Content uploaded by Matthew Wyon
Author content
All content in this area was uploaded by Matthew Wyon
Content may be subject to copyright.
Surface electromyography (sEMG) has been used in dance medicine
research since the 1970s, but normalization procedures are not con-
sistently employed in the field. The purpose of this project was to
develop a portable anchored dynamometer (PAD) specifically for
dance-related research. Due to the limited studies in the dance
research literature using normalization procedures for sEMG data, a
review of the procedures used in the exercise science literature was
conducted. A portable anchored dynamometer was then developed
and tested with dancers, using methods validated in previous litera-
ture. We collected sEMG maximum voluntary isometric contrac-
tions (MVIC, mV) from 10 female dancers (mean age 31.0 ± 15 yrs,
mean height 163 ± 7.6 cm, mean weight 57.6 ± 6.9 kg, and 17.0 ±
13.9 yrs of training in ballet and/or modern dance) over three trials
(5 sec each) for eight muscles bilaterally (quadriceps, tibialis ante-
rior, abductor hallucis, gastrocnemius, hamstrings, gluteus max-
imus, erector spinae, and rectus abdominus). Consistency of data
and feedback from dancers suggest that this dance-specific portable
anchored dynamometer is effective for future sEMG studies in
dance research. Medical Problems of Performing Artists 2011;
26(4):185–194.
Electromyography has been in use to study dancers’
muscle activity since the late 1970s,1with researchers
examining a broad range of dance movements including
pliés,2-6 relevés,7–9 degagés,10,11 développés,12,13 grand battement,1,14
forward stepping,15,16 and elevation work.17–19 In a recent
review of literature, Krasnow et al.20 stated that in the 21
dance research studies using surface electromyography
(sEMG) that they considered, less than half utilized any
method of data normalization in order to enable sEMG
amplitude comparisons across subjects or over time. The
studies that did not collect data for normalization only
assessed onset times of muscle activation in a given single
testing session and therefore did not require normalization
procedures. The studies that collected normalization data
used a variety of methods including average rectified values,
manual resistance testing, and use of isokinetic equipment.
To date, no dance research studies have used hand-held
dynamometers or dynamometer anchoring systems.
While many research questions do not require the assess-
ment of amplitudes, it is imperative to consider a method for
the collection of sEMG normalization data to provide clearer
insight into muscle activation patterns in dancers, a special-
ized subset of the physically active population. Therefore, the
project was in two stages: (1) the first stage examined the
existing exercise science literature using dynamometer sEMG
collection procedures to determine the potential for this pro-
cedure for dance research; (2) the second stage aimed to
develop a portable anchored dynamometer (PAD) that could
be easily constructed and implemented for dance-specific
EMG research, and to validate this system based on previous
methodology in exercise science research and pilot studies
with dancers.
STAGE 1: REVIEW OF LITERATURE
Dynamometers for sEMG Normalization
One preferred normalization procedure in sports and exer-
cise science literature is the data collection of maximum vol-
untary contractions (MVCs) or maximum voluntary isomet-
ric contractions (MVICs), using a percentage of maximum
values to compare subjects.21–25 Burden26 emphasized the
importance of normalizing sEMG data if comparisons were
made between different muscles and different individuals. In
December 2011 185
Articles
Development of a Portable Anchored Dynamometer
for Collection of Maximal Voluntary Isometric
Contractions in Biomechanics Research on Dancers
Donna Krasnow, MS, Jatin P. Ambegaonkar, PhD, Shane Stecyk, PhD, M. Virginia Wilmerding, PhD,
Matthew Wyon, PhD, and Yiannis Koutedakis, PhD
Ms. Krasnow is Professor, Department of Dance, Faculty of Fine Arts, York
University, Toronto, Ontario, Canada, and Lecturer, Department of Kine-
siology, California State University, Northridge, California, USA; Dr.
Ambegaonkar is Director, Sports Medicine Assessment Research and Test-
ing Laboratory, Director, GMU Performing Arts Medicine, George Mason
University, Manassas, Virginia, USA; Dr. Stecyk is Director, Athletic
Training Education Program, Department of Kinesiology, California State
University, Northridge, California, USA; Dr. Wilmerding is Adjunct Pro-
fessor, Department of Dance, and Assistant Research Professor, the Depart-
ment of Health, Exercise & Sports Sciences, University of New Mexico,
Albuquerque, New Mexico, USA; Dr. Wyon and Dr. Koutedakis are Pro-
fessors with the Research Centre for Sport, Exercise and Performance, School
of Sport, Performing Arts and Leisure, University of Wolverhampton, Wal-
sall, UK; Dr. Wyon is also with the Department of Dance, Artez, Arnhem,
The Netherlands; and Dr. Koutedakis is also with the Department of Exer-
cise Sciences, University of Thessaly, Trikala, Greece.
Address correspondence to: Donna Krasnow, MS, Professor, Department of
Dance, Faculty of Fine Arts, Accolade Building East 313, York University,
4700 Keele Street, Toronto, ON M3J 1P3, Canada. Tel 416-736-5137
x22130, fax 416-736-5743. dkrasnow1@aol.com.
© Copyright 2011 Science & Medicine, Inc. www.sciandmed.com/mppa
his review of the literature over the past 25 years, he assessed
eight normalization methods and concluded the following:
1) sEMG data from MVCs and MVICs are equally reliable, and
further, these values are as useful as using the dynamic maxi-
mum of the movement trial under investigation;
2) using either submaximal isometric values or maximal isometric
values at an arbitrary joint angle in mid-range is acceptable, as
both have good reliability;
3) evidence does not support the need to match the specific joint
angles during MVIC collection or joint ranges during MVC col-
lection to the movement trials in order to have reliable com-
parison data;
4) dynamic MVCs should be used only if it can be determined that
the task being used for the MVC collection can elicit maximal
contractions in all of the muscles under investigation; and
5) in conclusion, use of sEMG data from an MIVC is the recom-
mended method as a normalization reference value.
Standardized Equipment
The traditional method of collecting MVIC data has been
the use of standardized equipment designed for muscle test-
ing. However, there are pragmatic problems with the use of
equipment, such as the Biodex and Kin Com systems for
MVC and MVIC data collection, in dance research. First, the
available equipment does not allow for much flexibility in
terms of body positioning for data collection on a given
muscle. Second, if multiple electrodes are placed on the
body, it can be difficult to place the subject on the equip-
ment in various positions without disrupting some of the
electrode placements. Third, for dance researchers, access to
this equipment, particularly in a dance-suitable space, can
prove challenging. Finally, dancers often find using this type
of equipment so atypical of their normal training process
that it is questionable if they are able to elicit maximal or reli-
able levels of muscle activation.27
Hand-held Dynamometer
In seeking alternatives to the standardized equipment,
researchers in sport and exercise science have explored the
use of devices known as hand-held dynamometers
(HHD).28–34 The dynamometer measures the muscle force
that a subject can elicit, while a trained tester provides resist-
ance so that the subject can achieve high levels of muscle
contraction.
In an early study by Agre et al.,28 the HHD was found to
be reliable for upper extremity testing but not for lower
extremity testing, due to the lack of stability of the tester. The
variation coefficient of the methodology error (CV) was
between 5.1% and 8.3% for all upper extremity muscle tests,
which the authors considered acceptable reliability for clini-
cal muscle strength testing, but the lower extremity values
ranged from 11.3% to 17.8%, resulting in poor reliability.
Andrews et al.29 assessed an HHD device, examining eight
upper extremity and five lower extremity movements, and
used the results to determine normative values for popula-
tions 50 to 79 years old. The researchers concluded that
while the testing methodology is reliable, the training, expe-
rience, and strength of the tester are important factors.
Wikholm and Bohannon34 had three testers measure two
upper and three lower extremity muscles for 27 subjects.
They selected three testers with measurably different strength
levels and muscles with different maximum force produc-
tions. They found that there was considerable variability in
results. As the tested muscles increased in force production,
the interrater intra-class coefficients (ICCs) decreased in mag-
nitude (0.932 to 0.226). Similar results were seen for
intrarater/intrasession ICCs. They concluded that these
results were most likely due to differences in individual exam-
iner’s strength levels and the subsequent resistance they were
able to offer to the subjects during testing.
Bohannon30 used one tester to assess 6 upper extremity
and 4 lower extremity muscles for 106 men and 125 women
and confirmed that reliable measurements could be obtained
using an HHD. He observed, however, that the tester must be
strong enough to provide sufficient resistance to the subject’s
efforts, and the technique must be clearly defined, systematic,
and consistent. Kelln et al.32 tested 11 lower extremity muscles
of 20 subjects, using three testers on 2 separate days, with the
following results: intratester ICCs ranged between 0.77 to
0.97 with standard error of measurements (SEM) range of
0.01 to 0.44 kg. Mean intertester ICC range was 0.65 to 0.87
with SEM range of 0.11 to 1.05 kg. Mean intersession ICC
range was 0.62 to 0.92 with SEM range of 0.01 to 0.83 kg.
Similar to Bohannon,30 Kelln et al. suggested that the limita-
tion of such a hand-held device was in attempting to test
movements in which the subject could overpower the tester.
Bohannon31 reviewed 13 published articles in the litera-
ture using HHDs to determine the responsiveness of the test-
ing device over time. Using effect size as the measure of
responsiveness, he concluded that HHD could detect
changes in limb strength due to interventions. Thorborg et
al.33 used the HHD to assess hip abduction, hip adduction,
hip external rotation, hip internal rotation, hip flexion, and
hip extension, all of which would be highly applicable to
dance research. In test-retest trials, they examined measure-
ment variability and found highly reliable results, with meas-
urement variation between 3% and 12% for the various mus-
cles between sessions. It should be noted that the tester did
enthusiastic cueing during these data collection sessions,
which was seen as an important component in obtaining reli-
able results.
Portable Anchored Dynamometers
In order to mitigate the problem of the subject overpowering
the tester, and to provide more consistent positioning of the
resistance, researchers have designed portable anchoring sys-
tems, using solid apparatus and belts as the resistance modal-
ity. For example, Kramer, Vaz, and Vandervoort35 used a
combination of HHD and belt resistance, and they found
that this method required less strength on the part of the
examiner and greater stabilization of the subject and was pre-
ferred by the majority of subjects. Similarly, Bolgla and Uhl23
186 Medical Problems of Performing Artists
compared the reliability of three normalization methods for
testing hip abductor strength—maximum voluntary isometric
contraction (MVIC), mean dynamic activity, and peak
dynamic activity—all using a table and resistance belts. The
researchers concluded that this MVIC collection method
provided the highest level of reliability. They further com-
mented that factors that impacted reliability were body posi-
tioning, verbal encouragement, and task familiarization.
Nadler et al.36 designed a portable dynamometer anchor-
ing station that measured the strength of the hip extensors
and abductors. Ten subjects were tested twice, 2 weeks apart,
with the evaluators blinded. They computed the ICCs for
both maximum (ICC 1,1) and average (ICC 1,3) strength,
which ranged from 0.94 to 0.98. The average CV (coeffi-
cients of variation) for maximal abduction strength was
4.77% and for average abduction strength was 4%. Average
CV for maximal extension strength was 8.06% and for aver-
age extension strength was 7.83%. Thus they concluded that
this method of collection was highly reliable and particularly
useful for testing powerful muscles that might not be easily
assessed using an HHD device.
Finally, Scott et al.37 compared the inter- and intra-rater reli-
ability of a portable anchored dynamometer (PAD) to an
HHD, assessing hip abduction, extension, and flexion, using
two testers with a 1-hour break between sessions for the sub-
ject. Interrater ICCs of average peak strength ranged from 0.84
to 0.92 (hip flexors), 0.69 to 0.88 (hip abductors), and 0.56 to
0.80 (hip extensors). Intra-rater ICCs ranged from 0.59 to 0.89
for tester A and from 0.72 to 0.89 for tester B using the PAD,
and from 0.67 to 0.81 for the HHD across muscle groups. The
PAD was highly reliable for hip flexion and abduction,
whereas the HHD was more reliable for hip extension. They
concluded that both systems yielded reliable test results.
Summary of Literature Review
In summarizing the literature about sEMG normalization
procedures, MVICs are highly reliable for normalization of
sEMG data collected during movement trials.23,26 Standard-
ized equipment is problematic for dancers,27 but dynamome-
ters can reliably determine muscle strength for the purposes
of muscle testing,28,29,32,34 and the limitations due to tester
strength and variability can be overcome using PADs.23,35,36
Other recommendations include familiarity with procedures
and enthusiastic cueing.33
STAGE 2: NEED FOR DANCE-SPECIFIC PAD
To our knowledge, no PAD has been devised for dance med-
icine and science research. The anchoring systems presented
in the sports and exercise science literature do not always
replicate the typical movement patterns during dance move-
ment. For example, in sports and exercise science, the PADs
typically test the gluteus maximus in the seated position with
the subject pressing downward; however dancers are more
familiar with use of this muscle in movements such as the
arabesque, where the dancer is in a one-legged stance with the
free leg behind the body in hip and knee extension. There-
fore, we decided to address this absence of a dance-specific
dynamometer system by designing a customized PAD.
The second stage of the project was the development of a
portable anchored dynamometer (PAD) that can be easily
constructed and implemented for dance-specific EMG
research and the validation of this system based on previous
methodology in exercise science research and pilot studies
with dancers. The three steps were:
1) First, we chose to adapt a previously reported PAD, by modifying
the body positioning, to create a similarity to dance movements.
2) Second, we replicated procedures that are reported to result in
reliable results in previous literature.
3) Third, we tested this PAD on dancers, asking them for subjective
feedback on comfort and effort levels when using this apparatus.
METHODOLOGY
Subjects
Ten trained female dancers (mean age 31.0 ± 15 yrs, mean
height 163 ± 7.6 cm, mean weight 57.6 ± 6.9 kg, and 17.0 ±
13.9 yrs of training in ballet and/or modern dance) partici-
pated in this study. Subjects were only included if they had
no injuries that might impede successful completion of the
tasks. Dancers were volunteers from local colleges, universi-
ties, professional dance schools, and dance studios and were
recruited through announcements in dance classes, listings
in dance newsletters, and emails. Dancers were volunteers
from local colleges, universities, professional dance schools,
and dance studios and were recruited through announce-
ments in dance classes, listings in dance newsletters, and
emails. Dancers were intentionally selected with a broad
range of demographics, due to the potential variable subject
pool for future research. All test procedures were described
and explained to the subjects prior to preparation for testing
and data collection, whereupon they signed an informed con-
sent form. All procedures were approved by the university
institutional review board.
General Approach
The method first proposed by Nadler et al.36 was modified
for this study. The PAD incorporates several of the positive
variables described in previous research,23,31,33,35,36 including
a combination of a table and resistance belts for stability,
body positioning that is familiar to the dancer, practice trials,
and enthusiastic cueing during collections.
The Instrument
The apparatus components can be seen in Figure 1. The
system consisted of the following equipment: (1) a padded
treatment table; (2) a lightly padded removable back support
that was mounted on the table for the seated work and
adjusted so that the subject’s knees reached the end of the
table in the seated position; (3) an adjustable, padded board
December 2011 187
with clamps (Irwin®Quick-Grip® Clamps; Irwin Tools,
Huntersville, NC, www.irwin.com) that can be attached onto
the table legs and adjusted to adapt to the height and leg
length of the subject; (4) straps for stabilizing the subject for
the MVIC tests; and (5) a 6-inch diameter foam roller to assist
with knee flexion in some of the electrode placements and
some of the MVIC data collection.
Electrode Placement
All subjects wore sports bras and spandex shorts during the
testing and completed all trials in bare feet. Surface elec-
trodes (DE 2.3, Myomonitor Single Differential Ag elec-
trodes, skin contact size 10 1 mm, center-to-center distance
of 10 mm) (Delsys Inc., Boston, MA; www.delsys.com) were
applied over the skin after it was prepped with alcohol. Elec-
trodes placements were based on the SENIAM (Surface Elec-
troMyoGraphy for the Non-Invasive Assessment of Muscles)
Project standards (http://www.seniam.org/). The electrodes
were placed on the body in the following order:
Supine: quadriceps (QA), tibialis anterior (TA), abductor hallucis
(AH),
Prone: gastrocnemius (GA), hamstrings (HA), gluteus maximus
(GM), erector spinae (ES), and
Standing: rectus abdominus (AB).
This order was selected to require minimal movement
during the electrode placements, so that electrodes would
not be disturbed.
All sEMG data were collected using a combination of a 16-
channel Myomonitor IV wireless transmitter (Delsys Inc.,
Boston, MA) with an operating range of 25 to 350 m, pre-
amplifier gain 1000 V/V with a frequency bandwidth of 20
to 450 Hz, a common mode rejection ratio of 92 dBmin at 60
Hz, and an input impedance >1015 Ω//0.2 pF, and the
Vicon Nexus 1.416 system (Centennial, CO, USA). The elec-
trode wires were wrapped around the Myomonitor belt to
eliminate excess wiring that might interfere with movement,
and absence of crosstalk was confirmed.
Data Collection Protocol
MVICs were collected in the order listed under Testing Pro-
tocol and Collection Order (see below). Note that data for
AB and ES were collected bilaterally, i.e., right and left sides
were recorded at the same time. Data for AH, GM, QA, HA,
and TA were collected unilaterally, but alternating right and
left, always starting with the right side for consistency. GA
MVIC data were collected with all trials on the right side,
then all trials on the left side, due to the complexity of
moving the stabilizing straps. Joint angles for muscle testing
were determined as per other previous studies in the litera-
ture using either HHD or PAD.30,34 Only one muscle (AH)
required an investigator tester to provide manual resistance.
Due to the small force provided by the AH, the possibility of
a subject overpowering the investigator was ruled out. Still, to
reduce inter-tester variability, the same investigator provided
the resistance for all subjects.
For supine position data collection, the Myomonitor belt
was held off the subject’s body; for prone position data col-
lection, the wireless transmitter was not in the belt, but the
belt was attached to the subject; for seated position data col-
lection, the belt was again held just off the subject’s body.
Testing Protocol and Collection Order
After electrode placement, the subject was given 15 minutes
for a general warm-up. After warm-up, the investigator exam-
ined the electrodes to ensure that none had moved or dis-
lodged. Prior to MVIC collection for each muscle, the sub-
ject was given practice trials until she informed the
investigator that she was familiar with the procedure. After
the practice trials, the subject performed three MVICs using
the “make test” for each muscle,23,29,30 with 30-sec rest
between collections. For the ‘‘make test,’’ subjects generated
maximum muscle force over a 2-sec period and held the
maximum contraction for a 5-sec period. The principal
investigator provided enthusiastic verbal encouragement
during all data collections.33 The specific positioning for
188 Medical Problems of Performing Artists
FIGURE 1. Components of the PAD, showing positioning of the dancer and electrode placement during testing of the eight muscles: testing
of the abdominals (panel A) is shown here, and others (panels B–H) are on the facing page.
A
December 2011 189
FIGURE 1 (cont.). Components of the PAD, showing positioning of the dancer and electrode placement during muscle testing; B, right gas-
trocnemius; C, left abductor hallucis; D, erector spinae; E, left gluteus maximus; F, left quadriceps; G, left hamstrings; H, left tibialis anterior.
B
C
D E
FGH
each muscle can be generally seen in Figure 1 and is
described in detail below:
Supine
1. Spine flexion, MVIC for abdominals (AB): The subject was
supine on the table in a hook lying position, with the toes at the
edge of the table’s end (hips flexed to 45° degrees and knees
flexed to 90°). Arms were placed at the sides of the body, and the
strap crossed the chest just below both clavicles and over the
humeral heads. A second strap was placed over the distal femurs,
just superior to the patellae, and attached to the end of the table,
parallel to the tibias. The subject attempted to flex the spine (i.e.,
to curl the shoulders and knees together, while performing a pos-
terior pelvic tilt). See Figure 1A.
2. Ankle plantar flexion, MVIC for gastrocnemius (GA): The sub-
ject was supine on the table. Two straps were placed around the
distal metatarsal heads, and each strap was then placed over the
acromioclavicular joint of each shoulder. Straps were initially
tightened with the ankle in dorsiflexion so that the slack was
taken up during the muscle contraction, and the ankle was at
90° for the MVIC. A 6-inch foam roller was placed under the
knee, creating approximately 30° of knee flexion. The subject
pressed against the straps, attempting to point the foot (i.e., plan-
tar flex the ankle). Subjects were allowed to wear ballet slippers
if they so chose. See Figure 1B.
3. Big toe (hallux) abduction, MVIC for abductor hallucis (AH):
The subject was supine on the table with a small pillow or folded
towel under the head, and a foam roller under the knees for
support. The subject began in full active dorsiflexion, spread the
toes, and then abducted the hallux against the hand of the inves-
tigator. The investigator stabilized the subject’s heel with the
other hand. (Note that some inversion may also occur and is
acceptable.) See Figure 1C.
Prone
4. Spine extension, MVIC for erector spinae (ES): The subject was
prone on the table. One strap was placed over the scapulae and
thoracic spine at the level of the axilla. A second strap was placed
at the posterior distal femurs, superior to the knee joints. The
arms were folded, hands placed under the forehead, elbows out
to the side. The subject attempted to extend the spine (lift the
upper torso) off the table. The arms lifted off the table, while the
hands remained in contact with the forehead. Dancers were
instructed to raise their torso off of the table with the greatest
effort possible, and they could use the lower extremities as they
saw fit. See Figure 1D.
5. Hip extension, MVIC for gluteus maximus (GM): The subject
was prone on the table. One strap was placed over the scapulae
and thoracic spine at the level of the axilla. A second strap was
placed at the posterior distal femurs, superior to the knee joints.
A third strap was placed just above the posterior superior iliac
spines (PSISs). The arms were folded, hands placed under the
forehead, elbows out to the side. The subject extended one hip
with maximal effort. See Figure 1E.
Seated
6. Knee extension—MVIC for quadriceps (QA): The subject sat
off the end of the table with hips and knees flexed to approxi-
mately 90°. The upper thighs were stabilized to the table by a
strap placed at the mid-femurs. The trunk was stabilized to the
back support by a strap across the upper trunk just below the
axilla, arms relaxed at the sides of the body. The anchoring
system was attached by clamps to the table legs at level of sub-
ject’s lower legs, anterior to the tibia. The subject extended one
knee with maximal effort. See Figure 1F.
7. Knee flexion, MVIC for hamstring (HA): The subject sat off
the end of the table with hips and knees flexed to approximately
90°. The upper thighs were stabilized to the table by a strap
placed at the mid-femurs. The trunk was stabilized to the back
support by a strap across the upper trunk just below the axilla,
arms relaxed at the sides of the body. The anchoring system was
attached by clamps to the table legs at the level of the subject’s
lower legs, posterior to the tibia. The subject flexed one knee
with maximal effort. See Figure 1G.
8. Ankle dorsiflexion, MVIC for tibialis anterior (TA): The sub-
ject sat off the end of the table with hips and knees flexed to
approximately 90°. The upper thighs were stabilized to the table
by a strap placed at the mid-femurs. The trunk was stabilized to
the back support by a strap across the upper trunk just below the
axilla, arms relaxed at the sides of the body. The anchoring
system was attached by clamps to the table legs at the level of the
subject’s foot, anterior to the foot, resting on the metatarsals.
The subject began with the ankle in approximately 35° of plan-
tar flexion and dorsiflexed the ankle with maximal effort, which
caused the foot to push against the anchoring system with the
ankle at 90°. See Figure 1H.
Data Processing
All sEMG data were processed using Visual 3D (C-Motion
Inc. Germantown, MD). The sEMG signals were first
processed with a band pass filter from 10 to 450 Hz, using a
fourth-order, zero-lag Butterworth filter. The sEMG signals
for the MVICs were then smoothed using a 99-ms-wide RMS
time window to obtain steady-state results. Three trials for
each muscle were ensemble averaged to obtain one compos-
ite representative trial for each muscle using a customized
pipeline.
RESULTS
The purpose of developing the PAD described in this article
was to design a normalization procedure for sEMG data col-
lection for dance-related research. In this context, the system
needed to be portable for use in dance spaces, to be modified
in terms of body positioning for dancers, and to provide con-
sistent and reliable results.
All subjects reported that the PAD was comfortable yet
challenging. Subjects also indicated that they provided their
maximal effort and were pleased that the testing was per-
formed using dance-specific positions. Figure 2 represents a
single representative MVIC data collection trial for the left
gastrocnemius of one exemplar subject. The figure exhibits
raw data for all eight muscles on the left side during this trial.
The gastrocnemius graph in Figure 2 clearly demonstrates a
specific onset and activity above baseline for this muscle.
Other active muscles in this trial include abdominals, erector
spinae, and hamstrings. This muscle activation pattern is
what would be expected for a gastrocnemius trial, with the
190 Medical Problems of Performing Artists
December 2011 191
muscles contributing to trunk stabilization and knee flexion
active during this MVIC data collection.
Figure 3 represents the conversion of raw data to filtered
data for four of the muscles seen in Figure 2 during the
MVIC data collection for the left gastrocnemius. It can be
seen that the gastrocnemius and hamstrings are both active
in this trial, whereas the tibialis anterior and abductor hallu-
cis are unchanged relative to baseline. These results are con-
sistent with what would be expected in dance trials for the
gastrocnemius muscle. These graphs are representative of the
graphs for the muscles tested in these pilot studies, with clear
bursts of activation for the target muscle, supporting activity
in muscles contributing to stabilization, and little or no activ-
ity in remaining muscles.
Figure 4 represents all three MVIC trials for the left gas-
trocnemius for the same subject, with the bolded line being
the average of the three trials. Again, this graph is represen-
tative of the three-trial and average graphs for the tested mus-
cles in the pilot study. The individual trial lines and the
bolded average line are similar to results found in previous
FIGURE 2. Left gastrocnemius trial: Raw data from sEMG recordings for AB (abdominals), AH (abductor hallucis), ES (erector spinae), GA
(gastrocnemius), GM (gluteus maximus), HA (hamstrings), QA (quadriceps), and TA (tibialis anterior) muscles. X-axis is in seconds, Y-axis is
in millivolts.
AB
AH
ES
GA
GM
HA
QA
TA
192 Medical Problems of Performing Artists
literature. They demonstrate the general consistency of our
PAD across the three trials, as well as within each trial.
Although we did not do repeat trials in a second testing ses-
sion due to time constraints, we were satisfied that these
results suggest consistency with this procedure.
DISCUSSION
This project was undertaken in two stages: (1) The first stage
was to examine the existing exercise science literature using
dynamometer sEMG collection procedures to determine the
potential for this procedure for dance research. (2) The
second stage was the development of a portable anchored
dynamometer (PAD) that can be easily constructed and
implemented for dance-specific EMG research, and the vali-
dation of this system based on previous methodology in exer-
cise science research and pilot studies with dancers. The lack
of normalization procedures in the dance science literature
caused us to examine the exercise science research for back-
ground in developing our system.
While the exercise science background provided methods
that have been tested and found reliable and consistent, they
also presented barriers to use with dancers. One constraint
was the issue of equipment that would be too unwieldy to
move easily in and out of dance spaces. If testing is to be done
on dance movement, the space must be appropriate for
dance in terms of floor surface, amount of space, and avail-
ability of dance apparatus such as barres and sound systems.
Therefore, we needed to design a PAD that was easy to
assemble quickly and lightweight. A second issue was the
need to modify body positioning that would be familiar to
dancers, allowing them to achieve high levels of muscle acti-
vation in a known context.
Although we sought volunteers from a wide range of
dance schools and studios and university dance programs,
only one male volunteered for the study. Therefore, we
decided to conduct the investigation on female volunteers
only. First, using only one male might be potentially prob-
lematic, if he were atypical, and second, the development of
the PAD is for use in a larger study that will be conducted
exclusively on female dancers. We would recommend test-
ing the PAD on male subjects at a future time, so that it has
broader applications for a more general dance population.
Although there were only 10 subjects in this investigation,
small subject pools are not uncommon in research on
dynamometers and PADs.23,28,36 Given the consistency of
the results and feedback from the subjects, we are confident
that the system is reliable for an adult female dance popu-
lation, and larger numbers are not necessary to validate
these results.
FIGURE 3. Left gastrocnemius trial: Raw (left) and filtered (right) data from sEMG recordings for GA (gastrocnemius), HA (hamstrings), TA
(tibialis anterior), and AH (abductor hallucis). X-axis is in seconds, Y-axis is in millivolts.
GA
TA
HA
AH
In developing the PAD, we incorporated methods
described and found reliable in the literature, but also con-
sidered the weight and assembly ease of the apparatus and
dancers’ comments about comfort and best positioning for
eliciting maximal effort. The PAD described in this article
can be stored in a closet space and assembled in 5 minutes.
It can be moved easily by two people of moderate strength
levels. The verbal feedback from the subjects suggested that
they felt comfortable in the positions described above, and
they were confident that they were contributing their best
efforts. Additionally, Figures 2 to 4 depicted above are typical
of the data collection from the 10 subjects for all muscles
tested and suggest that the dancers were capable of eliciting
consistent and maximal efforts in the positions selected
through this entire testing of the PAD. The results and the
dancers’ responses support the use of this apparatus for
MVIC data collection on dancers.
CONCLUSION
sEMG is a useful tool for dance medicine and science
research, and as the technology improves, it will be increas-
ingly beneficial in the understanding of dance biomechan-
ics. It is essential that dance medicine and science
researchers begin to understand methods of normalization
of sEMG data across subjects and across time, so that
researchers can conduct studies across subjects, days, groups,
and locations. The dance-specific PAD described in this arti-
cle builds upon methods described in previous literature
and provides a reasonably inexpensive, practical way to col-
lect MVICs in this population. We acknowledge that the
reliability of this system is currently presumed for single-ses-
sion, single-tester studies. Thus, we recommend that future
testing of this PAD include multiple-session, multiple-tester
testing. The PAD developed, modified for dancers, and
tested in this study provides a standardized procedure that
dance researchers can use for future research involving the
use of sEMG.
ACKNOWLEDGMENTS
We would like to thank our subjects for their participation, Lilia Kibarska,
the dancer in the photographs, and Dr. Lori Bolgla for her assistance in
preparing this article.
REFERENCES
1. Ryman R, Ranney D. A preliminary investigation of two variations of
the grand battement devant. Dance Res J 1978/79;11(1/2):2–11.
2. Clippinger-Robertson KS, Hutton RS, et al. Mechanical and anatomi-
cal factors relating to the incidence and etiology of patellofemoral pain
in dancers. In: Shell CG, ed. The Dancer as Athlete (1984 Olympic
Scientific Congress Proceedings, Vol 8). Champaign, IL: Human
Kinetics; 1986: pp53–72.
3. Couillandre A, Lewton-Brain P, Portero P. Exploring the effects of
kinesiological awareness and mental imagery on movement intention
in the performance of demi-plié. J Dance Med Sci 2008;12(3):91–98.
4. Ferland G, Gardener P, Lèbe-Néron RM. Analysis of the electromyo-
graphic profile of the rectus femoris and biceps femoris during the
demi-plié in dancers [abstract]. Med Sci Sport Exerc 1983;15:159.
5. Trepman E, Gellman RE, Micheli LJ, De Luca CJ. Electromyographic
analysis of grand-plié in ballet and modern dancers. Med Sci Sport
Exerc 1998;30(12):1708–1720.
6. Trepman E, Gellman RE, Solomon R, et al. Electromyographic analy-
sis standing posture and demi-plié in ballet and modern dancers. Med
Sci Sport Exerc 1994;26(6):771–782.
7. Bartolomeo O De, Sette MM, Sloten JV, Albisetti W. Electromyo-
graphic study on the biomechanics of the lower limb during the exe-
cution of technical fundamentals of dance: the releve [poster]. J Bio-
mech 2007;40(2):S789.
8. Massó N, Germán A, Rey F, et al. Study of muscle activity during
relevé in first and sixth positions. J Dance Med Sci 2004;8(4):101–107.
9. Yoshida M, Kuno-Mizumura M. The changes of EMG activity with
fatigue during heel-rise test in Japanese female dance students
[abstract]. J Dance Med Sci 2003;7(2):66.
10. Lepelley MC, Thullier F, Koral J, Lestienne FG. Muscle coordination
in complex movements during jete in skilled ballet dancers. Exp Brain
Res 2006;175(2):321–331.
11. Mouchnino L, Aurenty R, Massion J, Pedotti A. Coordination
between equilibrium and head–trunk orientation during leg move-
ment: a new strategy built up by training. J Neurophysiol 1992;
67(6):1587–1598.
12. Monasterio RA, Chatfield SJ, Jensen JL, Barr S. Postural adjustments
for voluntary leg movements in dancers [thesis]. Eugene, OR: Univer-
sity of Oregon, Microform Publications; 1994.
13.Wilmerding M, Heyward VH, King M, et al. Electromyographic com-
parison of the développé devant at barre and centre. J Dance Med Sci
2001;5(3):69–74.
14.Wang A-T, Huang C-C, Hsieh C-C, Hu C-C, Lu T-W. Electromyogra-
phy analysis of grand battement in Chinese dance [translation]. In:
Lim CT, Goh JCH, eds. Proceedings of the 13th International Con-
ference on Biomedical Engineering. Singapore: IFMBE; 2008:
pp2064–2067.
15. Chatfield SJ, Krasnow DH, Herman A, Blessing G. A descriptive analy-
sis of kinematic and electromyographic relationships of the core
during forward stepping in beginning and expert dancers. J Dance
Med Sci 2007;11(3):76–84.
16. Krasnow D, Chatfield SJ, Blessing G. A preliminary investigation of
the relationship of alignment and abdominal activity during transfer of
weight through space in dancers [abstract]. J Dance Med Sci 2002;
6(1):27.
17. Harley YXR, Gibson AS, Harley EH, et al. Quadriceps strength and
jumping efficiency in dancers. J Dance Med Sci 2002;6(3):87–94.
18.Ravn S, Voigt M, Simonsen EB, et al. Choice of jumping strategy in
two standard jumps, squat and countermovement jump—effect of train-
ing background or inherited preference? Scand J Med Sci Sport 1999;
9(4):201–208.
19. Simpson KJ, Jameson EG, Odum S. Estimated patellofemoral com-
pressive forces and contact pressures during dance landings. J Appl
Biomech 1996;12(1):1–14.
20. Krasnow D, Wilmerding MV, Stecyk S, et al. Biomechanical research
in dance: a literature review. Med Probl Perform Art 2011;26(1):3–23.
21. Ambegaonkar JP, Shultz SJ, Perrin DH, et al. Lower body stiffness and
muscle activity differences between female dancers and basketball play-
December 2011 193
FIGURE 4. Left gastrocnemius trial: three MVIC trials and ensem-
ble graph. This graph shows the composite data of the three trials,
with the bold line showing the average. X-axis is in seconds, Y-axis
is in millivolts.
194 Medical Problems of Performing Artists
ers during drop jumps. Sports Health 2011;3(1):89–96.
22. Bolgla LA, Malone TR, Umberger BR, Uhl TL. Reliability of elec-
tromyographic methods used for assessing hip and knee neuromuscu-
lar activity in females diagnosed with patellofemoral pain syndrome. J
Electromyogr Kinesiol 2010;20(1):142–147.
23. Bolgla LA, Uhl TL. Reliability of electromyographic normalization
methods for evaluating the hip musculature. J Electromyogr Kinesiol
2007;17(1):102–111.
24.Kulas AS, Schmitz RJ, Shultz SJ, et al. Specific abdominal activation
strategies during landing. J Athl Train 2006;41(4):381–386.
25. Worrell TW, Crisp E, LaRosa C. Electromyographic reliability and
analysis of selected lower extremity muscles during lateral step-up con-
ditions. J Athl Train 1998;33(2):156–162.
26. Burden A. How should we normalize electromyograms obtained from
healthy participants?: what we have learned from over 25 years of
research. J Electromyogr Kinesiol 2010;20:1023–1035.
27. Chatfield S. Electromyographic response of dancers to isokinetic work
and select dance movements. Kinesiol Med Dance 1993/1994;
16(1):60–82.
28. Agre JC, Magness JL, Hull SZ, et al. Strength testing with a portable
dynamometer: reliability for upper and lower extremities. Arch Phys
Med Rehabil 1987;68:454–458.
29. Andrews AW, Thomas MW, Bohannon RW. Normative values for iso-
metric muscle force measurements obtained with hand-held
dynamometers. Phys Ther 1996;76(3):248–259.
30. Bohannon RW. Reference values for extremity muscle strength
obtained by hand-held dynamometry from adults aged 20 to 79 years.
Arch Phys Med Rehabil 1997;78:26–32.
31. Bohannon RW. Responsiveness of hand-held dynamometry to changes
in limb muscle strength: a retrospective investigation of published
research. Isokinet Exerc Sci 2009;17(4):221.
32. Kelln BM, McKeon PO, Lauren M, et al. Hand-held dynamometry:
reliability of lower extremity muscle testing in healthy, physically active,
young adults. J Sport Rehabil 2008;17:160–170.
33. Thorborg K, Petersen J, Magnusson SP, Hölmich P. Clinical assess-
ment of hip strength using a hand-held dynamometer is reliable. Scand
J Med Sci Sports 2010;20:493–501.
34.Wikholm JB, Bohannon RW. Hand-held dynamometer measure-
ments: tester strength makes a difference. J Orthop Sports Phys Ther
1991;13(4):191–198.
35. Kramer JF, Vaz MD, Vandervoort AA. Reliability of isometric hip
abductor torques during examiner and belt-resisted tests. J Gerontol
1991;46:M47–51.
36. Nadler SF, DePrince ML, Hauesien N, et al. Portable dynamometer
anchoring station for measuring strength of the hip extensors and
abductors. Arch Phys Med Rehabil 2000;81(8):1072–1076.
37. Scott DA, Bond EQ, Sisto SA, Nadler SF. The intra- and interrater reli-
ability of hip muscle strength assessments using a handheld versus a
portable dynamometer anchoring station. Arch Phys Med Rehabil
2004;85(4):598–603.
