Medical Engineering & Physics 26 (2004) 873–878

www.elsevier.com/locate/medengphy

Technical note

An investigation of the effect of electrode size and electrode location

on comfort during stimulation of the gastrocnemius muscle
We dedicate this paper to the memory of our late colleague and co-author, Jacinta O’Brien, who died suddenly on the 3rd of February 2004:
‘‘Ar dheis Dé go raibh a hanam dı́lis’’

G.M. Lyons a, G.E. Leane a, M. Clarke-Moloney b,
, J.V. O’Brien c, P.A. Grace b
a
Biomedical Electronics Laboratory, Department of Electronic and Computer Engineering, University of Limerick, Limerick, Ireland
b
Department of Vascular Surgery, Mid-Western Regional Hospital, Dooradoyle, Limerick, Ireland
c
Department of Physical Education and Sport Sciences, University of Limerick, Limerick, Ireland

Received 6 November 2003; received in revised form 4 June 2004; accepted 16 August 2004

Abstract

The use of surface neuromuscular electrical stimulation (SNMES) in medicine is well established. However, discomfort has
been identified as limiting the use of SNMES in these applications. This pilot study investigated the influence of various electrode
sizes and their positioning on perceived pain and discomfort during neuromuscular electrical stimulation (NMES) of the gastro-
cnemius muscle using surface electrodes. This study formed part of a research project to develop a therapeutic device for calf
muscle blood flow assist applications. Twelve healthy subjects (n¼ 12) participated in this pilot study. Each participant attended
the trial centre for testing which consisted of SNMES to four different electrode stimulation sites using two electrode sizes
(round with areas 19.63 and 38.48 cm2). Comfort was assessed by asking the subjects to indicate the stimulation amplitude corre-
sponding to the onset of discomfort (pain threshold) and the amplitude at which the discomfort became unbearable (pain toler-
ance).
Of the four stimulation sites tested, two were deemed unsuccessful as it was very difficult to obtain a muscle contraction using
these sites, while the remaining two sites elicited good muscle contraction. The most comfortable stimulation was achieved by pla-
cing the cathode electrode high on the calf, below the proximal end of the muscle heads and the anode electrode towards the end
of the muscle belly and when the 19.63 cm2 electrodes were used at these sites (p ¼< 0:001).
# 2004 IPEM. Published by Elsevier Ltd. All rights reserved.

Keywords: Neuromuscular electrical stimulation (NMES); Gastrocnemius muscle; Comfort

1. Introduction in medicine is well established. Examples of the appli-

cation of SNMES include the correction of drop foot
Neuromuscular electrical stimulation (NMES) refers
[2], muscle strengthening [3], mobilisation after spinal
to the application of current pulses, through surface or
cord injury [4] and maintaining range of motion in
implanted electrodes to trigger generation of a neural
immobilised subjects [5]. However, discomfort has been
action potential train to produce artificial muscle con-
identified as one of the factors in limiting the use of
traction in the associated muscle [1]. As the use of sur-
SNMES [2].
face electrodes with NMES is less invasive, surface
SNMES involves surface stimulation of neuromuscu-
NMES is more routinely applied in clinical practice.
The potential of surface neuromuscular electrical lar structures, which inadvertently results in activation
stimulation (SNMES) as a therapeutic and orthotic aid of sensory receptors on the skin surface. This acute
sensory excitation can cause discomfort which often


limits the effectiveness of stimulation programmes [6].
Corresponding author. Tel.: +353-61-482736; fax: +353-61-
482212. If stimulation intensity is increased beyond that
E-mail address: mclarkemoloney@mwhb.ie (M. Clarke-Moloney). required for muscle activation, pain nerve fibres are
1350-4533/$ - see front matter # 2004 IPEM. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.medengphy.2004.08.003
874 G.M. Lyons et al. / Medical Engineering & Physics 26 (2004) 873–878

additionally recruited, which can result in severe dis- a comfortable contraction. Muscle fatigue is shown to
comfort during high intensity stimulation. increase rapidly with increasing stimulation frequency
Discomfort can have a profound effect on treatment and has associated implications for subject comfort.
programmes [7]. For some people, the sensation of Low stimulation frequency results in less pronounced
electrical stimulation is unbearable and they totally fatigue of the neuromuscular system [4]. Therefore, the
reject SNMES as a treatment option, others are unable chosen frequency should be the lowest value capable of
to tolerate the optimal or necessary stimulus amplitude producing a smooth tetanic contraction.
and treatment duration due to perceived stimulation
discomfort. If the required stimulation could be deliv- 1.1.4. Pulse amplitude
ered in a fashion that is more comfortable, these sub- If pulse duration is low, high amplitude is required
jects may be able to fully benefit from SNMES. Thus to stimulate the muscle and vice versa [1]. This is based
the comfort characteristic of a SNMES programme is on the strength–duration curve which shows a relation-
important to the successful implementation of that pro- ship between stimulation amplitude and pulse duration.
gramme. Electrical stimulation delivered transcutaneously at
Several factors influence perceived comfort during high amplitude is expected to stimulate both cutaneous
SNMES, including: and muscular pain receptors and fibres [13,14].

1. Stimulation parameters 1.1.5. Ramp-up time

2. Electrode size The ramp-up time allows for the gradual recruitment
3. Electrode placement. of motor nerve fibres and a gradual increase in muscle
fibre contraction [11]. This gradual onset of stimulation
1.1. Stimulation parameters is more comfortable than sudden onset and is used
mainly to affect the comfort of stimulation. It may be
The stimulation parameters used in applying necessary to have a ramp-up time of 0.6 to 0.8 s to
SNMES are important in optimising treatment out- achieve an adequate slow stretch of a spastic muscle
come and maintaining subject comfort. The parameters [1], thereby reducing the effect of a rapid rising con-
include: stimulus waveform shape, pulse duration, traction.
pulse frequency, pulse amplitude, ramp-up time, ramp-
down time, on time and off time. 1.1.6. Ramp-down time
Ramp-down time allows the quality of the excitation
1.1.1. Stimulus waveform to gradually decrease in intensity, thus allowing the
The findings in the literature reveal that the stimulus muscle to return to its resting position more comfort-
waveform perceived as most comfortable is the sym- ably [11]. To enable the desired muscle response, the
metrical biphasic waveform [1,8–10]. Both asymmetri- ramp-up and ramp-down times must be shorter than
cal and symmetrical biphasic waveforms minimise the total on time [1].
irritation, with the symmetrical biphasic waveform
being most comfortable [1]. 1.1.7. On–off time
On:off time is the ratio of contraction time to relax-
1.1.2. Pulse duration ation time and is more concerned with the muscle fati-
There is clear support for pulse durations in the gue aspect of comfort. The off time is thought to allow
range of 200–400 ls [1,6–8]. Bowman et al. [7] investi- ionic gradients and neurotransmitters to recover in
gated a range of pulse durations during electrical nerve and muscle [15]. By providing an off time, the sti-
stimulation of the quadriceps; findings showed the pre- mulated muscle is allowed to rest and recuperate before
ferred pulse duration was 300 ls. Similarly, Bowman the stimulus is applied once again [16].
and Baker [8] compared pulse durations of 50 and
300 ls during quadriceps stimulation, again subjects 1.2. Electrode size
preferred the 300 ls pulse duration.
The size of the electrode used for SNMES has a
1.1.3. Pulse frequency direct effect on current density, with current density
It is known that stimulation pulse frequency affects decreasing as electrode size increases and vice versa. As
the quality of the muscle contraction and the rate of current density stimulates the contractions in the mus-
muscle fatigue [11]. Usually, a frequency of around 20– cle, its minimum value must be greater than the thresh-
40 Hz is selected which enables a smooth tetanic mus- old value for the muscle, this is achieved by changing
cle contraction. This is supported by Mourselas and the size of the electrodes to the cross-sectional area of
Granat [12] who stated that a relatively high stimu- the muscle [17]. Studies on electrode size have shown
lation frequency is required, 25 Hz or above to obtain that the optimum size for use depends on the muscle
 G.M. Lyons et al. / Medical Engineering & Physics 26 (2004) 873–878 875

stimulated. McNeal and Baker [9] compared an elec- example, during activation of the hamstring muscles
trode area of 20 cm2 to an area of 36 cm2, placed over [9].
the quadriceps and hamstrings. Results from this study The NMES Practical Guide from the Rancho Los
showed no statistically significant difference in the Amigos Rehabilitation Centre in Downey, CA [1], sug-
movements produced using small and large electrodes, gests four possible positions for the electrode pair dur-
however, several subjects expressed a preference for the ing the surface stimulation of the gastrocnemius muscle
larger electrodes (36 cm2), finding them more comfort- and the second objective of this study was to evaluate
able than the smaller electrodes (20 cm2). Larger elec- these four positions for comfort.
trodes were also favoured in a study carried out by
Alon [18] and also induced stronger muscle contrac-
tions compared with the smaller electrodes. Similar 2. Subjects and methods
findings were reported by Patterson and Lockwood [19]
who varied the electrode area from 20 to 60 cm2 on the Twelve healthy subjects (n ¼ 12), five females and
quadriceps. The smaller electrodes elicited more pain seven males with a mean age of 23 years (SD 3.2) parti-
during 25% of an isometric maximum voluntary con- cipated in the study.
traction compared with the larger electrodes. The ethical considerations adopted were in accord-
These studies have shown that larger electrodes are ance with the ‘‘World Medical Association Declaration
most comfortable, but their effectiveness in treatment of Helsinki: Recommendations guiding physicians in
also needs to be considered. A study by Alon et al. [20] biomedical research involving human subjects’’. This
found the use of larger sized electrodes most comfort- study received ethical approval from the University of
able and most effective in stimulating the gastro- Limerick and subjects gave informed consent to par-
cnemius muscle. The two largest electrodes tested ticipate in the study.
(20.25 and 40.3 cm2) contributed to more comfortable
stimulation and greater muscle force generation than 2.1. Equipment and material used
the smaller sizes (2.25 and 9 cm2). Both these electrodes
produced the same force of contraction and comfort; A BMR1 NeuroTech NT2000 muscle stimulator was
therefore, the 20.25 cm2 electrode appeared to be the used to stimulate the calf muscle. The muscle stimu-
size of choice for comfort, as selecting the 40.3 cm2 lator was set to deliver a pulse duration of 300 ls, a
electrode did not add to optimal excitation [20]. The pulse frequency of 35 Hz, a contraction time of 2 s, a
findings from these studies indicate that optimum elec- relaxation time of 6 s, a ramp-up time of 0.5 s and a
trode size is related to the size of the muscle stimulated. ramp-down time of 0.2 s. These parameters, as already
One objective of the study described in this paper discussed, have been shown to be most effective and
was to evaluate two almost identical sized electrodes cause least discomfort [6,7,10,11,15].
to that used by Alon et al. [20] (19.93 and 38.48 cm2) An asymmetrical biphasic waveform was selected
for comfort during stimulation of the gastrocnemius over the monophasic waveform as the literature
muscle. showed that biphasic waveforms minimised skin irri-
tation compared to monophasic waveforms [1].
All subjects wore loose clothing on their lower anat-
1.3. Electrode placement
omy to facilitate attachment of the electrodes to the
It is critical that the optimum position to elicit the calf muscle. The electrode site at the posterior of the
desired muscle contraction is used for each electrode, lower leg was prepared by clipping the hairs around
and it is noted that this position may vary with each the site of stimulation with a scissors and washing the
individual due to anatomical changes from subject to skin surface. Two different sized round PALS#2 elec-
subject. trodes, (area of 19.63 and 38.48 cm2) were used for
If an electrode pair is not optimally placed for stimu- each of the four placement locations. Care was taken
lation of a muscle, the stimulation amplitude required to ensure that the electrodes were not placed over a
to elicit a muscle contraction of sufficient strength will bony prominence, cuts or an area of skin irritation.
be higher than otherwise necessary, with consequences
for comfort. 2.2. Experimental protocol
Electrodes may be positioned at opposite ends of the
muscle and at motor points, and should be placed over Prior to conducting the experiment, a trial period of
muscle bulk rather than a bony prominence, cuts or approximately 20 min of electrical stimulation allowed
skin irritation [1,11].
There also exists large inter-subject variability in pre- 1
BMR Ltd., Galway, Ireland.
2
dicting surface regions of greatest excitability, for PALS suppliers.
876 G.M. Lyons et al. / Medical Engineering & Physics 26 (2004) 873–878

Site 2 The cathode electrode was placed at the

centre of the lateral side of the muscle belly
and the anode electrode was placed at the
centre of the medial side of the muscle belly
(Fig. 1b).
Site 3 The cathode was located between the centre
of the two muscle heads. The anode was
placed laterally to the centre of the lateral
head (Fig. 1c). Stimulation of only a single
muscle head, the lateral head, occurred with
this placement.
Site 4 The cathode was located at the centre of the
two muscle heads. The anode was placed
medial to the medial head (Fig. 1d). This
placement stimulated the medial head of the
gastrocnemius only.

Comfort was assessed by gradually increasing the

stimulation amplitude and asking the subjects to indi-
cate the stimulation amplitude corresponding to the
initial perception of the stimulus sensation (sensory
threshold), when a minimal muscle contraction was
observed (motor threshold), the onset of discomfort
(pain threshold) and the amplitude at which the dis-
comfort became unbearable (pain tolerance). The two
stimulation amplitude settings for pain threshold and
pain tolerance were recorded.

2.3. Statistical analysis

For all test conditions, 95% confidence intervals with

the p-value set at 0.05 were calculated. The means for
the two placement sites and two electrode sizes most
favoured were compared using repeated measures two-
way ANOVA. All statistical analyses were carried out
using SPSS, Version 9.

3. Results
Fig. 1. Placement of electrodes for sites 1, 2, 3 and 4.
It was extremely difficult to obtain a contraction
using stimulation sites 3 and 4, thus these electrode
the subject to become accustomed to the sensation of placement sites were deemed unsuccessful.
SNMES. The data for placement sites 1 and 2 are summarised
Four different electrode placement sites were investi- in Table 1, which shows that an electrode area of
gated: two of the sites stimulated the whole of the 19.63 cm2 provides superior comfort performance to
gastrocnemius muscle (sites 1 and 2) and two stimu- the other configurations for the two comfort para-
lated only half of the muscle. meters (pain threshold and pain tolerance).
The placement sites investigated were: Table 2 shows the mean and standard deviation for
the two placement sites and two electrode areas.
Site 1 The cathode (negative) electrode was placed The results show that site 1 was the favoured elec-
high on the calf, distally about 1.25 cm below trode placement with an electrode area of 19.63 cm2.
the proximal end of the muscle heads and the Site 1 positioned the cathode electrode high on the calf,
anode (positive) electrode was placed well distally about 1.25 cm below the proximal end of the
above the Achilles tendon, towards the end of muscle heads and the anode electrode placed well
the muscle belly, to maximise comfort (Fig. 1a). above the Achilles tendon, towards the end of the mus-
 G.M. Lyons et al. / Medical Engineering & Physics 26 (2004) 873–878 877

Table 1
Percentage of amplitude corresponding to pain threshold (PTH) and pain tolerance (PTO) for sites 1 and 2 for electrode areas of 19.63 and 38.48 cm2

Sex Age Electrode area #1, 19.63 cm2 (round) Electrode area #2, 38.48 cm2 (round)
Electrode site #1 Electrode site #2 Electrode site # 1 Electrode site # 2
PTH (%) PTO (%) PTH (%) PTO (%) PTH (%) PTO (%) PTH (%) PTO (%)
F 22 50 72 48 64 40 64 38 58
F 19 55 76 52 60 46 58 42 60
F 19 33 60 45 70 36 65 32 55
F 23 55 80 40 54 50 68 40 64
F 22 50 85 45 68 42 72 38 60

M 20 55 86 45 80 45 70 50 68
M 26 62 88 40 74 40 62 42 60
M 31 50 70 50 65 38 56 36 62
M 25 65 90 56 88 52 74 46 65
M 23 55 75 40 65 42 65 35 55
M 23 60 85 55 76 50 64 46 70
M 23 52 80 42 64 48 70 42 65
Site 1: The cathode electrode placed high on the calf, distally about 1.25 cm below the proximal end of the muscle heads and the anode electrode
placed well above the Achilles tendon, towards the end of the muscle belly.
Site 2: The cathode electrode placed at the centre of the lateral side of the muscle belly and the anode electrode placed at the centre of the medial
side of the muscle belly.

cle belly. Pain or discomfort was not experienced for ger electrode, thus the electrode area of 38.48 cm2 for
this site using the smaller electrode size until approxi- site 2 was perceived as least comfortable.
mately 53% of maximum amplitude (pain threshold) Analysis of pain threshold, for the measures of elec-
was reached. This is compared to an electrode area of trode area and electrode site are shown to be signifi-
cant (p < 0:005 and p ¼ 0:007, respectively). The mean
38.48 cm2 for the same electrode placement site, where
for electrode area #1 was 49.6 and for electrode area
pain threshold was approximately 44% of maximum
#2 was 42.3. The mean for electrode site 1 was 48.4
amplitude. In addition, the percentage amplitude set- and for electrode site 2 was 43.5. Analysis of pain tol-
ting corresponding to pain tolerance was higher for the erance measurements also showed significant differ-
smaller electrode area, 79% compared to 66%, thus ences between electrode areas and electrode sites
these subjects could withstand higher levels of stimu- (p < 0:005 and p ¼ 0:001, respectively).
lation with the smaller size electrodes.
Site 2 positioned the cathode electrode at the centre
of the lateral side of the muscle belly and the anode 4. Discussion
electrode placed at the centre of the medial side of the
The results from this study show that changing
muscle belly. With site 2, pain threshold was reported
either the electrode placement site or the electrode area
at 46% and 41% of maximum amplitude for electrode results in a change in pain threshold and pain tolerance
areas of 19.63 and 38.48 cm2, respectively. Pain toler- (p ¼ 0:001 and p < 0:005, respectively). A similar study
ance was 69% of maximum amplitude for the smaller by Alon et al. [20], also looked at stimulation of the
electrode and 62% of maximum amplitude for the lar- gastrocnemius muscle, concluded that the two largest

Table 2
Mean and standard deviation (in brackets) of amplitude corresponding to pain threshold and pain tolerance for sites 1 and 2 for an electrode area
of 19.63 and 38.48 cm2

Comfort parameters Electrode area #1 Electrode area #2

Electrode site #1, mean, Electrode site #2, mean, Electrode site #1, mean, Electrode site #2, mean,
% (SD) % (SD) % (SD) % (SD)
Pain threshold (PTH) 52.7 (7.84) 46.5 (5.73) 44.1 (5.20) 40.6 (5.16)
Pain tolerance (PTO) 78.9 (8.74) 69.0 (9.27) 65.7 (5.43) 61.8 (4.75)
Site 1: The cathode electrode placed high on the calf, distally about 1.25 cm below the proximal end of the muscle heads and the anode electrode
placed well above the Achilles tendon, towards the end of the muscle belly.
Site 2: The cathode electrode placed at the centre of the lateral side of the muscle belly and the anode electrode placed at the centre of the medial
side of the muscle belly.
878 G.M. Lyons et al. / Medical Engineering & Physics 26 (2004) 873–878

electrodes tested (20.25 and 40.3 cm2) were more Downey (CA): Rancho Los Amigos Research and Education
comfortable and produced greater muscle force gener- Institute; 1993.
[2] Lyons GM, Sinkjaer T, Burridge JH, Wilcox D. A review of
ation than the smaller sizes (2.25 and 9 cm2). Both lar- portable FES-based neural orthoses for the correction of drop
ger electrode sizes produced the same force of foot. IEEE Transactions on Neural Systems and Rehabilitation
contraction and comfort, therefore the 20.25 cm2 elec- Engineering 2002;10(4):260–79.
trode appeared to be the size of choice, as the 40.3 cm2 [3] Merletti R, Zelaschi F, Latella D, Galli M, Angeli S, Sessa MB.
electrode did not add to optimal excitation or improve A control study of muscle force recovery in hemiparetic patients
during treatment with functional electrical stimulation. Scandi-
comfort levels. The findings in this present study differ navian Journal of Rehabilitation Medicine 1978;10(3):147–54.
from Alon et al. [20] as a significant difference was [4] Kralj A, Bajd T, editors. Functional electrical stimulation:
shown in favour of the smaller electrode (19.63 cm2), standing and walking after spinal cord injury. Boca Raton, FL:
this may be attributable to the different parameters CRC Press; 1989.
applied in these studies. This study used a pulse dur- [5] Baker LL, Yeh C, Wilson D, Waters RL. Electrical stimulation
of wrist and fingers for hemiplegic patients. Physical Therapy
ation of 300 ls and a pulse frequency of 35 Hz, 1979;59(12):1495–9.
whereas Alon et al. [20] used a pulse duration of 200 [6] Gracanin F, Trnkoczy A. Optimal stimulus parameters for mini-
ls, and a pulse repetition rate of 50 Hz. The pulse dur- mum pain in the chronic stimulation of innervated muscle.
ation and pulse frequency used for our study were Archives of Physical Medicine and Rehabilitation 1975;56:243–9.
selected as more appropriate for comfort, based on evi- [7] Bowman B, Meadows P, Su N. Effects of stimulus pulse dur-
ation on comfort during controlled motor contractions. In:
dence in the literature as discussed previously. Popovic D, editor. Proceedings of the Eighth International Sym-
To conclude, of the four stimulation electrode sites tes- posium on ECHE, Dubrovnik. Belgrade: Tanjun; 1984, p. 429–
ted for comfort, the optimum stimulation site is electrode 39.
site 1, where the cathode electrode is placed high on the [8] Bowman BR, Baker LL. Effects of waveform parameters on
calf, distally approximately 1.25 cm below the proximal comfort during transcutaneous neuromuscular electrical stimu-
lation. Annals of Biomedical Engineering 1985;13(1):59–74.
end of the muscle heads and the anode is placed well [9] McNeal DR, Baker LL. Effects of joint angle, electrodes and
above the Achilles tendon, towards the end of the muscle waveform in electrical stimulation of the quadriceps and ham-
belly. Of the two electrode sizes tested (round with area strings. Annals of Biomedical Engineering 1988;16:299–310.
of 19.63 and 38.48 cm2), the smaller electrode was found [10] Kantor G, Alon G, Ho HS. The effects of selected stimulus
to be most comfortable and effective. waveforms on pulse and phase characteristics at sensory and
motor thresholds. Physical Therapy 1994;74(10):951–62.
These are important findings as stimulation of the [11] Robinson AJ, Snyder-Mackler L, editors. Clinical electro-
gastrocnemius muscle has wide application from gait physiology, electrotherapy and electrophysiologic testing. Wil-
correction to blood flow assistance using the calf mus- liams & Wilkins; 1994.
cle pump. As discomfort can have a profound effect on [12] Mourselas N, Granat MH. Evaluation of patterned stimulation
treatment programmes, leading some subjects to totally for use in surface functional electrical stimulation systems. Medi-
cal Engineering and Physics 1998;20:319–24.
reject SNMES as a treatment option and others to not [13] Torebjork HE, Hallin RG. Perceptual changes accompanying
tolerate the optimal or necessary stimulus amplitude controlled preferential blocking of A and C fiber responses in
and treatment duration, by delivering SNMES in a intact human skin. Experimental Brain Research 1973;16:321–
more comfortable fashion, using the findings of this 32.
study, more subjects may be able to fully benefit from [14] Belanger AY, Allen ME, Chapman AE. Cutaneous versus mus-
cular perception of electrically evoked tetanic pain. Journal of
SNMES of the gastrocnemius. Orthopaedic and Sports Physical Therapy 1992;16(4):162–8.
[15] Reed B. The physiology of neuromuscular electrical stimulation.
Pediatric Physical Therapy 1997;9:96–102.
Acknowledgements [16] Baker LL, Cole K, Hart J, Smith D, Wong T, McNeal DR.
Effects of duty cycle and frequency on muscle fatigue during iso-
The authors would like to acknowledge the assistance metric electrically stimulated quadriceps contractions. Physical
Therapy 1988;68:835.
of BMR Ltd., Galway, Ireland who provided funding
[17] Rattay F. Modeling the excitation of fibers under surface elec-
for the completion of this project. We would also like to trodes. IEEE Transactions on Biomedical Engineering
acknowledge the assistance of Dr. Jean Saunders, Man- 1988;35(3):199–202.
ager Statistical Consulting Unit, University of Limerick [18] Alon G. High voltage stimulation: effects of electrode size on
who provided statistical advice on this study. basic excitatory responses. Physical Therapy 1985;65:890–5.
[19] Patterson RP, Lockwood JS. The current requirements and the
pain responses for various sizes of surface stimulation electrodes.
Proceedings IEEE-EMBS 1991;13:1809–10.
References [20] Alon G, Kantor G, Ho HS. Effects of electrode size on basic
excitatory responses and on selected stimulus parameters. Jour-
[1] Baker LL, McNeal DR, Benton LA, Bowman BR, Waters RL. nal of Orthopaedic and Sports Physical Therapy 1994;20(1):
Neuromuscular electrical stimulation—a practical guide, 3rd ed. 29–35.