The 2014 Biomedical Engineering International Conference (BMEiCON-2014)

Transcutaneous Electrical Nerve Stimulation based

pain control system via Q-EEG validation
Pradkij Panavaranan Yodchanan Wongsawat, Ph.D.
Department of Biomedical Engineering Department of Biomedical Engineering
Faculty of Engineering, Mahidol University Faculty of Engineering, Mahidol University
Nakhon Pathom, Thailand Nakhon Pathom, Thailand
[email protected] [email protected]

Abstract—Pain is unpleasant sensory which is always a in the spinal cord. The result is a second-order neurons is
subjective measurement. Various studies try to find a solution to inhibited with the gate control theory [7].
have life without pain. In critical situation, the treatment of burn Pain is always a subjective measurement. The traditional
patients need some criteria to understand a level of pain pain level evaluation uses a questionnaire to describe a feeling
sensation. This study focuses on an acute thermal pain. The
of pain. Some technician uses some symbols, pictograms, and
purpose of this study is to block the pain pathway using the
transcutaneous electrical nerve stimulation (TENS). The results numeric rating scale. Visual Analog Scale (VAS) is a numeric
are validated using the visual analog scale (VAS), oxygen rating scale that mostly uses for measuring level of pain [8].
saturation (SpO2), pulse rate (PR), and the quantitative The Quantitative Electroencephalogram (Q-EEG)
electroencephalogram (Q-EEG). The experiment is taken with measures the neural connectivity in cortical and sub-cortical
normal healthy volunteers. During the experiment, acute thermal dysfunction from the EEG [9]. The Z-scored FFT training of
pain is stimulated by a hot thermal pad which can generate sixty QEEG monitors the distribution of EEG data which has
degree Celsius. The TENS generates electrical stimuli for pain absolute power, amplitude asymmetric, coherence, and phase
alleviation in high frequency (around a hundred hertz). The lag. The topographic brain mapping displays the quantitative
result of pain sensation during the used of TENS is mild while
information about brain activity. Recent clinical indicates
without using it give modulate sensation. The resulting oxygen
saturation, pulse rate as well as the Q-EEG are also revealed related experiences condition with Q-EEG analysis including
synchronous results. pain.
Keywords—Quantitative-EEG, Transcutaneous Electrical The propose study is to investigate acute thermal pain via
Nerve Stimulation, acute thermal pain, Visual Analog Scale, SpO2, Q-EEG validation compare between using TENS and without
Pulse Rate. using TENS. The study proposes the pain marker and the
position of TENS stimulation for alleviation acute thermal
I. INTRODUCTION pain.
Pain is a human sensation which refers to an unpleasant
sensory, discomfort, illness, and disease. This sensation is an II. MATERIAL AND METHODS
integration of sensory, perceptual, affective, and cognitive
processing in the brain [1]. Pain categories in two groups of a A. Paticipants
duration criteria which is an acute pain, and a chronic pain [2]. The participants were normal healthy volunteers (n= 5; 5
This study focuses on the acute thermal pain that happened male). The age-wise was recruited in this study (mean±SD;
from burnt. The nature of this acute pain is described as 23.67±6.71 years). All of the participants were right-handed,
lancing, tingling, shooting and like electric shocks [3]. Burn drug-free, and no mental disorder.
patients have highly affected by side effects of a B. Living Environment
pharmacological treatment [3].
The sensation of pain happens when the noxious stimuli The room size was 2.45×2.45×2.00 m (weight × length ×
are stimulated in a nociceptive nerve ending. Free nerve height). The room temperature was averaged in 25 degree
ending, which has a myelinated fiber and non-myelinated Celsius and normal humidity. The perceiving color was
fiber, produces an action potential to sense the pain signal performed in black wall. The light (Philips Inc.) was installed
along ascending pain pathway [4]. Pain signal travels along at the center of the room. Light produced 6500K temperature
spinal cord, medulla, thalamus, and somatosensory cortex [5]. color with 806 Lumen. Fig 1 was showed the living
The Transcutaneous Electrical Nerve Stimulation (TENS) environment.
is an alternative medicine for the therapeutic proposes [6]. C. Pain Stimulation
TENS electrode contact with the surface of the skin with non-
The pain stimulation was performed in an acute thermal
invasive method. They produce high frequency electrical
pain sensation. The pain stimulator was a hot thermal pad
stimuli. The high frequency electrical stimuli activates C
which can control a heat temperature around 30-80 degree
fibers, an inhibitory fibers that blocks the pain signal pathway
978-1-4799-6801-5/14/$31.00 ©2014 IEEE

Authorized licensed use limited to: NED UNIV OF ENGINEERING AND TECHNOLOGY. Downloaded on October 28,2024 at 08:46:57 UTC from IEEE Xplore. Restrictions apply.
 Celsius. The pad size was 30×45 cm. This stimulator
contacted to right forearm of the participants. The average of
temperature was performed in 65 degree Celsius.
D. Transcutaneous Electrical Nerve Stimulation Fig. 3. The numeric ranking scale was represented to index the level of pain
on the Visual Analog Scale.
Transcutaneous Electrical Nerve Stimulation (TENS) was
an alternative medicine which designed to reduce pain in a
therapeutic purposes. The TENS designed to produce a high 30 secs 30 secs 30 secs
frequency (mean±SD; 101.6±0.0 Hz), and adjustable voltage
intensity as in Fig 2. The electrode of the TENS warped
around the right upper forearm of each subject. Participants
were adjusted TENS themselves in a tingling sensation.
5mins 5mins 5mins 5mins
E. Data Collection Baseline ES Hot Hot+ES

The collected data were a Pain questionnaire, Visual

Analog Scale (VAS), Oxygen saturation (SpO2), Pulse Rate, Fig. 4. The procedure of experiment. VAS was asked after each testing.
Quantitative-Electroencephalogram (Q-EEG). (“Baseline” do nothing, “ES” stimulated by the TENS, “Hot” stimulated by
the hot thermal pad, “Hot+ES” stimulated by both of the TEN and the hot
The Pain questionnaire and VAS were collected from an thermal pad)
answer of each participant. The questionnaire asked to explain
the participants’ feeling after each testing. The VAS was a
visual imaginary to guide a level of pain (0 means no F. Feature Extraction
sensation of pain, 4-6 means mild pain, and 10 means severe The Pain questionnaire was extracted to describe the pain
pain sensation). The VAS was shown in Fig. 3. sensation. The VAS was extracted by using the average of the
The Oxygen saturation and Pulse Rate were measured by data and the standard deviation.
The SpO2 and pulse rate were extracted by the statistical
the Radical7® (Masimo). The sampling frequency was 0.5 Hz. method which included an average of data, a standard
The collected data were analyzed respectively in a min deviation (SD), and P-values (n = 5; 5 male).
interval. The Q-EEG was analyzed by Z-scored FFT and a
The Q-EEG measured a function of brain activity related to normative database. The sampling of Q-EEG was 256 Hz, and
participants’ behavior. The BrainMaster discovery 24E was used a fifth order of IIR Butterworth bandpass filter at 1-40Hz.
used to record 19-channels EEG (a montage of the 10-20 The Z-scored FFT was calculated by a mean of the population
international system electrode placement). The EEG channels divided by the standard deviation of the population as in (1).
included Fp1, F3, C3, P3, O1, F7, T3, T5, Fp2, F4, C4, P4,
O2, F8, T4, T6, Cz, and Pz. The bandpass filter was 0.5-40 Hz Z-scored FFT = (1)
and the sampling frequency was 256 Hz. The Z-Scored FFT
was used to analysis Q-EEG by the NeuroGuide software. The Equation (2) was the amplitude asymmetry which the A
Q-EEG results showed an absolute power, a relative power, an and B were represented an instantaneous value of a given
estimate EEG.
amplitude asymmetry, a coherence, and a phase lag.
Asymmetry = (2)

The coherences were computed as in (3). The Numerator

was the cross-spectrum between two signals. The denominator
was the auto-spectrum of each individual signal.

Coherence = (3)

Fig. 1. The living environment was in a black color wall. The Phase lag was calculated as in (4). The ‘a’ and ‘b’ was
determined by Fourier Transform which ‘b’ was an imaginary
number and ‘a’ was a real number.

Phase lag = arctan( ) – arctan( ) (4)

Finally, the topographical brain mapping was displayed the

color-based statistical analysis which included the absolute
power, the amplitude asymmetry, the coherence, and the phase
lag.

Fig. 2. The TENS electrical waveform recorded from the oscilloscope.

Authorized licensed use limited to: NED UNIV OF ENGINEERING AND TECHNOLOGY. Downloaded on October 28,2024 at 08:46:57 UTC from IEEE Xplore. Restrictions apply.
 G. Experiment procedure
Participants were seated at the center of the living
environment with a comfortable position. The sensor of SpO2
was connected to the left index finger. The Electrode of the Q-
EEG scalp put on their head and connected to the
BrainMaster. The surface electrode of TENS connected to the
right upper forearm of the participants. The Pain stimulation
was contacted to the right forearm. Fig. 4 was a summarized
of the experimental procedure. Fig. 5. The visual Analog Scale of each testing condition.
III. EXPERIMENTAL RESULTS TABLE I. OXYGEN SATURATION
A. Pain questionnaire and Visual Analoge Scale Testing condition
Subject
The result of the pain questionnaire about their feeling to Baseline ES Hot Hot+ES

the TENS was mild and tingling sensation. The feeling of 97.04 97.67 97.58 96.99
1
±0.27 ±0.58 ±0.50 ±0.68
thermal sensation was mild and modulate sensation. The
98.09 98.72 99.00 98.97
feeling of both TENS and thermal sensation was mild and 2
±0.30 ±0.45 ±0.00 ±0.23
modulate sensation.
95.25 96.02 96.13 96.22
Analysis of statistical distribution of VAS was shown in 3
±0.55 ±0.59 ±0.58 ±0.62
each testing condition, Baseline (mean±SD; 0.00±0.00), and 97.31 97.74 97.75 97.42
ES (mean±SD; 3.08±0.80), and Hot (mean±SD; 4.67±1.63), 4
±0.51 ±0.51 ±0.45 ±0.86
and Hot+ES (mean±SD; 4.75±1.72). Analysis of variance 97.86 97.95 97.68 97.47
5
revealed that the difference of the mean in each testing ±0.40 ±0.28 ±0.76 ±0.53
(Values represent in mean±SD)
condition of ES (p=0.000113), and Hot (p=0.0004584), and
Hot+ES (p=0.000543). Analysis of variance in between each TABLE II. PULSE RATE
testing condition revealed that the mean difference of ES to Testing condition
Hot (p=0.045417), ES to Hot+ES (p=0.0415049), Hot to Subject
Baseline ES Hot Hot+ES
Hot+ES (p=0.455692). The result from VAS was shown in
71.12 70.81 75.23 75.50
Fig.5. 1
±2.96 ±3.33 ±5.83 ±4.16
B. Oxygen saturation and Pulse Rate 2
75.94 72.75 74.17 74.13
Statistical analysis of oxygen saturation in each testing ±2.24 ±3.36 ±2.53 ±3.40
condition respectively revealed the mean and the average of 75.95 77.66 76.89 79.70
3
±4.03 ±3.25 ±2.81 ±3.05
the baseline, ES, HOT, and HOT+ES as in Table I, Table II.
75.74 69.91 70.50 75.89
In oxygen saturation, variance analysis revealed that the 4
±3.37 ±3.91 ±3.67 ±4.98
mean difference in each testing condition of 59.30 60.07 57.89 62.42
participant 1; ES (p=4.38E-23), Hot (p=5.16E-21), 5
±2.95 ±2.86 ±3.52 ±3.53
Hot+ES (p=0.16682), (Values represent in mean±SD)

participant 2; ES (p=7.24E-30), Hot (p=2.08E-12), C. Quantitative Electroencephalogram

Hot+ES (p=8.40E-59),
The Absolute Power Q-EEG result was represented
participant 3; ES (p=1.20E-32), Hot (p=4.66E-36),
(selected participant) in Fig 6.
Hot+ES (p=2.90E-32), Absolute Power
participant 4; ES (p=2.62E-10), Hot (p=3.11E-11),
Hot+ES (p=0.08646)
participant 5; ES (p=0.01847), Hot (p=0.00197),
Hot+ES (p=1.01E-15)
In pulse rate, variance analysis revealed that the mean
difference in each testing condition of
participant 1, ES (p=0.18907), Hot (p=2.40E-12),
Hot+ES (p=2.00E-30),
participant 2, ES (p=3.40E-17), Hot (p=3.05E-12),
Hot+ES (p=1.80E-08),
participant 3, ES (p=6.14E-05), Hot (p=2.03E-23),
Hot+ES (p=2.00E-13),
participant 4, ES (p=5.04E-38), Hot (p=3.11E-11),
Hot+ES (p=0.37432)
participant 5, ES (p=0.00212), Hot (p=7.82E-06), Fig. 6. The absolute power of the Q-EEG of each testing conditon.
Hot+ES (p=1.56E-16)

Authorized licensed use limited to: NED UNIV OF ENGINEERING AND TECHNOLOGY. Downloaded on October 28,2024 at 08:46:57 UTC from IEEE Xplore. Restrictions apply.
 The coherence of Q-EEG of each testing condition was V. CONCLUSION
represented in Fig. 7. TENS stimulated a high frequency at 101.6 Hz for
alleviating hot thermal sensation has been investigated, and
Coherence
validated by VAS, Oxygen saturation, Pulse Rate, and Q-
EEG.
The VAS showed the feeling of ES was a mild sensation to
compare with Hot, Hot+ES were moderate sensational. The
means and standard deviation of each testing condition was
generally similar to every participant.
The TENS related acute thermal pain has been validated
via QEEG. The Topographical Brain Mapping of the absolute
power decrease in beta band of hot condition. The result of
coherence shows the increase of brain connectivity in each
testing condition. The result of phase lag shows the phase
difference significantly change.
VI. ACKNOWLEDGEMENT
The project is supported by the government funding of
Fig. 7. The coherence of Q-EEG showed in each testing condition. Mahidol University.

Lastly, the phase lag of Q-EEG of each testing condition VII. REFERENCES
were represented in Fig. 8. [1] N. AndrewC, P. Chen Rappelsberger, Brain and Human Pain:
Topographic EEG Amplitude and Coherence Mapping. Brain
topography, Volume7, Number 2, 1994.
Phase lag
[2] H. Hoffman, G. Garcia-Palacios, A. Kapa, V., Beecher, J., & Sharar, S.
R., (2003). Immersive virtual reality for reducing experimental ischemic
pain: International Journal of Human-Computer Interaction Vol 15(3)
2003.
[3] P. Richardson, L. Mustard, The management of pain in the burns unit,
Burns, Volume 35, Issue 7, November 2009.
[4] F. , L.S., A New method To Quantitatively Describe Pain Behavior in
Infants. Nursing Research, 35(1). p. 28-31, 1986.
[5] F. Freudenrich, Ph.D., Craig. "How Pain Works" 09 November 2007.
[6] L. S Chesterton, N. E Foster, C. C Wright, G. David Baxter, Panos
Barlas, Effects of TENS frequency, intensity and stimulation site
parameter manipulation on pressure pain thresholds in healthy human
subjects, Pain, Volume 106, Issues 1–2, November 2003.
[7] C. Chung Chen, M. I. Johnson, An Investigation Into the Hypoalgesic
Effects of High- and Low-Frequency Transcutaneous Electrical Nerve
Stimulation (TENS) on Experimentally-Induced Blunt Pressure Pain in
Healthy Human Participants, The Journal of Pain, Volume 11, Issue 1,
Fig. 8. The phase lag of Q-EEG of each testing condition. January 2010.
[8] A.E.E. de Jong, M. Bremer, M. Schouten, W.E. Tuinebreijer, A.W.
Furthermore, there were more Q-EEG results which could Faber, Reliability and validity of the pain observation scale for young
be presented on request. children and the visual analogue scale in children with burns, Burns,
Volume 31, Issue 2, March 2005.
IV. DISCUSSION [9] T. H. Budzynski, H. Kogan Budzynski, J. R. Evans and A. Abarbanel,
Introduction, In Introduction to Quantitative EEG and Neurofeedback
The result of the VAS shows that the sensation of the ES (Second Edition), edited by Thomas H. Budzynski, Helen Kogan
and Hot has high contrast in order to compare with the ES and Budzynski, James R. Evans and Andrew Abarbanel, Academic Press,
both stimuli. San Diego, 2009.
The result of the oxygen saturation and the pulse rate show [10] F. T. Nickel, F. Seifert, S. Lanz, C. M., Mechanisms of neuropathic pain,
European Neuropsychopharmacology, 2012.
that each participant has subjective measurement. However,
[11] Y. Meng Xu, H. Ge, and L. Arendt-Nielsen, Sustained Nociceptive
the statistical result shows a low variance between baseline Mechanical Stimulation of Latent Myofascial Trigger Point Induces
and both stimuli. Central Sensitization in Subjects, The Journal of Pain, Vol 11, No. 12
The absolute power of Q-EEG show the power in each (December): pp 1348-1355, 2010.
location of the brain. The coherence shows the connectivity of [12] R. Reuven Nir, A. Sinai, E. Raz, E. Sprecher, D. Yarnitsky. Pain
the brain relate to acute thermal pain sensations. The Phase lag assessment by continuous EEG: Association betwwen subjective
perception of tionic pain and peak frequency of alpha oscillations during
of Q-EEG show the phase difference between each channel of stimulation and at rest, Brain research 1344, p. 77-86, 2010.
EEG.

Authorized licensed use limited to: NED UNIV OF ENGINEERING AND TECHNOLOGY. Downloaded on October 28,2024 at 08:46:57 UTC from IEEE Xplore. Restrictions apply.