug-in Gait”
marker set [15] (Fig. 1(d)). BioSI and Vicon systems were triggered simultaneously and sampled at 1000 Hz and 60 Hz, respectively. Right and left knee flexion/extension angles from motion
capture were processed according to Davis et al. [15]. BioSI data
was downsampled to 60 Hz using a spline function in MATLAB
(MathWorks Inc. Natick, MA). Participants performed the following activities:
(1) Static Standing Trial. Participants were asked to stand perfectly still, arms at sides, with eyes opened and looking
straight ahead and feet shoulder width apart.
(2) Free Gait Trial. From a standing position at the start of the
walkway, the participant walked at their preferred comfortable pace until reaching the opposite end of the walkway.
(3) Sit-to-stand Trial. Participants were seated upon an armless,
backless chair of adjustable height with arms folded across
2
Although both DOF were recorded from the Egon we only used the sagittal
plane angles for comparison to the Fgon.
the abdomen, and were instructed to rise from the chair on
cue.
(4) Deep Knee Bend Trial. From a standing position with feet
shoulder width apart, participants were instructed to lower
their center gravity by bending at the knees as much as possible to a crouch position, and then rising again to a standing position.
For each of the locomotion activities, the first trial captured
without marker loss (common for chair rise and deep knee bends)
was selected for data analysis.
2.5 Data Analysis. Fgon angles were compared to the right
knee angles estimated from the Vicon data, and Egon angles were
compared to the left knee angles estimated from Vicon data. The
device calibration curves were first applied to the raw goniometer
data, and then the static standing offsets were removed from both
the Vicon and goniometer signals. The absolute average error
between the two signals was then computed. A second analysis
was performed by first subtracting off the signal means, and then
calculating the absolute errors (signal bias removed). Finally, correlation coefficients for the two signals were computed.
In order to standardize the comparisons, for chair rise trials and
deep knee bends, the signals were compared between the start of
movement and erect standing. For gait trials, the signals were
compared between successive heel strike or toe-off events (of the
same limb), depending on the portion of knee flexion captured
with Vicon.
3
Results
3.1 Goniometer Characteristics. As previously described,
the Egon and Fgon were worn on the right and left knee, respectively, for all participants. Therefore only quadrant 1 of Fig. 2(a)
is relevant for the Egon, and quadrant 3 of Fig. 2(a) for the Fgon.
Differences between goniometers readings and known angles are
also shown in Fig. 2(b). Figure 2(b) shows asymmetric linearity
characteristics for both goniometers (meaning the goniometers do
not have the same degree of linearity throughout positive and negative ranges), but also indicates that our choice of mounting minimized device linearity errors.
3.2 Goniometry Compared to Optical Tracking. As shown
in Table 1, mean absolute errors relative to Vicon for gait and
chair rise were 4.2 deg and 4.7 deg, respectively for the Egon, and
5.1 deg and 8.9 deg, respectively for the Fgon. For deep knee
bends, the Fgon performed slightly better overall, having mean
error of 8.5 deg compared to >11 deg for the Egon. As shown in
Table 2, when signal biases were completely eliminated, errors
for gait and chair rise reduced to 2.3 deg and 3.4 deg, respectively
for the Egon, and 3.3 deg and 4.3 deg, respectively for the Fgon.
Little improvement was observed for deep knee bends, for either
Egon (10.5 deg) or Fgon (8.4 deg), when removing signal bias.
Figure 3(a) shows representative results of a gait trial for Participant #1. Here, the Fgon tracked the knee better than the Egon
(Egon: 4.3 deg; Fgon: 2.2 deg, Table 1), but when removing signal
bias the errors decreased for the Egon to less than the Fgon (Egon:
1.9 deg; Fgon: 2.3 deg, Table 2). Figure 3(b) shows representative
results of a chair rise trial for Participant #2. The Egon tracked the
knee more closely than the Fgon (Egon: 3.0 deg; Fgon: 4.4 deg,
Table 2), but little change was observed when removing signal bias
(Egon: 2.6 deg; Fgon: 4.4 deg, Table 3). For other participants (not
shown in Fig. 3), removing signal bias did reduce errors observed
(Participant #4 for example—see Tables 2 and 3).
Error in predicting knee angles during the deep knee bend trials
was quite high for both goniometer devices. Figure 3(c) shows
representative data of a deep knee bend trials for Participant #6.
The Fgon was able to track knee flexion angles better than the
Egon (Egon: 8.1 deg; Fgon: 3.0 deg, Table 2). This trend did not
084502-2 / Vol. 134, AUGUST 2012
Downloaded From: http://biomechanical.asmedigitalcollection.asme.org/ on 01/18/2015 Terms of Use: http://asme.org/terms
Transactions of the ASME
Table 2 Knee goniometer angle error relative to Vicon knee
angle, after removing signal bias
Egon
Fig. 2 (a) Goniometer linearity test results over 0 to 150 deg
and 0 to 150 deg range. Measured goniometer values are
shown on the vertical axis and reference goniometer values are
shown on the horizontal axis. (b) Difference between measured
goniometer readings and reference goniometer. Flexion ranges
used for each goniometer in the experiments are indicated.
Table 1
angle
Knee goniometer angle error relative to Vicon knee
Egon
Fgon
Subject
Gait
Chair
Bend
Mean
Gait
Chair
Bend
Mean
1
2
3
4
5
6
4.29
2.62
4.19
7.04
6.60
3.34
2.13
3.04
5.78
5.97
3.15
4.83
9.76
5.26
9.93
28.33
6.91
8.08
5.39
3.64
6.63
13.78
5.55
5.42
2.23
8.31
1.99
6.82
4.59
6.71
13.55
4.39
6.96
9.73
6.40
12.24
22.22
8.53
1.78
9.74
5.84
3.04
12.67
7.08
3.58
8.76
5.61
7.33
Mean
4.68
4.15
11.38
6.74
5.11
8.88
8.53
7.50
change when signal bias was removed. Nevertheless, data in
Tables 1 and 2 show each device experienced errors greater than
20 deg for both the Egon (Participant #4) and the Fgon (Participant #1).
Finally, correlation coefficients were very high for all trials and
subjects, ranging between 9.2–1.0. Due to the limited value of
these results, we did not include them in the tables.
4
Discussion
Although wearable sensor systems have great potential for
applications in physical medicine and rehabilitation [8,12,16,17],
Journal of Biomechanical Engineering
Fgon
Subject
Gait
Chair
Bend
Mean
Gait
Chair
Bend
Mean
1
2
3
4
5
6
1.94
2.59
4.63
3.22
4.17
3.52
0.62
2.61
4.26
4.68
1.21
0.79
10.94
6.15
9.06
22.03
7.32
6.60
4.50
3.78
5.99
9.98
4.23
3.64
2.29
5.00
2.02
5.95
3.22
6.95
5.47
4.38
2.09
3.86
1.79
2.05
24.22
9.47
2.60
8.44
5.04
1.56
10.66
6.28
2.24
6.08
3.35
3.52
Mean
3.35
2.36
10.35
5.35
4.24
3.27
8.56
5.35
there is no clear consensus regarding the type(s) of sensors that
provide the best solutions for measuring key outcomes variables
in remote environments. Inertial-based sensors show great promise in a range of applications, from identifying human activities
for monitoring purposes [12,18] to detecting important gait events
critical to assistive technologies for stroke and Parkinson’s disease
[19,20].
Although inertial-measurement units (IMUs) can be used to
measure knee flexion/extension kinematics [21–23], they are not
immune to artifact. deVries et al. [24] report significant magnetic
field artifacts when IMUs are used in close proximity to ferromagnetic structural supports in a motion laboratory. Iron structures are
common in residential apartment buildings and care centers, and
thus pose a threat to the validity of IMU instruments for remote
applications. Furthermore, Kendall and Lemaire [25] report that
ferromagnetic interference with mobility devices, such as walkers,
wheelchairs, orthotics and prosthetic devices significantly
decreases IMU measurement validity. The usability of IMU systems in remote rehabilitation environments may, therefore, be
problematic.
Flexible goniometers are comparatively simple to use, their data
easy to analyze and interpret, and they are not susceptible to ferromagnetic influence. As with any motion capture system, however,
proper mounting and alignment of sensors on the limbs, and static
calibration trials, are clearly still a requirement. Our data support
this fact, but also show that flexible goniometry provides a promising level of accuracy in measuring knee joint kinematics during
ADL relevant to physical medicine and rehabilitation.
4.1 Flexible Goniometry as an Assessment Tool for Knee
Mobility. When applying device calibration curves and a simple
static calibration trial (participant standing erect with knees at full
extension) to correct the joint angles, the Egon had slightly better
performance during gait and chair rise activities compared to the
Fgon (see Table 2). When removing all signal bias, errors were
reduced for both the Egon and Fgon (see Table 3). This data suggest that for locomotor ADL, where peak knee flexion angles are
typically in the 0–110 deg range (which includes chair and stair
activities along with gait [13]) both goniometers can achieve good
levels of accuracy, but that a more sophisticated static calibration
scheme may be required to achieve better levels of accuracy for
these ADLs.
For deep knee bends both goniometers showed a tendency to
under-predict knee flexion angles beyond 110 deg. Here the Fgon
error was lower than the Egon error (see Table 2). However, the
data suggest that device mounting is more critical when testing at
higher flexion angles, probably due to the deformation of the body
segments and mounting cuffs at higher degrees of knee flexion.
This might also explain why the Fgon performed better at higher
flexion angles: the neoprene cuffs were further apart for the Fgon
(see Fig. 1) and; as a result, end block mounts might not have
deformed as much at higher flexion angles. This suggests that
mounting strategy is more relevant when being used for highflexion applications.
AUGUST 2012, Vol. 134 / 084502-3
Downloaded From: http://biomechanical.asmedigitalcollection.asme.org/ on 01/18/2015 Terms of Use: http://asme.org/terms
Fig. 3 Representative trials for study participants. Plots depict a close agreement for gait
and chair rise, and illustrate the higher errors observed for deep knee bend trials. (a) Gait
data for Participant #1. (b) Chair rise data for Participant #2. (c) Deep knee bend data for
Participant #6.
4.2 Flexible Goniometery in Wearable Sensor Applications.
Huddleston et al. [12] used the Penny & GilesTM goniometer with
an activity monitor system [26] to document the range of knee flexion angles for different ADL. Although laboratory validation results
are given as correlation coefficients, rather than absolute errors, the
values (0.97–0.98) are in agreement with our study. Huddleston’s
study also deployed the system remotely for long-term capture, to
test whether “high-flexion” prosthetic knee components (capable of
>150 deg flexion) reflect the mobility needs of normal behaving
adults. They found that maximum knee angles rarely exceeded
120 deg during human activity. As discussed above, our data
suggest that end block mount deformation (soft tissue and/or cuffs)
at extreme flexion angles may be a significant factor in causing this
under-prediction.
Indramohan et al. [8] measured goniometer knee angles for chair
rise, gait, stair ascent/decent, and squatting (similar to our deep
084502-4 / Vol. 134, AUGUST 2012
Downloaded From: http://biomechanical.asmedigitalcollection.asme.org/ on 01/18/2015 Terms of Use: http://asme.org/terms
Transactions of the ASME
knee bend), but they only validated the goniometer data for gait trials captured simultaneously with an optical tracking system (Vicon,
Oxford, UK). The range of absolute error was 3–5 degrees, which
is consistent with our results. Interestingly, they also report that
squat knee angles were much lower (120 deg) than expected
(>150 deg [27]) for this activity. As noted above, the ability of
flexible goniometers to measure extreme knee flexion angles could
be questioned.
In terms of clinical relevance, many lower extremity pathologies can result in significant deviations of knee joint kinematics,
with peak differences greater than 10 deg during walking in
patients with stroke [28] and knee arthritis [29], and differences
up to 20 deg for high-flexion activities (knee bends, crouching,
etc.) in patients with total knee replacement [30]. The errors we
report for both goniometer systems generally suggest these devices could be useful clinically, however our results emphasize that
great care needs to be taken in calibrating the devices for specific
applications and more research is needed on the effects of different mounting strategies.
4.3 Limitations. There were several limitations to our study.
The sample size was relatively small, but similar to other studies
that have evaluated goniometers. We positioned the goniometers
on the medial side of the knee rather than the more traditional
lateral side; this was necessary to avoid interference between the
goniometer cable and knee marker. However, we have found no
studies indicating that medial mounting is more prone to goniometer errors than lateral mounting, and considering our results agree
with other studies, we are confident this is not a major limitation.
Our data suggest the mounting cuffs may have contributed to
the higher errors observed at extreme knee flexion angles.
Although taping the end blocks directly to skin might have
reduced this effect, this approach is less likely to be accepted in
remote sensing applications than the quickly donned and doffed
cuffs that we used. There are also limitations to the Plug-in Gait
marker set that may have contributed to errors observed. Because
this marker set was designed primarily for use in gait studies [15]
errors at higher flexion activities could also be due to uncertainly
in knee angle computations from marker data. However, we
observed no significant abnormalities in calculated angles. Also,
although crosstalk between joint axis angles due to marker misalignment is a common concern, the issue is typically more relevant to the quality of varus/valgus and axial rotation angles,
which are not analyzed in this paper.
Other issues not explored in this paper, but relevant to future
applications outside the laboratory (clinic, community or home
care settings; for example) include the influence of sensor positioning, sensitivity to off-axis motions (such as knee varus-valgus), and the potential for the cuffs to migrate during usage.
Sensor positioning is generally not an issue as trained personnel
are still required to help patients don and doff the devices, and the
ShapeSensor device exhibits low sensitivity to off-axis rotation.
The inner material of the cuffs adhered very well to skin preventing any observable cuff migration, but this would need to be
monitored for long-term wear applications.
5
Conclusions
Our study is the first we know of to compare a fiber-optic goniometer and a strain-gauge goniometer to a gold standard motion
analysis system, for a range of knee activities relevant to rehabilitation. Correlations were very high for both goniometers against
optical tracking methods for all activities studied, indicating that
patterns of movement are faithfully recorded by both devices.
Applications where more knee flexion range >120 deg is required
may need to incorporate different mounting strategies, and a more
sophisticated static calibration protocol. We also conclude from
this preliminary data that a fiber-optic goniometer may be advantageous when higher range of motion testing is required, but more
Journal of Biomechanical Engineering
studies are required to substantiate this finding, particularly when
different mounting strategies are being used.
Acknowledgment
Funding for this study was provided in part by the Natural Sciences and Engineering Research Council of Canada, Canadian
Institutes of Health Research Regional Partners Program, and the
Atlantic Canada Opportunities Agency.
References
[1] Krebs, D. E., Scarborough, D. M., and McGibbon, C. A., 2007, “Functional vs.
Strength Training in Disabled Elderly Outpatients,” Am. J. Phys. Med. Rehabil., 86(2) pp. 93–103.
[2] Alexander, N. B., Gross, M. M., and Medell, J. L., 2001, “Effects of Functional
Ability and Training on Chair-Rise Biomechanics in Older Adults,” J. Gerontol.
Ser. A 56(9), pp. M538–47.
[3] Liu, F., Ohdera, T., Miyamoto, H., 2009, “In Vivo Kinematic Determination of
Total Knee Arthroplasty From Squatting to Standing,” The Knee, 16(2), pp.
116–120.
[4] Ploegmakers, M. J., Ginsel, B., Meijerink, H. J., 2010, “Physical Examination
and In Vivo Kinematics in Two Posterior Cruciate Ligament Retaining Total
Knee Arthroplasty Designs,” The Knee, 17(3), pp. 204–209.
[5] Bonato, P., 2005, “Advances in Wearable Technology and Applications in
Physical Medicine and Rehabilitation,” J. Neuroengineering and Rehabilitation,
2(1).
[6] Dejnabadi, H., Jolles, B. M., and Aminian, K., 2005, “A New Approach to
Accurate Measurement of Uniaxial Joint Angles Based on a Combination of
Accelerometers and Gyroscopes,” IEEE Trans. Biomed. Eng., 52(8), pp.
1478–1484.
[7] Liu, K., Liu, T., and Shibata, K., 2009, “Novel Approach to Ambulatory
Assessment of Human Segmental Orientation on a Wearable Sensor System,”
J. Biomech., 42(16), pp. 2747–2752.
[8] Indramohan, V. P., Valsan, G., and Rowe, P. J., 2009, “Development and Validation of a User-Friendly Data Logger (SUDALS) for Use With Flexible Electrogoniometers to Measure Joint Movement in Clinical Trials,” J. Med. Eng.
Technol., pp. 1–6.
[9] Piriyaprasarth, P., Morris, M. E., and Winter, A., 2008, “The Reliability of
Knee Joint Position Testing Using Electrogoniometry,” BMC Musculoskeletal
Disorders, 9.
[10] van der Linden, M. L., Rowe, P. J., and Nutton, R. W., 2008, “Between-Day
Repeatability of Knee Kinematics during Functional Tasks Recorded Using
Flexible Electrogoniometry,” Gait and Posture, 28(2), pp. 292–296.
[11] Cleffken, B., van Breukelen, G., and Brink, P., 2007, “Digital Goniometric
Measurement of Knee Joint Motion. Evaluation of Usefulness for Research Settings and Clinical Practice,” The Knee, 14(5), pp. 385–389.
[12] Huddleston, J., Alaiti, A., and Goldvasser, D., 2006, “Ambulatory Measurement
of Knee Motion and Physical Activity: Preliminary Evaluation of a Smart Activity Monitor,” J. Neuroengineering and Rehabilitation, 3.
[13] Rowe, P. J., Myles, C. M., and Walker, C., 2000, “Knee Joint Kinematics in
Gait and Other Functional Activities Measured Using Flexible Electrogoniometry: How Much Knee Motion is Sufficient for Normal Daily Life?” Gait and
Posture, 12(12), pp. 143–155.
[14] Bell, J. A., and Stigant, M., 2007, “Development of a Fibre Optic Goniometer
System to Measure Lumbar and Hip Movement to Detect Activities and their
Lumbar Postures,” J. Med. Eng. Technol., 31(5), pp. 361–366.
[15] Davis, R., Ounpuu, S., and Tyburski, D., 1991, “A Gait Analysis Data Collection and Reduction Technique,” Hum. Mov. Sci., 10, pp. 575–587.
[16] Conklyn, D., Stough, D., and Novak, E., 2010, “A Home-Based Walking Program Using Rhythmic Auditory Stimulation Improves Gait Performance in
Patients With Multiple Sclerosis: A Pilot Study,” Neurorehabilitation and
Neural Repair, 24(9), pp. 835–842.
[17] Salarian, A., Russmann, H., and Vingerhoets, F. J., 2004, “Gait Assessment in
Parkinson’s Disease: Toward an Ambulatory System for Long-Term Monitoring,” IEEE Trans. Biomed. Eng., 51(8), pp. 1434–1443.
[18] Sherrill, D. M., Moy, M. L., and Reilly, J. J., 2005, “Using Hierarchical Clustering Methods to Classify Motor Activities of COPD Patients From Wearable
Sensor Data,” J. Neuroengineering and Rehabilitation, 2.
[19] Bachlin, M., Plotnik, M., and Roggen, D., 2010, “Wearable Assistant for Parkinson’s Disease Patients With the Freezing of Gait Symptom,” IEEE Trans. Inf.
Technol. Biomed., 14(2), pp. 436–446.
[20] Lau, H., and Tong, K., 2008, “The Reliability of Using Accelerometer and
Gyroscope for Gait Event Identification on Persons With Dropped Foot,” Gait
and Posture, 27(2), pp. 248–257.
[21] Favre, J., Jolles, B. M., and Aissaoui, R., 2008, “Ambulatory Measurement of
3D Knee Joint Angle,” J. Biomech., 41(5), pp. 1029–1035.
[22] Kun, L., Inoue, Y., and Shibata, K., 2011, “Ambulatory Estimation of KneeJoint Kinematics in Anatomical Coordinate System Using Accelerometers and
Magnetometers,” IEEE Trans. Biomed. Eng., 58(2), pp. 435–442.
[23] Cooper, G., Sheret, I., and McMillan, L., 2009, “Inertial Sensor-Based
Knee Flexion/Extension Angle Estimation,” J. Biomech., 42(16), pp.
2678–2685.
AUGUST 2012, Vol. 134 / 084502-5
Downloaded From: http://biomechanical.asmedigitalcollection.asme.org/ on 01/18/2015 Terms of Use: http://asme.org/terms
[24] de Vries, W. H., Veeger, H. E., and Baten, C. T., 2009, “Magnetic Distortion in
Motion Labs, Implications for Validating Inertial Magnetic Sensors,” Gait and
Posture, 29(4), pp. 535–541.
[25] Kendell, C., and Lemaire, E. D., 2009, “Effect of Mobility Devices on Orientation
Sensors that Contain Magnetometers,” J. Rehabil. Res. Dev., 46(7), pp. 957–962.
[26] Zhang, K., Werner, P., and Sun, M., 2003, “Measurement of Human Daily
Physical Activity,” Obes. Res., 11(1), pp. 33–40.
[27] Hemmerich, A., Brown, H., and Smith, S., 2006, “Hip, Knee, and Ankle Kinematics of High Range of Motion Activities of Daily Living,” J. Orthop. Res.,
24(4), pp. 770–781.
[28] Oken, O., Yavuzer, G., and Ergocen, S., 2008, “Repeatability and Variation of
Quantitative Gait Data in Subgroups of Patients With Stroke,” Gait and Posture,
27(3), pp. 506–511.
[29] Maly, M. R., Costigan, P. A., and Olney, S. J., 2006, “Role of Knee Kinematics
and Kinetics on Performance and Disability in People With Medial Compartment Knee Osteoarthritis,” Clin. Biomech. (Bristol, Avon), 21(10), pp.
1051–1059.
[30] Moynihan, A. L., Varadarajan, K. M., and Hanson, G. R., 2010, “In Vivo Knee
Kinematics During High Flexion After a Posterior-Substituting Total Knee
Arthroplasty,” Int. Orthop., 34(4), pp. 497–503.
084502-6 / Vol. 134, AUGUST 2012
Downloaded From: http://biomechanical.asmedigitalcollection.asme.org/ on 01/18/2015 Terms of Use: http://asme.org/terms
Transactions of the ASME