ArticlePDF Available

Surface electromyography for assessing triceps brachii muscle activities: A literature review

Authors:
  • American International University - Bangladesh (AIUB)
  • Universiti Teknikal Malaysia Melaka (UTeM)

Abstract and Figures

The goal of this review was to summarise the scientific findings of research conducted on the triceps brachii muscle using surface electromyography. To achieve this goal, we searched through several articles available from the online databases ScienceDirect and SpringerLink published in the English language between January 2008 and June 2012. We specifically searched for the phrases “EMG” and “triceps brachii” in the title, abstract, keywords or methods sections. From a total of 569 articles we identified 77 potentially relevant studies where 42 studies have been examined triceps brachii muscle activity using surface electromyography that applied in the field of rehabilitation, physiological exercise, sports, and prosthesis control. Among the 42 articles found, 16 studies have been examined triceps brachii muscle activity in rehabilitation, 13 for physiological exercise, 9 for sports, and 4 for prosthesis control in this literature review. We therefore believe that the information contained in this review will greatly assist and guide the progress of studies that use surface electromyography to measure triceps brachii muscle activity in the context of rehabilitation, physiological exercise, sports, and prosthesis control.
Content may be subject to copyright.
Review
Article
Surface
electromyography
for
assessing
triceps
brachii
muscle
activities:
A
literature
review
Asraf
Ali *,
Kenneth
Sundaraj,
R.
Badlishah
Ahmad,
Nizam
Uddin
Ahamed,
Anamul
Islam
AI-Rehab
Research
Group,
Universiti
Malaysia
Perlis
(UniMAP),
Kampus
Pauh
Putra,
Perlis,
Malaysia
1.
Introduction
The
activity
of
the
triceps
brachii
muscle
(TB)
is
an
important
role
for
the
movement
of
human
body.
There
are
many
examples
of
application
eld
involving
the
TB,
including
rehabilitation
[16],
physiological
exercise
[711],
sports
[12
15],
and
prosthesis
control
[1618].
To
study
the
TB
activities
in
different
application
elds,
researchers
have
utilised
surface
electromyography
(sEMG)
to
detect
the
TB
activity
[1923].
sEMG
consists
of
a
series
of
procedures
to
test
the
electrical
activity
of
a
muscle
[24].
sEMG
uses
non-invasive
techniques
to
measure
muscle
activity
by
a
bipolar
electrode,
which
is
easy
to
use
and
has
therefore
been
widely
adopted
in
both
clinical
and
non-clinical
settings
to
assess
physiological
and
patho-
logical
muscle.
For
example,
the
multi-channel
sEMG
is
utilised
to
evaluate
methods
for
predicting
muscle
forces
[25].
Moreover,
the
sEMG
is
also
able
to
record
myoelectric
signals
from
the
surfaces
of
skeletal
muscles
to
improve
the
performance
of
myoelectric
prosthesis
control
[17].
However,
with
the
aid
of
sEMG,
a
group
of
researchers
have
recently
characterised
TB
activity
for
the
rehabilitation,
including
stroke
rehabilitation
by
improving
motor
unit
[26,27],
and
force
[28,29];
the
spinal
cord
injury
rehabilitation
by
improving
motor
unit
[30,31];
and
the
cerebral
palsy
rehabilitation
by
improving
force
modulation
[4].
In
addition,
TB
of
healthy
subject
can
be
rehabilitated
by
reducing
fatigue
[32],
and
a
contribution
of
elbow
joint
stability
[6].
However,
b
i
o
c
y
b
e
r
n
e
t
i
c
s
a
n
d
b
i
o
m
e
d
i
c
a
l
e
n
g
i
n
e
e
r
i
n
g
3
3
(
2
0
1
3
)
1
8
7
1
9
5
a
r
t
i
c
l
e
i
n
f
o
Article
history:
Received
21
March
2013
Accepted
7
June
2013
Available
online
10
September
2013
Keywords:
Triceps
brachii
muscle
activity
Surface
electromyography
Rehabilitation
Physiological
exercise
Sports
Prosthesis
control
a
b
s
t
r
a
c
t
The
goal
of
this
review
was
to
summarise
the
scientic
ndings
of
research
conducted
on
the
triceps
brachii
muscle
using
surface
electromyography.
To
achieve
this
goal,
we
searched
through
several
articles
available
from
the
online
databases
ScienceDirect
and
SpringerLink
published
in
the
English
language
between
January
2008
and
June
2012.
We
specically
searched
for
the
phrases
‘‘EMG’’
and
‘‘triceps
brachii’’
in
the
title,
abstract,
keywords
or
methods
sections.
From
a
total
of
569
articles
we
identied
77
potentially
relevant
studies
where
42
studies
have
been
examined
triceps
brachii
muscle
activity
using
surface
electro-
myography
that
applied
in
the
eld
of
rehabilitation,
physiological
exercise,
sports,
and
prosthesis
control.
Among
the
42
articles
found,
16
studies
have
been
examined
triceps
brachii
muscle
activity
in
rehabilitation,
13
for
physiological
exercise,
9
for
sports,
and
4
for
prosthesis
control
in
this
literature
review.
We
therefore
believe
that
the
information
contained
in
this
review
will
greatly
assist
and
guide
the
progress
of
studies
that
use
surface
electromyography
to
measure
triceps
brachii
muscle
activity
in
the
context
of
rehabilitation,
physiological
exercise,
sports,
and
prosthesis
control.
#
2013
Nałęcz
Institute
of
Biocybernetics
and
Biomedical
Engineering.
Published
by
Elsevier
Urban
&
Partner
Sp.
z
o.o.
All
rights
reserved.
*
Corresponding
author
at:
AI-Rehab
Research
Group,
Universiti
Malaysia
Perlis
(UniMAP),
Compleks
Ulu
Pauh,
02600
Arau
Perlis,
Malaysia.
E-mail
address:
asrafbabu@hotmail.com
(M.A.
Ali).
Available
online
at
www.sciencedirect.com
ScienceDirect
journal
homepage:
www.elsevier.com/locate/bbe
0208-5216/$
see
front
matter
#
2013
Nałęcz
Institute
of
Biocybernetics
and
Biomedical
Engineering.
Published
by
Elsevier
Urban
&
Partner
Sp.
z
o.o.
All
rights
reserved.
http://dx.doi.org/10.1016/j.bbe.2013.09.001
with
the
physiological
exercise,
many
of
researchers
have
identied
the
TB
activity
using
sEMG,
including
eccentric
elbow
exion
or
extension
exercise
[33,34],
exercise-induced
muscle
damage
protocol
[7],
concentric
and
eccentric
fatigue
protocol
[19],
isometric
maximum
voluntary
exercise
[8,35],
bench
press
exercise
[9,36],
discrete
bimanual
aiming
move-
ment
exercise
[10,37],
during
lat-pull
exercise
[11].
Moreover,
another
group
of
researchers
have
examine
the
TB
activity
using
sEMG
in
sports,
including
tennis
forehand
drive
[12,23,38],
karate
[13,22],
swimming
[5,39],
dart
throwing
[15],
pole
vault
weight
bearing
[40].
On
the
other
hand,
some
researchers
have
veried
signal
processing
using
sEMG
on
TB
for
the
prosthesis
control
[16,41].
We
identied
six
literature
reviews
related
to
sEMG:
three
on
sEMG
signal
processing
[24,42,43],
one
describing
sEMG
pattern
recognition
[44],
one
measuring
the
EMG
variables
of
a
golf
swing
[45],
and
one
concerning
sEMG
for
motor
unit
analysis
[46].
To
our
knowledge,
there
is
no
review
article
that
evaluates
only
the
TB
muscle
activities
using
sEMG.
Therefore,
the
purpose
of
this
review
is
to
assess
the
different
aspects
of
TB
activity-oriented
research
using
sEMG.
The
possibilities
for
future
work
on
TB
activity
using
sEMG
will
also
be
discussed
in
this
study.
2.
Method
2.1.
Article
searching
procedure
We
used
a
systematic
searching
procedure
to
identify
available
articles
on
TB
activities
conducted
using
sEMG
from
the
online
digital
databases
ScienceDirect
and
SpringerLink.
In
our
systematic
searching
procedure,
we
used
the
keyword
‘‘EMG’’
only
on
journal
articles
published
in
the
English
language
between
January
2008
and
June
2012.
We
then
used
the
keyword
‘‘triceps
brachii’’
within
this
set
of
results
to
further
narrow
the
set
of
publications.
2.2.
Article
selection
criteria
For
the
nal
selection
of
articles
to
assess
TB
activity
using
sEMG,
we
used
some
criteria
to
exclude
some
articles
from
the
available
set
of
both
the
ScienceDirect
and
SpringerLink
online
databases.
The
exclusion
criteria
were
based
on
the
article's
title,
abstract,
and
relevance
to
measuring
TB
activity
using
sEMG.
3.
Results
3.1.
Article
search
results
A
search
of
the
ScienceDirect
and
SpringerLink
electronic
databases
retrieved
15335
and
4710
articles,
respectively,
that
t
with
the
keyword
‘‘EMG’’.
A
rened
search
was
then
run
using
the
keyword
‘‘triceps
brachii’’,
which
retrieved
375
and
194
articles,
respectively,
from
the
ScienceDirect
and
Spring-
erLink
databases.
Finally,
based
on
the
exclusion
criteria,
we
selected
39
articles
from
the
ScienceDirect
database
and
38
articles
from
the
SpringerLink
database.
The
article
search
results
are
summarised
in
Fig.
1.
Finally,
we
found
77
articles
studying
TB
activities
using
sEMG
for
use
in
this
review.
3.2.
Article
study
results
We
analysed
all
of
the
collected
articles
individually
and
focused
on
the
TB
activities
measured
using
sEMG
mainly
on
four
application
elds:
rehabilitation,
physiological
exercise,
sports,
and
prosthesis
control.
Fig.
1
Article
search
results.
b
i
o
c
y
b
e
r
n
e
t
i
c
s
a
n
d
b
i
o
m
e
d
i
c
a
l
e
n
g
i
n
e
e
r
i
n
g
3
3
(
2
0
1
3
)
1
8
7
1
9
5188
From
the
77
retrieved
articles,
16
studies
applied
in
rehabilitation
where
6
studies
[1,3,26,27,30,31]
analysed
the
motor
unit,
1
study
[5]
measured
fatigue,
6
studies
[2,4,25,28,29,32]
considered
force,
1
study
[6]
evaluated
joint
stability,
and
2
studies
[47,48]
estimated
movement;
12
studies
applied
in
physiological
exercise
where
2
studies
[7,33]
analysed
the
motor
unit,
6
studies
[8,9,19,3436]
measured
fatigue,
1
study
[49]
evaluated
joint
stability,
and
3
studies
[10,11,37]
estimated
movement;
10
studies
applied
in
sports
where
2
studies
[13,38]
analysed
the
motor
unit,
3
studies
[12,39,50]
measured
fatigue,
1
study
[51]
considered
force,
2
studies
[23,40]
evaluated
muscle
coordination,
and
2
studies
[15,22]
estimated
movement;
4
studies
[1618,41]
applied
in
prosthesis
control
by
signal
processing;
and
other
35
studies
were
not
related
to
our
main
focus
where
8
studies
[20,5258]
analysed
the
motor
unit,
4
studies
[5962]
measured
fatigue,
3
studies
[21,63,64]
considered
force,
4
studies
[14,6567]
evaluated
joint
stability,
and
4
studies
[6871]
estimated
movement,
6
studies
[7277]
assessed
muscular
or
EMG
activity,
and
6
studies
[24,4246]
reviewed
literature.
Those
studies
were
not
related
to
our
main
focus
did
not
consider
to
discuss
further
in
this
study.
Therefore,
42
studies
out
of
our
selected
77
studies
have
been
considered
to
discuss
further
for
assessing
TB
activities
using
sEMG.
3.3.
TB
activity
sEMG
results
in
rehabilitation
The
sEMG
recorded
values
of
TB
found
in
the
retrieved
articles
for
the
rehabilitation
are
resulted
in
this
section.
For
the
stroke
rehabilitation,
the
study
[1]
evaluated
reex
probability
and
size
values
from
the
recording
of
sEMG
on
TB
between
stroke
patient
and
healthy
subject.
This
study
showed
that
the
stroke
patient's
TB
activity
higher
than
healthy
subject
due
to
motor
dysfunction.
The
authors
also
suggested
that
the
stroke
patient
was
rehabilitated
by
improving
motor
unit
due
to
loss
of
capacity
to
dynamically
modulate
the
nociceptive
withdrawal
reex
according
to
the
movement
variables.
In
addition,
another
two
studies
[19,66]
also
suggested
for
the
rehabilitation
of
stroke
patient
by
improving
the
motor
unit.
The
study
[26]
found
that
sEMG
on
TB
activity
of
the
stroke
patient
lower
corticomuscular
coherence
compared
to
healthy
subject.
The
study
[27]
evaluated
the
maximum
rate
of
EMG
rise,
peak
amplitude,
and
root
mean
square
of
TB
and
found
all
the
results
signicantly
lower
for
stroke
patient
than
healthy
subject.
Moreover,
force
is
another
factor
for
the
stroke
rehabilitation.
For
example,
the
study
[2]
evaluated
the
activation
level
of
the
lateral
head
of
TB
in
the
stroke
patient
using
EMG-driven
modelling
for
the
prediction
of
individual
elbow
muscle
forces.
This
study
found
that
the
patient
increased
in
the
elbow
exion
movement
and
demonstrated
that
the
EMG-driven
model
might
be
useful
for
the
rehabilitation
of
TB
by
force
generation.
The
study
[29]
evaluated
peak
force
of
TB
by
EMG
reading
and
found
that
minimising
peak
force
increased
cadence
and
recovery
power
of
wheelchair
user
stroke
patient.
This
study
mentioned
that
minimising
cadence
appears
to
have
the
most
potential
for
reducing
muscle
demand
and
fatigue
which
could
decrease
upper
extremity
injuries
and
pain.
For
the
spinal
cord
injury
rehabilitation,
the
study
[3]
evaluated
the
group
root
mean
square
EMG
amplitudes
of
the
passive
upper
limb
muscles
were
signicantly
greater
during
the
passive
upper
and
active
lower
conditions
compared
to
the
TB
activity
due
to
motor
dysfunction.
This
study
suggested
that
spinal
cord
injury
patient
can
be
rehabilitated
by
improving
motor
unit.
In
addition,
another
two
studies
[30,31]
also
suggested
the
rehabilitation
of
spinal
cord
injury
patient
with
improving
the
motor
unit.
The
study
[30]
evaluated
EMG
activation
levels
of
TB
between
paraplegic
patient
and
healthy
subject.
This
study
found
that
EMG
activation
levels
of
paraplegic
subject
higher
compared
to
healthy
subjects
in
the
early
push
and
late
push
phases
during
the
recovery
period
using
different
types
of
wheelchair
propulsion
mechanisms
due
to
motor
dysfunction.
The
study
[31]
evaluated
the
EMG
reading
on
the
lateral
head
of
TB
where
TB
was
active
during
approximately
70%
of
the
movement
of
the
tetraplegia
patient
that
was
not
similar
to
the
healthy
subject
due
motor
dysfunction.
For
the
rehabilitation
of
cerebral
palsy,
the
study
[4]
evaluated
root
mean
square
EMG
activity
on
TB
and
resulted
that
the
affected
arm
had
signicantly
lower
compared
to
the
contralateral
arm
due
to
force
modulation.
This
study
suggested
that
cerebral
palsy
patient
can
be
rehabilitated
by
improving
force
modulation.
For
the
fatigue
rehabilitation
of
healthy
subject,
the
study
[5]
found
that
after
repetitive
arm
movement
the
EMG
variability
on
TB
was
approximately
76%
higher
in
the
shoulder
angular
position
sense
task
compared
to
the
upper
limb
endpoint
position
sense
task
due
to
fatigue.
This
study
also
suggested
that
healthy
subject
individually
able
to
develop
strategies
to
compensate
for
fatigue-induced
decits
at
one
joint
property
to
maintain
the
endpoint
accuracy
of
a
multi-joint
task
constant.
Moreover,
force
is
another
factor
for
the
fatigue
rehabilitation
of
healthy
subject.
For
example,
the
study
[32]
evaluated
the
TB
contractions
by
EMG
reading
and
found
that
sustained
contractions
led
to
decreased
twitch
force
24-h
post-exercise,
whereas
recovery
was
completed
within
1-h
after
intermittent
contractions.
This
study
sug-
gested
that
time-varying
force
may
be
a
useful
intervention
to
reduce
local
fatigue
in
healthy
works
performing
low-load
task.
For
the
joint
stability
rehabilitation
of
healthy
subject,
the
study
[6]
investigated
the
effects
of
arm
posture
and
hand
load
on
TB
activity
during
baseline
and
reex
perturbations
of
the
arm
and
found
that
during
baseline
period,
known
perturba-
tion
timing
resulted
in
greater
muscular
activity
than
for
unknown
timing,
while
the
opposite
was
found
for
the
reex
period.
This
study
recommended
that
body
orientation
and
hand
loading
inuenced
muscular
response
to
elbow
pertur-
bation
and
muscle
co-contraction
at
the
elbow
during
known
timing
which
suggested
a
contribution
to
elbow
joint
stability
that
may
be
reduce
injury
risk
cause
by
sudden
elbow
loading
of
healthy
subject.
For
the
gait
disorder
rehabilitation,
the
study
[47]
examined
the
EMG
activity
on
TB
and
found
that
anti-normal
gait
involved
stronger
EMG
activity
than
normal
gait.
This
result
indicated
that
normal
arm
swing
during
human
walking
involves
active
and
passive
components
and
suggested
that
gait
disorder
can
be
rehabilitated
with
reduced
or
altered
arm
swing
movements.
In
addition,
another
study
[48]
also
suggested
for
the
rehabilitation
of
gait
disorder
patient
by
contribution
of
arm
movement.
This
study
evaluated
EMG
b
i
o
c
y
b
e
r
n
e
t
i
c
s
a
n
d
b
i
o
m
e
d
i
c
a
l
e
n
g
i
n
e
e
r
i
n
g
3
3
(
2
0
1
3
)
1
8
7
1
9
5
189
amplitude
of
TB
activity
after
a
trip
initiation
and
found
that
EMG
amplitude
of
TB
activity
was
larger
in
the
retroexors
on
the
tripped
side,
which
indicated
that
the
TB
is
the
most
important
contributor
to
arm
movements
for
balance
recovery
from
a
trip.
3.4.
TB
activity
sEMG
results
in
physiological
exercise
The
sEMG
recorded
values
of
TB
found
in
the
retrieved
articles
for
the
physiological
exercise
are
resulted
in
this
section.
For
the
eccentric
elbow
exion/extension
exercise,
the
study
[33]
evaluated
EMG
burst
duration,
and
segments
timing
and
found
that
TB
relative
duration
was
decreased
delayed
post
exercise
due
to
shorten
activity
of
TB.
This
study
also
mentioned
that
TB
activity
started
later
2
h
post-exercise
from
the
beginning
segment
and
was
associated
with
parallel
changes
of
next
two
segments.
The
study
[7]
examined
EMG
on
TB
activity
with
the
exercise-induced
muscle
damage
protocol
and
found
that
after
132
h
exercise,
TB
activity
of
experimental
group
signicant
changes
during
maximal
force,
isometric
low
force,
submaximal
force,
and
maximum
voluntary
contrac-
tion,
but
there
were
no
changes
in
the
control
group.
These
changes
suggested
that
central
regulation
inuenced
the
neural
ring
patterns
and
motor
unit
activity.
Fatigue
observing
on
TB
muscle
in
exercise,
the
study
[19]
examined
low-frequency
fatigue
on
TB
after
concentric
and
eccentric
fatigue
protocols.
This
study
found
the
level
of
fatigue-induced
excitationcontraction
coupling
compromise
was
similar
across
protocol,
but
eccentrically-induced
fatigue
creates
greater
unsteadiness
during
a
precision
task
and
suggested
that
low-frequency
fatigue
is
induced
to
similar
extent
regardless
of
how
the
fatigue
is
induced
across
different
contraction
types.
Fatigue
also
able
to
measure
with
the
isometric
elbow
extensor
exercise,
for
example,
the
study
[8]
examined
normalised
interference
EMG
amplitude
and
found
that
sex
differences
were
not
observed
before
the
fatigue
task
but
a
main
effect
was
observed
for
fatigue-state.
These
results
indicated
that
interference
EMG
amplitude
increased
during
the
fatigue
task.
Another
study
[35]
evaluated
the
root
mean
square
normalised
values
and
found
that
signicantly
decreased
between
the
initial
and
nal
values
recorded
at
06:00
h
and
18:00
h;
and
no
signicant
interactions
between
time
of
day
and
fatigue
effect
were
noted
on
either
root
mean
square
normalised
values
or
ratio
between
nal
and
initial
state.
This
study
indicated
that
the
time
of
day
effect
is
not
the
actual
cause
of
fatigue
and
has
no
effect
on
recovery
during
isometric
maximal
voluntary
exercise.
The
study
[36]
mea-
sured
EMG
activity
on
TB
before
and
through
48
h
of
recovery
after
performing
100
eccentric
actions
of
the
barbell
bench
press
exercise.
This
study
identied
that
frequency
median
demonstrated
time
main
effects
but
not
repeated
bout
effect
and
suggested
that
repeated
bout
effect
is
absent
in
resistance
trained
subject.
Another
study
[9]
recorded
EMG
on
TB
during
bench
press
exercise
and
evaluated
root
mean
square
frequency
and
median
frequency
under
varying
speeds
and
intensities.
This
study
found
that
median
frequencies
before
fatigue
were
similar
among
speeds
and
intensities
due
to
effect
of
motor
unit
synchronisation,
and
the
fall
in
median
frequency
after
fatigue
was
mostly
similar
across
speeds,
but
was
greater
for
lighter
intensities.
These
results
indicated
that
fatigue
levels
were
similar
across
speeds
but
were
greater
for
lighter
intensities.
On
the
other
hand
the
study
[34]
examined
the
time
course
of
adaptation
through
20
days
of
eccentric
training
of
TB
activity.
This
study
found
TB
activity
decreased
after
20
days
of
training
and
suggested
that
antagonist
co-
activation
decreased
due
to
fatigue.
TB
muscle
can
prepare
in
advance
an
asymmetrical
bimanual
movement
with
the
practice
of
discrete
bimanual
aiming
movement.
For
example,
the
studies
[10,37]
examined
EMG
amplitude
and
EMG
triphasic
pattern
and
found
that
EMG
activity
was
greater
for
the
startle
trail
group
than
control
trail
group.
These
studies
identied
that
discrete
bimanual
aiming
movement
exercise
can
be
able
to
produce
the
correct
asymmetrical
EMG
amplitude,
rather
than
changing
the
timing
of
the
triphasic
pattern.
Another
study
[11]
examined
peak
EMG
activity
on
TB
during
lat-pull
exercise
performed
on
machines
that
one,
two,
or
three
degrees
of
freedom
for
the
movement.
This
study
found
that
the
amount
of
TB
EMG
increased
across
machines
with
a
signicantly
less
integrated
EMG
on
type-1
machines
than
on
type-3
machines,
and
the
amount
of
integrated
EMG
for
the
TB
was
signicantly
greater
than
the
biceps
brachii
on
the
type-2
and
-3
machines
but
not
on
type-1
machines.
These
results
suggested
that
the
degrees
of
freedom
on
a
weight
training
machine
can
alter
the
EMG
activity
on
TB
to
achieve
a
better
match
with
the
patterns
observed
in
everyday
activities
such
as
throwing
and
striking
actions.
3.5.
TB
activity
sEMG
results
in
sports
The
sEMG
recorded
values
of
TB
found
in
the
retrieved
articles
for
the
sports
are
resulted
in
this
section.
During
forehand
drives
in
tennis,
The
study
[12]
evaluated
EMG
bust
onset
and
offset
and
found
that
signicant
relation-
ships
were
observed
for
the
erector
spinae,
latissimus
dorsi,
pectoralis
major,
TB,
and
extension
carpi
radialis
muscles
in
EMG
burst
onset
and
for
the
erector
spinae,
anterior
deltoid,
middle
deltoid,
biceps
brachii,
TB,
and
exor
carpi
radialis
muscles
in
EMG
burst
offset,
indicating
that
the
TB
has
a
role
in
the
strength
of
the
forehand
drive
in
tennis
with
the
coordination
of
the
trunk
and
other
upper
limb
muscles.
On
the
other
hand
the
study
[23]
individually
calculated
the
onset
before
impact
and
EMG
burst
values,
and
resulted
that
the
ball
speed
and
the
TB
activation
temporal
sequences
were
similar,
whatever
the
increase
in
racket
mass.
This
study
identied
that
the
temporal
sequence
of
TB
activation
were
contracted
early
before
impact
for
elbow
extension.
Moreover,
the
study
[38]
calculated
the
normalised
root
mean
square
of
EMG
as
an
index
of
EMG
amplitude
during
TB
exion
at
different
angle.
This
study
resulted
that
antagonist
activation
of
the
TB
was
signicantly
lower
in
tennis
players
compared
to
non-players
at
all
angular
velocities
during
maximal
isokinetic
elbow
exion
and
suggested
that
a
reduced
coactivation
indicated
the
achievement
of
motor
skill.
Karate
is
the
one
more
important
sport
for
assessment
TB
activity
measured
by
EMG
recording.
The
study
[22]
evaluated
the
peak
angular
speed
and
peak
EMG
during
the
execution
of
specic
choku-zuki
punch
and
found
that
in
the
forearm
extension
and
pronation,
the
TB
had
larger
activation
intensity
in
the
karate
group.
Finally,
this
study
showed
a
better
ballistic
b
i
o
c
y
b
e
r
n
e
t
i
c
s
a
n
d
b
i
o
m
e
d
i
c
a
l
e
n
g
i
n
e
e
r
i
n
g
3
3
(
2
0
1
3
)
1
8
7
1
9
5190
performance
in
the
execution
of
the
choku-zuki
punch
than
in
the
non-karate
group.
On
the
other
hand
the
study
[13]
examined
EMG
root
mean
square
and
frequency
wavelet
transform
domain
during
strike
time
of
Kung
Fu.
This
study
identied
the
greater
values
of
root
mean
square
and
sum
of
signicant
power
of
the
TB
activity
in
the
rst
half
of
the
movements.
This
is
mentioned
that
EMG
characteristic
of
Kung
Fu
is
important
not
only
for
improving
the
performance
of
practitioners
but
also
to
demonstrate
the
applicability
of
this
martial
art
in
the
process
of
motor
control
development.
TB
muscle
having
active
role
in
swimming,
for
example,
the
study
[50]
analysed
EMG
amplitude
and
frequency
parameters
and
found
that
the
EMG
amplitude
signicantly
increased
due
to
recruitment
of
additional
motor
unit
during
swimming,
but
power
frequency
signicantly
decreased.
These
changes
in
EMG
amplitude
and
frequency
parameters
at
the
100-m
of
front
crawl
swim
with
respect
of
the
beginning
of
swimming
indicated
the
presence
of
muscle
fatigue.
Another
study
[39]
evaluated
changes
in
TB
activity
during
200-m
of
front
crawl
swimming
associated
with
physiological
fatigue
and
de-
creased
swimming
velocity,
and
also
examine
the
relationship
between
the
decreased
swimming
velocity
and
EMG
param-
eters.
This
study
found
that
the
mean
amplitude
value
decreased
with
a
decrease
in
swimming
velocity,
and
positive
correlations
were
identied
between
mean
amplitude
values.
These
results
suggested
that
the
decrease
in
swimming
velocity
was
related
to
decrease
in
the
TB
activities
that
coordinated
with
each
other.
With
the
dart
throwing,
the
study
[15]
calculated
the
integrated
EMG
activity
of
TB
for
the
onset-to-release
interval
and
found
signicantly
less
integrated
EMG
activity
of
TB
during
external
focus
than
during
internal
focus.
This
result
indicated
that
subjects
should
be
focus
on
the
effect
of
their
actions,
rather
than
on
body
movement
for
improving
performance
during
training
and
retention
testing.
With
the
pole
vault
weight
bearing,
the
study
[40]
evaluated
the
wavelet
transformed
EMG
signal
on
TB
was
quantied
by
its
wavelet
intensities
between
take-off
and
complete
pole
straightening
times
for
each
pole
vault
and
observed
that
TB
had
less
sensitive
to
the
different
sub
phases
of
the
vault
but
suggested
that
the
EMG
proles
allowed
dening
the
chrono-
logical
muscle
coordination
of
both
upper
limbs
during
the
pole
support
phase
of
the
vault.
One
study
[51]
developed
a
valid
model
that
predicts
isometric
force
from
EMG
signal
on
TB.
The
relationship
between
EMG
and
force
predicted
by
this
model
is
highly
non-
linear,
with
an
over-estimation
of
the
increase
in
force
at
low
activation
levels
and
an
under-estimation
of
the
increase
in
force
at
high
activation
levels.
These
results
indicated
that
the
simple
physiological
model
is
able
to
reproduce
and
predict
measured
contractions
using
sEMG
on
the
TB.
3.6.
TB
activity
sEMG
results
in
prosthesis
control
The
sEMG
recorded
values
of
TB
found
in
the
retrieved
articles
for
the
prosthesis
control
are
resulted
in
this
section.
For
the
classication
of
signal
processing,
the
support
vector
machine
classier
gives
a
very
good
average
accuracy
rate
(9698%)
that
can
be
used
to
classify
EMG
signals
for
proper
arm
prosthesis
control
[16].
With
isometric
and
anisometric
contractions,
most
of
the
error
values
of
the
higher
order
frequency
moments
(HOFMS)
had
less
than
root
mean
square
features
and
suggested
that
HOFMs
of
the
signal
up
to
the
fourth
order
could
be
used
instead
of
the
entire
frequency
content
of
the
signal
to
decrease
the
computational
cost.
Therefore,
the
spectral
features
of
the
EMG
signal
might
be
used
as
input
parameters
for
the
myoelectric
prostheses
control
[41].
With
the
different
types
of
arm
and
hand
movements,
it
was
reported
that
the
combination
of
a
sampling
rate
of
500
Hz
and
high-pass
cut-off
frequency
of
60
Hz
is
an
optimal
selection
in
EMG
recordings
for
the
recognition
of
different
arm
movements
without
sacricing
too
much
classication
accuracy
and
can
also
remove
most
motion
artefacts
and
power-line
interferences
to
improve
the
performance
of
myoelectric
prosthesis
control
[17].
However,
during
the
alternate
dumbbell
curl
motion
test,
the
degree
of
muscular
fatigue
of
the
TB
decreased
in
cases
where
the
elbow
orthosis
was
not
used,
and
the
muscular
activities
wearing
the
elbow
orthosis
were
reduced
and
elbow
joint
torque
wearing
the
elbow
orthosis
was
higher
because
of
the
assist
of
the
orthosis.
These
results
indicated
that
the
effect
of
elbow
orthosis
in
exercise
could
be
performed
with
the
use
of
a
smaller
degree
of
muscular
activity
[18].
4.
Discussion
We
identied
four
main
application
areas
for
the
assessment
of
TB
activity
using
sEMG.
First,
we
found
several
studies
on
TB
activity
using
sEMG
in
rehabilitation.
Second,
there
were
some
studies
in
which
sEMG
were
used
to
assess
TB
activity
in
physiological
exercise.
Third,
TB
activity
can
be
analysed
using
sEMG
in
sports.
Fourth,
there
were
few
studies
processing
sEMG
signals
that
demonstrated
that
TB
activity
assessment
is
possible
by
processing
the
sEMG
signal
for
prosthesis
control.
4.1.
TB
activity
in
rehabilitation
We
already
mentioned
that
several
studies
have
been
utilised
sEMG
for
the
rehabilitation
of
the
TB.
In
one
study
[4],
cerebral
palsy
patients
were
rehabilitated
by
recovering
the
co-
activation
in
maximal
force
generation
and
sub-maximal
force
tracing
with
performing
elbow
extension
and
exion.
Another
study
[30]
showed
that
paraplegic
patients
recovered
by
improving
muscle
activation
using
handrim
wheelchair
propulsion.
The
third
study
[31]
demonstrated
that
a
spinal
cord
injury
patient
recovered
motor
function
by
performing
a
standardised
gross
upper
extremity
task.
However,
the
studies
[1,2,26,27,29]
concerned
stroke
rehabilitation,
but
these
patients
were
rehabilitated
by
recovering
motor
units
[1,26]
or
improving
co-ordination
between
the
TB
and
the
biceps
brachii
[27]
with
a
reaching
task.
For
example,
Barker
et
al.
[27]
noted
that
TB
EMG
onset
time
relative
to
movement
onset
was
signicantly
delayed,
and
the
maximum
rate
of
EMG
rise,
peak
amplitude,
root
mean
square,
and
the
ratio
of
TB
to
biceps
brachii
were
signicantly
lower
for
stroke
patients
compared
to
healthy
patients.
This
study
trained
the
stroke
patients
for
four
weeks
with
a
reaching
task
using
a
SMART
Arm
with
and
without
EMG-triggered
functional
electrical
stimulation.
They
noted
an
increase
in
TB
activation
and
improved
co-ordination
b
i
o
c
y
b
e
r
n
e
t
i
c
s
a
n
d
b
i
o
m
e
d
i
c
a
l
e
n
g
i
n
e
e
r
i
n
g
3
3
(
2
0
1
3
)
1
8
7
1
9
5
191
of
the
TB
with
biceps
brachii
activity.
Moreover,
the
TB
muscles
of
stroke
patients
were
rehabilitated
by
recovering
force
and
joint
properties
with
dynamic
movement
[2]
and
by
reducing
muscle
demand
and
fatigue
with
dynamic
simulations
of
wheelchair
propulsion
[29].
Hence,
research
can
be
conducted
on
the
rehabilitation
of
TB
activity
in
patients
with
cerebral
palsy,
paraplegia,
spinal
cord
injury,
stroke
and
other
disorders
like
gait
disorder.
4.2.
TB
activity
in
physiological
exercise
One
study
[9]
examining
TB
fatigue
during
a
bench
press
physiological
exercise
showed
that
EMG
amplitude
in-
creased
with
the
speed
effect
after
fatigue
and
that
the
median
frequency
was
similar
among
speeds
and
intensities
before
fatigue
because
higher
speed
and
intensity
require
greater
motor
unit
recruitments.
Thus,
the
motor
units
of
the
TB
can
be
recruited
by
physiological
exercise.
However,
Krentz
and
Farthing
[34]
found
that
successive
intense
eccentric
training
of
the
elbow
exors
performed
every
2nd
day
for
a
20-day
period
decreases
strength
in
previously
untrained
individuals
and
suggested
that
the
time
course
of
early
adaptations
to
eccentric
training
can
be
used
to
recover
the
muscle
strength.
One
study
[7]
also
suggested
that
exercise-induced
muscle
damage
disturbs
neuromus-
cular
function
during
maximal
and
submaximal
isometric
and
submaximal
exion-extension
contractions
over
a
132
h
exercise
period
due
to
the
inuences
of
neural
ring
patterns
and
motor
unit
activity,
but
these
changes
do
not
seem
to
be
directly
related
to
clinical
symptoms.
Additionally,
one
study
[33]
emphasised
that
unaccustomed
repeated
maximal
eccentric
actions
cause
exercise-induced
muscle
damage
that
disturbs
normal
neuromuscular
func-
tion,
especially
in
the
rst
few
hours
post-exercise.
Moreover,
with
the
joint
angle,
the
TB
can
have
an
effect
on
shoulder
extension
function,
with
the
shoulder
angle
being
the
most
important
factor
in
the
extension
moment;
exercises
that
can
strengthen
and
stretch
the
TB
as
a
shoulder
extensor
were
suggested
[49].
One
study
[11]
using
the
beginning
movement
load
training
exercise
reported
a
progressive
proximal-to-distal
sequence
of
EMG
activation
involving
the
TB
and
suggested
that
beginning
movement
load
training
exercises
with
greater
degrees
of
freedom
can
enhance
the
association
between
training
actions
and
functional
activities.
Hence,
research
can
be
performed
to
improve
TB
activity
with
physiological
exercise.
4.3.
TB
activity
in
sports
One
study
[50]
used
swimming
to
evaluate
TB
muscle
fatigue
during
a
100-m
all-out
front
crawl
and
observed
that
the
mean
power
frequency
decreased
between
the
lower
part
of
the
latissimus
and
the
TB.
It
was
suggested
that
the
EMG
parameters
did
not
indicate
a
signicant
difference
on
the
TB
after
swimming
because
changes
in
the
EMG
signal
amplitude
and
mean
power
frequency
could
not
be
attributed
solely
to
the
fatigue
process.
Three
studies
[12,23,38]
exam-
ined
TB
activity
while
playing
tennis.
Rogowski
et
al.
[12,23]
investigated
TB
activity
during
a
forehand
drive
and
showed
that
the
TB
recruitment
order
was
not
substantially
modied
by
changes
in
movement
velocity,
which
had
a
signicant
inuence
on
the
EMG
root
mean
square
burst
activation/
deactivation
timing,
and
the
activation
levels
of
some
muscles.
These
results
might
bring
new
knowledge
for
strength
and
tennis
coaches
to
improve
resistance
training
protocols
from
a
performance
and
prophylaxis
perspective.
Rogowski
et
al.
also
investigated
the
relationship
between
muscle
coordination
and
mass
of
the
tennis
racket
and
suggested
that
individual
muscle
coordination
may
be
altered
by
an
increase
in
the
mass
of
the
tennis
racket.
Moreover,
Bazzucchi
et
al.
[38]
found
that
the
antagonist
activation
of
the
TB
is
signicantly
lower
in
tennis
players
than
non-players
at
all
angular
velocities
during
isokinetic
elbow
exion,
suggest-
ing
that
the
lower
elbow
extensor
coactivation
in
tennis
players
can
adapt,
playing
an
important
role
in
the
motor
learning
process.
Two
studies
[13,22]
used
karate
as
a
model.
One
study
[13]
found
that
during
the
Kung
Fu
Yau-Man
palm
strike,
the
EMG
data
showed
well-developed
muscle
coordi-
nation
of
the
subject,
in
agreement
with
kinematic
results.
These
results
are
important
not
only
for
improving
the
performance
of
the
subject
but
also
to
demonstrate
the
applicability
of
Kung
Fu
in
the
process
of
motor
control
development.
In
another
study
[22]
of
karate
punching,
the
author
found
that
the
triceps
had
larger
activation
intensities
during
forearm
extension
and
pronation
and
identied
improved
ballistic
performance
in
the
execution
of
the
choku-zuki
karate
through
the
peak
angular
speed
and
peak
EMG
closer
to
contact.
Hence,
research
on
TB
activity
can
improve
performance
in
various
types
of
sports.
4.4.
TB
activity
in
prosthesis
control
The
EMG
signal
is
a
complicated
biomedical
signal
due
to
the
anatomical
and
physiological
properties
of
the
muscles
and
their
noisy
environment.
It
is
necessary
to
classify
these
signals
for
prosthesis
control.
Alkan
and
Gunay
[16]
used
discriminant
analysis
and
a
support
vector
machine
classier
to
determine
that
a
support
vector
machine
classier
gives
a
very
good
average
accuracy
rate
(99%)
for
movement
with
a
classication
error
rate
of
1%
from
the
TB.
These
classication
rates
are
very
high,
which
allows
them
to
classify
EMG
signals
for
arm
prosthesis
control.
Guanglin
et
al.
[17]
also
investigated
the
effects
of
EMG
conditioning
and
sampling
rate
on
the
performance
of
EMG
pattern
recognition
in
identifying
differ-
ent
classes
of
arm
and
hand
motions
for
the
development
of
practical
multifunctional
myoelectric
prostheses
for
limb
amputees.
Hence,
research
on
EMG
signal
processing
from
TB
activity
can
improve
the
control
of
prosthetic
arm
move-
ments.
From
the
above
discussion,
we
found
that
TB
activity
can
be
researched
using
sEMG
in
rehabilitation,
physiological
exer-
cise,
sports
science,
and
prosthesis
control
for
further
development.
5.
Limitations
of
this
review
Because
we
included
only
journal
articles
published
in
English
and
published
between
January
2008
and
June
2012,
a
potential
limitation
of
this
study
would
be
publication
bias.
b
i
o
c
y
b
e
r
n
e
t
i
c
s
a
n
d
b
i
o
m
e
d
i
c
a
l
e
n
g
i
n
e
e
r
i
n
g
3
3
(
2
0
1
3
)
1
8
7
1
9
5192
6.
Conclusions
This
review
identied
77
articles
that
matched
our
literature
search
criteria:
71
studies
were
original
results-based
articles
for
assessing
TB
activity,
and
6
studies
were
literature
reviews.
Researchers
studied
TB
activity
using
sEMG
in
various
situations,
such
as
muscle
fatigue,
musculoskeletal
joint
angle,
motor
unit
activity,
movement
activity,
force
activity,
and
co-contraction
or
coordination
with
others
muscles.
The
ndings
from
the
various
studies
can
be
most
effectively
applied
to
rehabilitation,
physiological
exercises,
sports
science,
and
the
development
of
a
suitable
model
or
method
for
prostheses
control.
Thus,
the
TB
is
a
vital
muscle
to
examine
diverse
activities
of
the
human
body.
Finally,
we
propose
that
the
current
review
can
be
used
as
a
guide
for
further
improving
TB
activity
using
sEMG
for
rehabilitation,
physiological
exercise,
sports,
and
prosthesis
control.
r
e
f
e
r
e
n
c
e
s
[1]
Serrao
M,
Ranavolo
A,
Andersen
OK,
Don
R,
Draicchio
F,
Conte
C,
et
al.
Reorganization
of
multi-muscle
and
joint
withdrawal
reex
during
arm
movements
in
post-stroke
hemiparetic
patients.
Clin
Neurophysiol
2012;123:52740.
[2]
Li
L,
Tong
KY,
Hu
XL,
Hung
LK,
Koo
TKK.
Incorporating
ultrasound-measured
musculotendon
parameters
to
subject-specic
EMG-driven
model
to
simulate
voluntary
elbow
exion
for
persons
after
stroke.
Clin
Biomech
2009;24:1019.
[3]
Huang
HJ,
Ferris
DP.
Upper
limb
effort
does
not
increase
maximal
voluntary
muscle
activation
in
individuals
with
incomplete
spinal
cord
injury.
Clin
Neurophysiol
2009;120:17419.
[4]
Brændvik
SM,
Roeleveld
K.
The
role
of
co-activation
in
strength
and
force
modulation
in
the
elbow
of
children
with
unilateral
cerebral
palsy.
J
Electromyogr
Kinesiol
2012;22:13744.
[5]
Emery
K,
Côté
J.
Repetitive
arm
motion-induced
fatigue
affects
shoulder
but
not
endpoint
position
sense.
Exp
Brain
Res
2012;216:55364.
[6]
Holmes
MWR,
Keir
PJ.
Posture
and
hand
load
alter
muscular
response
to
sudden
elbow
perturbations.
J
Electromyogr
Kinesiol
2012;22:1918.
[7]
Plattner
K,
Baumeister
J,
Lamberts
RP,
Lambert
MI.
Dissociation
in
changes
in
EMG
activation
during
maximal
isometric
and
submaximal
low
force
dynamic
contractions
after
exercise-induced
muscle
damage.
J
Electromyogr
Kinesiol
2011;21:54250.
[8]
Dearth
D,
Umbel
J,
Hoffman
R,
Russ
D,
Wilson
T,
Clark
B.
Men
and
women
exhibit
a
similar
time
to
task
failure
for
a
sustained,
submaximal
elbow
extensor
contraction.
Eur
J
Appl
Physiol
2010;108:108998.
[9]
Sakamoto
A,
Sinclair
P.
Muscle
activations
under
varying
lifting
speeds
and
intensities
during
bench
press.
Eur
J
Appl
Physiol
2012;112:101525.
[10]
Maslovat
D,
Carlsen
A,
Chua
R,
Franks
I.
Response
preparation
changes
during
practice
of
an
asynchronous
bimanual
movement.
Exp
Brain
Res
2009;195:38392.
[11]
Koyama
Y,
Kobayashi
H,
Suzuki
S,
Enoka
R.
Enhancing
the
weight
training
experience:
a
comparison
of
limb
kinematics
and
EMG
activity
on
three
machines.
Eur
J
Appl
Physiol
2010;109:789801.
[12]
Rota
S,
Hautier
C,
Creveaux
T,
Champely
S,
Guillot
A,
Rogowski
I.
Relationship
between
muscle
coordination
and
forehand
drive
velocity
in
tennis.
J
Electromyogr
Kinesiol
2012;22:294300.
[13]
Neto
OP,
Magini
M.
Electromiographic
and
kinematic
characteristics
of
Kung
Fu
Yau-Man
palm
strike.
J
Electromyogr
Kinesiol
2008;18:104752.
[14]
Praagman
M,
Chadwick
EKJ,
van
der
Helm
FCT,
Veeger
HEJ.
The
effect
of
elbow
angle
and
external
moment
on
load
sharing
of
elbow
muscles.
J
Electromyogr
Kinesiol
2010;20:91222.
[15]
Lohse
KR,
Sherwood
DE,
Healy
AF.
How
changing
the
focus
of
attention
affects
performance,
kinematics,
and
electromyography
in
dart
throwing.
Hum
Movement
Sci
2010;29:54255.
[16]
Alkan
A,
Günay
M.
Identication
of
EMG
signals
using
discriminant
analysis
and
SVM
classier.
Expert
Syst
Appl
2012;39:447.
[17]
Li
G,
Li
Y,
Yu
L,
Geng
Y.
Conditioning
and
sampling
issues
of
EMG
signals
in
motion
recognition
of
multifunctional
myoelectric
prostheses.
Ann
Biomed
Eng
2011;39:
177987.
[18]
Kim
K,
Hong
K-J,
Kim
N-G,
Kwon
T-K.
Assistance
of
the
elbow
exion
motion
on
the
active
elbow
orthosis
using
muscular
stiffness
force
feedback.
J
Mech
Sci
Technol
2011;25:3195203.
[19]
Meszaros
AJ,
Iguchi
M,
Chang
S-H,
Shields
RK.
Repetitive
eccentric
muscle
contractions
increase
torque
unsteadiness
in
the
human
triceps
brachii.
J
Electromyogr
Kinesiol
2010;20:61926.
[20]
Wheaton
L,
Fridman
E,
Bohlhalter
S,
Vorbach
S,
Hallett
M.
Left
parietal
activation
related
to
planning,
executing
and
suppressing
praxis
hand
movements.
Clin
Neurophysiol
2009;120:9806.
[21]
Hearn
J,
Cahill
F,
Behm
D.
An
inverted
seated
posture
decreases
elbow
exion
force
and
muscle
activation.
Eur
J
Appl
Physiol
2009;106:13947.
[22]
VencesBrito
AM,
Rodrigues-Ferreira
MA,
Cortes
N,
Fernandes
O,
Pezarat-Correia
P.
Kinematic
and
electromyographic
analyses
of
a
karate
punch.
J
Electromyogr
Kinesiol
2011;21:10239.
[23]
Rogowski
I,
Creveaux
T,
Faucon
A,
Rota
S,
Champely
S,
Guillot
A,
et
al.
Relationship
between
muscle
coordination
and
racket
mass
during
forehand
drive
in
tennis.
Eur
J
Appl
Physiol
2009;107:28998.
[24]