=Paper=
{{Paper
|id=Vol-3200/paper26
|storemode=property
|title=Information Technology for Identification of Electric Stimulating Effects Parameters
|pdfUrl=https://ceur-ws.org/Vol-3200/paper26.pdf
|volume=Vol-3200
|authors=Volodymyr Fedorchenko,Igor Prasol,Olha Yeroshenko
}}
==Information Technology for Identification of Electric Stimulating Effects Parameters ==
https://ceur-ws.org/Vol-3200/paper26.pdf
Information Technology for Identification of Electric Stimulating
Effects Parameters
Volodymyr Fedorchenko, Igor Prasol and Olha Yeroshenko
Kharkiv National University of Radio Electronics, Nauky Ave. 14, Kharkiv, 61166, Ukraine
Abstract
A wide range of modern therapeutic devices based on various physical principles, widely used in
medicine, cosmetology, sports. Among them, electric massage devices occupy a worthy place,
alternative to classic manual massage. Therapeutic electromassage procedures are popular,
convenient and beneficial for the recovery of the body. They are widely used in the treatment of
chronic diseases of the circulatory system, musculoskeletal system, internal organs, etc. The
restoration of damaged muscles is especially effective, provided that the parameters of
stimulating effects are chosen correctly. Therefore, in this work, it is proposed to use an
information method for studying the neuromuscular system based on electromyography.
The parameters of the stimulating effect do not always optimally correspond to a specific patient
or a selected area of the body, which leads to insufficient effectiveness of therapeutic procedures,
prolongation of rehabilitation. Elimination of shortcomings is possible due to the adjustment of
the parameters of electrical stimuli depending on the data of myographic studies of a particular
patient.
Based on the data obtained by EMG, specific parameters of stimulating effects (electrical
impulses) are selected, such as amplitude, frequency, duty cycle, etc., which makes it possible to
implement a technical device for carrying out rehabilitation procedures. Therefore, an
electromassage apparatus is proposed, built on the basis of a modern microcontroller, which
allows, on the basis of EMG data, to change stimulating impulses of exposure in a fairly wide
range, thereby realizing an individual approach to each patient and increasing the efficiency of
therapeutic procedures.
Keywords 1
Biomedical parameters, electromyostimulator, total electromyography, electromyogram,
neuromuscular system, musculoskeletal system, time-frequency analysis
1. Introduction The effectiveness of the use of electrotherapy
devices is largely based on the use of methods and
means of diagnostic support, which would give
In the modern world, the number of factors
objective information about the patient's
negatively affecting human health is becoming
condition, contributing to the successful solution
more and more.The human body ceases to have
of the problem localization of zones of influence
time to heal itself.All this requires a search for
for electrostimulation, correct setting and
new combinations of recovery methods., when
achievement of treatment goals.
medical devices are used in conjunction with drug
In order to improve the quality and speed of
methods, implementing various types of
treatment, system development required, in which
electrotherapy.
III International Scientific And Practical Conference “Information
Security And Information Technologies”, September 13–19, 2021,
Odesa, Ukraine
EMAIL: volodymyr.fedorchenko@nure.ua (A. 1);
igor.prasol@nure.ua (A. 2); olha.yeroshenko@nure.ua(A. 3)
ORCID: 0000-0001-7359-1460 (A. 1); 0000-0003-2537-7376
(A. 2); 0000-0001-6221-7158 (A. 3)
©️ 2021 Copyright for this paper by its authors. Use permitted under Creative
Commons License Attribution 4.0 International (CC BY 4.0).
CEUR Workshop Proceedings (CEUR-WS.org)
automation will be provided, allowingprovide the
most effective treatment result.
The ultimate goal of creating an automated
electrotherapy system is to develop modeling
methodsand research of control systems and
devices percutaneous electroneurostimulation,
characterized by adaptation to changes in
biological objects.
The novelty is the development of a
methodology for analyzing the functions of Figure 1: Dependence of the signal amplitude
electrostimulating devices, which makes it and the excitation threshold of the
possible to minimize negative effects during the neuromuscular structure( а) - muscle fiber, б) –
stimulation procedure. muscle, 1) subthreshold stimulus, 2) threshold
stimulus, 3) submaximal suprathreshold
2. Electrostimulation stimulus, 4) maximum suprathreshold stimulus)
Electrical stimulation in this approach causes The dependence of the amplitude of muscle
minimal changes in the treated area of the skinand contraction on the strength of the stimulus occurs
nearby tissues, which allows to increase the according to the law of power relations:
efficiency of the treatment process. • Each excitatory tissue has its own
Skeletal muscle electrical stimulation, which functional reserve.
are the basis of the musculoskeletal system, gives • Each excitatory tissue has its own
a positive healing, preventive and training effects. functional boundary.
During electrical stimulation of the With the help of electrical stimulation, you can
neuromuscular system, a rational choice of modes temporarily change the muscle composition. In all
is importantand a combination of tonic and kinetic cases, the strength of the electrical stimulation
contractions, which significantly affect the current depends on its density per unit area of the
increase in mass, development of strength, electrodes, resistance value on the electrode-skin
increased excitability and muscle performance [1, section, excitability of those muscles, which are
2]. subject to stimulation and individual
Electrical stimulation is successfully characteristics of the human body. The most
combined with traditional drug therapy. To important factor that determines the shape of
enhance metabolic and trophic processes, muscle muscle contraction is the frequency of irritation.
tissue stimulation is performed using targeted Single muscle contractions are possible only with
stimulation and contraction of a specific muscle a low frequency of irritation. At a high frequency
group. of stimulation, the muscle contracts tetanically.
An important property of neuromuscular For human skeletal muscles, the optimum
structures when irritated by electric currents, the frequency of stimulation is different. The
dependence of excitability on the rate of change optimum frequency of irritation also changes
in the amplitude of the stimulating signal [1]. when the state of the body changes (fitness, time
Depending on the signal amplitude and the of day, previous load, etc.). Pulse electric current
excitation threshold of the neuromuscular used in electrostimulation has a wide variety of
structure, the following electrostimulation modes characteristics (frequency, shape, pulse duration,
are distinguished: subthreshold, threshold and character of the current, the ratio of the periods of
suprathreshold(fig. 1) [3-5]. stimulation and pauses, etc.), which leads to a
great variety of options for conducting electrical
stimulation of the locomotor system. Optimal
electrical stimulation is only possible if, when the
repetition rate and shape of stimulating signals
correspond to the physiological properties of
neuromuscular structures.
3. Characteristics of muscle
contraction
The strength and speed of contraction are • Relative strength - the ratio of maximum
important characteristics of a muscle. The strength to anatomical diameter.
equations expressing these characteristics were The greatest work and power is achieved at
empirically obtained by A. Hill and subsequently medium loads [4-6].
confirmed by the kinetic theory of muscle
contraction (Deshcherevsky's model). 4. Electromyographic signal
Hill's equation, which relates the strength and
speed of muscle contraction, has the following
processing method
form [6-8]:
For a qualitative and quantitative assessment
(P+a)(v+b) = (P0+a)b = a(vmax+b), (1)
of the state of the human neuromuscular system
where v – muscle shortening rate; P – muscle
using electromyogram (EMG) the information
force or load applied to it; vmax - maximum speed method of time-frequency analysis based on
of muscle shortening; P0 - strength developed by a spectrograms can be used (fig. 2, fig. 3) [6-12].
muscle in isometric contraction mode; a, b - This method is implemented on the basis of the
constants. fast windowed Fourier transform. In this case, the
The total power developed by the muscle is signal is divided into time intervals ("windows")
determined by the formula: of short duration, within which it can be
Ngen = (P+a)v = b(P0-P). (2) considered stationary. Time intervals are called
The muscle efficiency remains constant (about quasi-stationary segments, and the approach to
40%) in the range of strength values from 0,2 P0 processing is analysis over short intervals [13-30].
to 0,8 P0. In the process of muscle contraction, a The original signal on the selected segment is
certain amount of heat is released. This value is multiplied by the window function and undergoes
a fast Fourier transform in accordance with the
called heat production.
expression:
Heat production depends only on the change in
STFT = ∫ [x(t)⋅ω∗ (t –τ)] ⋅ e−2jftπ dt, (4)
muscle length and does not depend on the load.
where x(t) – original signal, ω(t) – window
Constants a and b have constant values for a given function, kτ – time shift amount, k – the ordinal
muscle. Constant a has the dimension of force, a number of the window shift, f – frequency, t –
and b - velocity. Constant b is highly temperature time, ω∗ (t) – complex conjugate window function
dependent. The constant a is in the range of values [13-14].
from 0,25 P0 to 0,4 P0. Based on these data, the Next, we obtain a portion of the spectrogram
maximum rate of contraction for a given muscle for the analyzed window by squaring the real part
is estimated. [7-13]: (amplitude) of the windowed Fourier transform:
vmax = b•( P0 / a). (3) X(t) = | STFT(τk,f)|2. (5)
Muscle strength depends on the morphological To conduct a quantitative analysis of EMG
properties and physiological state of the muscle: signals, it is necessary to calculate the following
• The original muscle length (resting length). parameters of the time-frequency representation
of the total EMG: lower and upper cutoff
The more the muscle is stretched at rest, the
frequency, median frequency, effective spectrum
stronger the contraction (Frank-Starling law).
width and a number of others [13-41].
• Muscle diameter or cross section. Allocate The average signal amplitude is also calculated
two diameters: by the formula:
- anatomical diameter - muscle cross 1
𝐴ср = ∑|𝐴 [𝑖 ]|, (6)
section. 𝑁
- physiological diameter - the where A [i] - amplitude of the i-th sample of the
perpendicular section of each muscle fiber. The registered signal, N – signal counts.
larger the physiological section, the more strength The lower and upper cutoff frequencies
the muscle has. determine the effective spectrum width, i.e., the
There are two types of muscle strength: frequency range in which at least 90% of the
signal power is concentrated [17-20]. The median
• Absolute strength - the ratio of maximum
is the frequency dividing the area under the energy
strength to physiological diameter. spectral density curve into two equal parts [16].
Determination of frequency parameters is
performed automatically based on the results of
calculating the spectrogram of the EMG signal.
For this, the value of the EMG signal energy in
each cell of the spectrogram is calculated:
𝐸 [𝑖, 𝑗] = 𝐴[𝑖, 𝑗]2 , (7)
[ ]
where A i, j - the amplitude of the
electromyogram in the i-th row and j-th column.
Next, we determine the median frequencyfm .
For this, a column with a serial number j is
allocated, which corresponds to the spectral
energy density of the signal at the j-th moment of
time.
Signal energy concentrated in effective Figure 3: Corresponding spectrogramof the
spectrum widthEэфф [j] is more than 90%, muscle m. biceps brachii
calculated by the formula:
𝐹 Let us represent the parameters of the EMG
𝐸эфф [𝑗] = 0,95 ∑ 𝐸 [𝑘, 𝑗]. (8) signal in the form of a certain finite set
𝑘=1 𝐴𝑚 = {𝑎𝑖 }(𝑖 = ̅̅̅̅̅̅
1, 𝑚), (9)
Lower cutoff frequencyfнj is determined from where А - the designation of this set; m –
the condition: the difference between the sum of cardinality multitudes; аi – elements of the set.
the elements of the column with indices from fнj The elements of the set can be amplitudes,
1 frequencies of the spectrum components, phase
tо fmj and the value Eэфф [j] minimal modulo. shifts, etc.
2
Upper cutoff frequencyfвj is determined from Let us represent the parameters of stimulating
the condition: the difference between the sum of influences also in the form of a finite set
the elements of the column with indices from fmj 𝐵𝑛 = {𝑏𝑖 }(𝑖 = ̅̅̅̅̅
1, 𝑛), (10)
1 where B - the designation of this set; n –
tо fвj and the value Eэфф [j] minimal modulo. cardinality multitudes; bi – elements of the set.
2
These processing parameters make it possible The elements of the set can be the amplitude
to fully assess the frequency content of the EMG and frequency of stimuli, the type of modulation,
signal. modulation parameters, time intervals, etc.
Thus, the task is to determine such a
transformation ω, which provides an
unambiguous display of the elements of the
number А to the corresponding elements В
𝜔
𝐴 →𝐵 , (11)
𝑚 𝑛
EMG signal processing allows for ongoing
monitoring the effectiveness of therapeutic effects
due to the optimal selection parameters of
stimulating effects.
5. Conclusions
Quantitative analysis of the total
Figure 2: Electromyogramof the muscle m. biceps electromyogram of the trained and untrained
brachii muscle m. biceps brachii revealed the following
patterns:
– the average amplitude of the EMG signal for
trained subjects reaches the highest values
345,62±148,10 μV, for the untrained equals
189,27±84,00 μV;
– the median frequency is characterized by the "KhPI", № 13 (1338). (2019)165–175.doi:
lowest values 111,44±27,62 Hz for trained 10.20998/2411-0558.2019.13.15
subjects and 123,00±30,07 Hz for untrained; [3] P. P. Pestrikov, T.V. Pestrikova, Measuring
– the upper cutoff frequency for trained system for recording signals from surface
subjects is 409,07±69,91 Hz and electromyography of forearm
446,90±66,22 Hz for untrained; muscles,Electronic scientific publication
"Scientific notes of PNU". Volume 10. No.
– the effective spectrum width for trained
2. (2019) 173–180.
subjects is 382,09±71,42 Hz and
[4] O. Yeroshenko, I. Prasol, O. Datsok,
415,92±65,35 Hz for untrained. Simulationofanelectromyographicsignalcon
Comparative analysis of the calculated verterforadaptiveelectricalstimulationtasks,
parameters for trained and untrained subjects: The current state of research and technology
indicators of the upper cutoff frequency, median in industry. № 1 (15). (2021) 113–119. doi:
frequency and effective spectrum width for 10.30837/ITSSI.2021.15.113
untrained subjects exceed the corresponding [5] S. S. Nikitin, Electromyographic stages of
values for trained subjects. the denervation-reinnervation process in
The proposed amplitude-frequency criterion neuromuscular diseases: the need for
allows one to take into account basic parameters revision, Neuromuscular diseases. Moscow.
of non-stationary bioelectric signal (amplitude №2. (2015) 16–24.
and frequency) and thus to carry out quick and [6] C. J.De Luca, The use of surface
electromyography in biomechanics, Journal
effective express diagnostics of the functional
of Applied Biomechanics. № 13 (2). (1997).
state of the neuromuscular system using [7] S. H.Roy, G. DeLuca, S.Cheng,
automated complexes of time-frequency A.Johansson, L. D.Gilmore, C. J.De Luca,
processing of EMG signals. Electro-Mechanical stability of surface EMG
Thus, carrying out qualitative and quantitative sensors, Medical and biological engineering
analyzes the structure of an EMG signal that is and computing. № 45. (2007).
unsteady in natureand the dynamics of its [8] M.Voelker Implantable EMG measuring
parameters in the process of muscle contraction is system, AMA Conferences. (2015).
performed based on the spectrogram, realizing [9] O. Yeroshenko, I. Prasol, O. Trubitsyn, and
graphical visualization of the amplitude, L. Rebezyuk, Organization of a Wireless
frequency and time components of the biomedical System for Individual Biomedical Data
signal in real time. Consequently, specific Collection, International Journal of
parameters of stimulating effects can be selected Innovative Technology and Exploring
based on the data of the EMG signal, which makes Engineering, vol. 9, no. 4, (2020)2418–
it possible to implement an effective technical 2421.doi: 10.35940/ijitee.D1870.029420
device for carrying out individual therapeutic [10] A. N. Osipov, S. K. Dick, K. G. Senkovsky,
procedures. Complex biotechnical feedback in
electrostimulation systems, Moscow:
6. References Medical technology, № 6. (2007) 27– 29.
[11] K. Jermey, Atlas of Musculoskeletal
[1] О. А. Yeroshenko, I. V. Prasol, Anatomy, AST Publishing House. (2008)
V. V. Semenets, About building a system of 382 p.
muscle electrical stimulation for cadets,The [12] S.H.Roy, G.Luca De, S.Cheng, A.Johansson,
use of information technology in the training L.D. Gilmore, C.J.Luca De, Electro-
and operation of law enforcement: materials Mechanical stability of surface EMG
International. scientific-practical conf. Mar sensors, Medical and biological engineering
14-15 2018 Kharkiv: NANGU (2018) 120– and computing. Vol. 45. (2007).
122. [13] М. М. Mezhennaya, Time-frequency
[2] О. M. Datsok, І. V. Prasol, О.А. Yeroshenko, analysis of the total electromyogram in the
Construction of a biotechnical system of qualitative and quantitative assessment of the
muscular electrical stimulation,Bulletin of functional state of the human neuromuscular
NTU "KhPI". Series: system. Biomedical radioelectronics. № 2.
Informaticsandmodeling. Kharkiv: NTU (2012) 3-11.
[14] S. G. Nikolaev, Workshop on clinical medicine (2015) 185–
electromyography, Ivanovo.(2001) 264 p. 197.doi:10.17691/stm2015.7.2.22.
[15] B. M. Gekht, Theoretical and clinical [26] H. Kawamoto, Y. Sankai, Power assist
electromyography. (1990) 229 p. methodbased on phase sequence and muscle
[16] A. V. Sidorenko, V. I. Khodulev, A. P. force conditionfor HAL, Adv Robotic
Selitskiy, Nonlinear analysis of (2005)717–734.doi:
electromyograms, Biomedical technologies 10.1163/1568553054455103.
and electronics. №11. (200) 53–59. [27] D.P. Ferris, G.S. Sawicki, M.A. Daley, A
[17] M. M.Mezhennaya, Choice of parameters of physiologist’sperspective on robotic
time-frequency processing of exoskeletons for human locomotion,Int J HR
electromyograms of the neuromuscular (2007) 507–528.doi:
apparatus, RT-2010: materials of the 6th Int. 10.1142/s0219843607001138
youth scientific-tech. conf. Sevastopol: [28] K.E. Gordon, M. Wu, J.H. Kahn, B.D.
SevNTU. (2010) 464 p. Schmit, Feedbackand feedforward
[18] I. Perova, Ye. Bodyanskiy, Adaptive Human locomotor adaptations to ankle-foot load
Machine Interaction Approach for Feature inpeople with incomplete spinal cord injury,
Selection-Extraction Task in Medical Data J Neurophysiol (2010)1325–1338.Doi:
Mining, International Journal of Computing, 10.1152/jn.00604.2009.
no. 17(2). (2018) 113-119. [29] C.K. Battye, A. Nightengale, J. Whillis The
[19] M. Akay, Time-frequency representationsof use of myoelectric current in the operation of
signals, Detection and estimation methods prostheses, J Bone JointSurg Br 37-B(3).
forbiomedical signals. San Diego: Academic (1995) 506–510.
Press. (1996) 111–152. [30] F.R. Finley, R.W. Wirta,Myocoder studies of
[20] M. Hosokawa, Time-Frequency Analysis of multiplemyopotential response, Arch Phys
Electronystagmogram Signals in Patients Med Rehabil 48(11).(1967) 598–601.
with Congenital Nystagmus, Japanese [31] B.Peerdeman, D.Boere, H.Witteveen,
Ophthalmological Society. Vol. 48. (2004) R.Huis in`tVeld, H.Hermens, i S.Stramigiol,
262–267 H.Rietman, P.Veltink,S. Misra,Myoelectric
[21] J. Kaipio, Simulation and Estimation forearm prostheses: state of the art froma
ofNonstationary EEG, Natural and user-centered perspective, J Rehabil Res Dev
Environmental Sciences. Vol. 40. (1996) 48(6).(2011)719.doi:
110. 10.1682/jrrd.2010.08.0161.
[22] Z.Y Lin, D.Z Chen, Time-frequency [32] M. Aminoff Electromyography in clinical
representation ofthe electrogastrogram – practice,Addison-Wesley (1978).
application of the exponential distributions, [33] J.M. Wakeling Spectral properties of the
IEEE TransBiomed Eng. Vol. 41. (1994) surface EMGcan characterize motor unit
267–275. recruitment strategies, J ApplPhysiol;
[23] Rohtash Dhiman et al. Detecting the useful 105(5).(2008) 1676–1677.
electromyogram signals–extracting, [34] C. Fleischer, A. Wege, K. Kondak,
conditioning andclassification, IJCSE. – G.Hommel, Application of EMG signals for
Aug.–Sep. 2011. V. 2. № 4. (2011) 634–637. controlling exoskeletonrobots, Biomed Tech
[24] A.S.Borgul, A.A.Margun, K.A.Zimenko, 51(5–6). (2006)314–319.doc:
A.SKremlev., A.Y. Krasnov Intuitive 10.1515/BMT.2006.063.
Control for Robotic Rehabilitation Devices [35] D. Farina, L. Mesin, S. Marina, R.A.
by Human-Machine Interface with EMG and Merletti, SurfaceEMG generation model
EEG Signals, 17th international with multilayer cylindrical descriptionof the
conferenceon Methods and Models in volume conductor, IEEE Trans Biomed Eng
Automation and Robotics (MMAR 2012). 51(3).(2004) 415–426.doi:
Proceedings. Międzyzdroje: IEEEXplore 10.1109/TBME.2003.820998.
digital library.(2012) 308–311. [36] H.J. Hermens, B. Freriks, C. Disselhorst-
[25] A.A. Vorobyev, A.V. Petrukhin, O.A. Klug, G. Rau,Development of
Zasypkina,P.S. Krivonozhkina, A.M. recommendations for SEMG sensors
Pozdnyakov, Exoskeleton as a newmeans in andsensor placement procedures, J
habilitation and rehabilitation of invalids Electromyogr Kinesiol 10(5).(2000)361–
(review),Sovremennye tehnologii v 374.doi: 10.1016/S1050-6411(00)00027-4.
[37] F. Sylos-Labini, V La Scaleia, A. d’Avella, I.
Pisotta,F. Tamburella, G. Scivoletto, M.
Molinari, S. Wang, L. Wang,E. van
Asseldonk, H. van der Kooij, T. Hoellinger,
G. Cheron,F. Thorsteinsson, M. Ilzkovitz,
J.Gancet, R.Hauffe, F.Zanov,F.Lacquaniti,
Y.P. Ivanenko,EMG patterns during
assistedwalking in the exoskeleton, Front
Hum Neurosci 8.(2014)423.doi:
10.3389/fnhum.2014.00423.
[38] R.Merletti, M.Aventaggiato, A.Botter,
A.Holobar,H.Marateb, T. Vieira,Advances
in surface EMG: recentprogress in detection
and processing techniques, Crit RevBiomed
Eng 38(4).(2011) 305–345.doi:
10.1615/CritRevBiomedEng.v38.i4.10.
[39] D. Farina, C. Cescon, Concentric-ring
electrode systemfor noninvasive detection of
single motor unit activity, IEEETrans
Biomed Eng 48(11).(2001) 1326–1334.doi:
10.1109/10.959328.
[40] J.L. Nielsen, S.Holmgaard, N.Jiang,
K.Englehart,D.Farina, P. Parker,Enhanced
EMG signal processing forsimultaneous and
proportional myoelectric control, Conf
ProcIEEE Eng Med Biol Soc (2009) 4335–
4338. doi: 10.1109/IEMBS.2009.5332745.
[41] D.P.Ferris, C.L. Lewis,Robotic lower limb
exoskeletonsusing proportional myoelectric
control, Conf Proc IEEEEng Med Biol Soc
(2009) 2119–2124.doi:
10.1109/IEMBS.2009.5333984.