# I. Introduction uman movements are the result of the coordinated work of functional mechanisms and processes. Such coordinated work develops during and as a result of constructing an action with the leading role of the higher parts of the central nervous system and ensures the consolidation of all body systems involved in its implementation [1][2].This complicated organization of the human motion system determines the demand of integrated analysis of movements based on the interpretation of both biomechanical and electrophysiological motion parameters. Biomechanical parameters describe the external spatio-temporal components, and electrophysiological parameters describe internal control mechanisms [3][4]. Existing methods and technical solutions for the study of human movements based on electrophysiological signals are used primarily in clinical practice to diagnose the functional state and diseases of the human musculoskeletal system [5][6]. Moreover, the list of movements in clinical practice is limited to simple and working movements (walking, grabbing, cyclic motion acts) [7][8][9][10]. In the field of sports, the use of biomechanical methods for analyzing human movements predominates. Such methods allow to estimate the technique of implementation sports exercises, but do not provide information about the physiological mechanisms of the motion act [11][12][13].Electrophysiological methods in sports are still not enough used, because they are mainly intended for laboratory use of rather complex and expensive neurophysiological systems, require highly qualified personnel and special methods for analyzing the obtained data [14][15][16][17].Diagnostic systems for the study of sports movements must have certain design and technical features: multichannel, compact, autonomous power, wireless data transfer [18][19][20]. In addition, the sport movements are characterized by a complicated structural and functional organization [21], which requires the development of new algorithms for the complex processing of biomechanical and electrophysiological data. Each human movement corresponds to a certain spatio-temporal pattern, which characterize the direction, power and sequence of inclusion of muscles involved in performing the movement. Automated movements (walking, running, cycling, special and sports movements) are motion skills. They have a constant pattern of muscle involvement not only when the movement is repeated by one person, but also in different people. Selection from the aggregate of the muscles electrical activity data such combinations, which characterize a specific movement, make it possible to create the electromyographic pattern of human movement [22][23][24]. In this work, we propose the complex analysis of human movements based on the identification of amplitude-time characteristics of electromyographic patterns, and describe studies that confirm the effectiveness of the proposed approach. # II. Research Methods and Data Processing a) Subjects The study was carried out among participations of the basketball team at the Sports Complex of the Belarusian State University of Informatics and Radio Electronics (Minsk, Belarus). The study involved 42 people (7 women, 35 men) aged 17 to 25 years, of which 8 people are the professional sportsmen. The essence of the research were previously explained for all subjects. The subjects were in full health and did not report any feelings of pain when performing the tests. The movement "jump up" was selected as a test for the study of person motion skills. This movement is a speed-strength exercise, andit' sused as a test movement for estimation of motion abilities in highspeed power sports, or as a training exercise to develop a muscle strength of legs, jumping ability and the ability to concentrate muscle efforts combining strength with speed [25][26]. The main muscles of the legs (m. rectus femoris, m. biceps femoris, m. gastrocnemius lateralis, m. soleus) were studied [27] (table 1) due to the specificity of the test movement. # Table1: The studied muscles and their functions # Musclename Locationandfunction It is located on the front of the thigh. Unbends the leg in the knee joint, flexes the hip in the hip joint. It is located on the back of the thigh. Flexes the leg in the knee joint, unbends the leg in the hip joint. It is located on the back of the shin. Flexes the foot, helps in the flexing of the leg in the knee joint. It is located deep in the shin. Flexes the foot. The method of multichannel interference electromyography was chosen for creation of the motion innervation structure. The method of podography was chosen for creation of the motion spatio-temporal structure. The methods, which are used to create the motion innervation structure, should be associated with the registration of muscles control signals in the process of motion activity. The most convenient, simple and painless of these methods is multichannel interference electromyography, which allows studying the bioelectric activity of several muscles at the same time. The summary electromyogram (EMG) is formed as a result of interference of many motion unit signals located in the area of recording. The method of multichannel interference electromyography allows to estimate the participation of muscles in various movements, the sequence of their on and off, the integrated level of muscle activation, the correlation of muscle activity in different periods of motion skills formation [28][29][30]. Thus, the motion innervation structure can be represented by multichannel electromyographic signals of the studied muscles. Thus, the motion spatio-temporal structure can be represented by multichannel podographic signals of the studied movement. The complex application of these methods (multichannel interference electromyography and podography) allows to create an electromyographic pattern of motion, which is an electrophysiological analogue of motion and is a combination of time, spatial and amplitude electromyographic characteristics of motion. # c) Data collection Registration of electrophysiological and biomechanical signals were realized by the proprietary system "Mio Sport" [33]. The developed device includes a multi-channel electromyography system with synchronous registration multi-channel podographic signals, and a software for visualization and analysis of the results (Figure 1). The system allows real-time The spatio-temporal structure of movement allows to divide the motion activity into separate phases and then to research it in parts. In the presented paper, the method of podography was chosen for creation of the motion spatio-temporal structure. The method of podography (PDG) allows to record the time moments of contact of various parts of the foot with the ground, and to identifymotion phases [31][32]. recording of 4EMG channels (minimum detection limit of 30 ? V, frequency range of 0.1-800 Hz), 4 PDG channels The registration of summary bioelectric activity of the muscles (electromyograms) was carried out using specialized surface medical adhesive electrodes. Fixation and positioning of electrodes on the studied muscles, orientation relative to motor fibers, the quality of skin surface preparation for recording of surface EMG were carried out in accordance with the recommendations of SENIAM [36]. The multi-channel podographic signals were recorded synchronously with EMG using the developed sensor electrodes [33][34] located on the soles of both feet. The dimensions of the podographic electrodes are small and do not disturb with the acting of exercises. The sensory manufacturing technology of PDG electrodes allows to study motions on the floor with any coating. The PDG electrodes are fixed to the heel and toe of each foot. Thus, it is possible to record the moments of separation and touching the ground by the heel and toe of each leg. Figure 2 shows an example of the signals registration: the muscle electromyogram of m. rectus femoris of the right leg and two channels of the right foot podograms for the test move "jump up". The podograms of each leg are presented as two curves: with a smaller amplitude -the reaction of the heel, with a larger amplitude -the reaction of the toe. In this case, the zero level of the curves matches to the touching of the foot part with the ground, and the maximum amplitude of the curves matches to the lack of touching (Figure 2). In the presented work, the sampling frequency of the electromyographic signals is 1600 Hz for, and the sampling frequency of the podographic signals is 320 Hz. Accordingly, the minimum sampling recovery time interval for the correct displaying of electrophysiological and biomechanical signals is selected as 0.625 ms for the EMG signal, and 3.125 ms-for the PDG signal. 2. The filtration of multichannel electrophysiological signals. 3. Time rationing of multichannel electromyographic signals. Time rationing of signals of multichannel electromyograms is necessary for unification of the EMG pattern of the same movement in different people. The creation of a time motion structure using the method of podography is possible for such locomotion as walking, running, jumping. Thus, any such exercise will contain phases of movement, when one or both legs do not touch the ground. Those phases are chosen as the norm. Further, the signals of multichannel EMG are normalized according to the selected norm. In the work, "jump up" was selected as a test motion. For this exercise, the phase of flight (Î?"t_flight) is taken as the norm. The time interval corresponding to the two phases of flight (2Î?"t_flight) is captured to the left of the moment the toe separate from the ground and to the right of the moment the toe touch with the ground (Figure 3). Thus, after processing of the multichannel electrophysiological and biomechanical signals in accordance with the described algorithm, the obtained data are the EMG pattern of the studied movement [38]. # 18 Global Journal of Researches in Engineering ( ) Volume XIX Issue V Version I F As a result of the electrode system movement during the motion test, low-frequency components (motion artifacts) appear in the spectrum of the EMG signal, which leads to incorrect analysis of the EMG signal [37]. In the presented work, EMG signals are filtered using a fourth-order Chebyshev digital high-pass filter for removing of the motion artifacts from the useful signal spectrum. The cutoff frequency of the filter is 10 Hz. ground'', "heel touching with the ground'' and ''end of the movement''. The following phases were detected for the test movement: the squatting phase (1) -from the motion beginning to the heel separation from the ground, the push phase (2) -from the heel separation from the ground to toe separation from the ground, the flight phase (3) -from toe separation from the ground to the toe touching with the ground, the touchdown phase (4) -from the toe touching with the ground to the heel touching with the ground, the rise phase (5) -from the heel touching with the ground to the movement end. 1 -the squatting phase, 2 -the push phase, 3 -the flight phase, 4 -the touchdown phase, 5 -the rise phase Figure 4: The EMG pattern of the movement "jump up" The processing of registered electromyographic signals of muscles is carried out in the digital form. The initial analog electrical EMG signal of muscles is sampled in time and quantized in amplitude. In the theory of digital signal processing, the energy of a digital signal is considered not as a physical quantity but as a means of comparing different signals. So, for the initial electrical signal, the load resistance is assumed R=1 [39]. Accordingly, the dimension of the received energy of the digital signal is not measured in Joules. However, the energy of the digital signal can be recalculated to a physical value if the load resistance is known [40]. The EMG signal energy m E is calculated for each muscle m: ? ? = = = = n i i m n i i m m n x E E 1 2 , 1 , ) ( ,(2) where i m E , -i-th discrete sample of the EMG signal energy of the muscle m; Similarly, the energy of the EMG signal of muscle m is calculated for each phase f of the studied motion f m E , . The using of the average, not summary energy of the EMG signal of muscles allows to take account of the frequency of sampling of the initial electromyographic signal, and, accordingly, to standardize the proposed method of integrated analysis of human motions in case of using technical systems of multichannel electromyography with different parameters of the sampling frequency of EMG signals. In each phase of motion f for all muscles m, the fraction of the EMG signal energy % , f m E is calculated as a percentage of the total signal energy: % 100 * , % , m f m f m E E E = , (%)(3) Where f-the number of the motion phase; f m E , -the energy of the EMG signal of the muscle m in the phase f. III. # Results The estimation of energy contribution of the muscles in the process of motion. In order to identify types of motion skills, the estimation of the energy contribution of muscles during the motion "jump up" was carried out according to the following algorithm: 1. For each subject, the EMG pattern of the test motion was created in accordance with the above method. 2. The summary energy of the EMG motion pattern ? E was calculated as the sum of the EMG signal energies of all studied muscles: ? = ? = M m m E E 1 ,(4) 3. For each muscle, the percentage of energy from the summary EMG energy of the motion pattern % m E was determined: % 100 * % ? = E E E m m ,(5) 4. For each muscle, the mean value and standard deviation (RMS) of the energy fraction from the summary EMG energy of the motion pattern were calculated. The percentage of the energy from the summary EMG energy of the motion pattern for each muscle is presented in Table 2. It was found out that among the studied muscles the following muscles mainly provide for the test movement: m. rectus femoris muscles (the share of EMG energy of the motion pattern is 38.82%±14.75%), m. gastrocnemius lateralis and m. soleus (the share of EMG energy of the motion pattern is 28.62%±12.33% and 28.23%±13.43%). The level of involvement of the muscle m. biceps femoris in the motion is not significant (the share of EMG energy of the motion pattern is 4.33%±2.86%). In addition, the largest variation of the EMG energy fraction of the motion pattern is observed for the muscles m. rectus femoris, m. gastrocnemius lateralis and m. soleus (m. rectus femoris -±14.75%, m. gastrocnemius lateralis -±12.33%, m. soleus -±13.43%). Thus, the muscles m. rectus femoris, m. gastrocnemius lateralis and m. soleus are the leaders in the formation of the test motion and were chosen as the basis for the detection of jump types. The muscles m. gastrocnemius lateralis and m. soleus belong to the triceps of the shin and perform the same functions (shin and foot bending/unbending, shin rotation). Accordingly, they can be combined into one group for consideration of energy contribution of the muscles in the process of motion. For this muscle group (m. gastrocnemius lateralis + m. soleus), the share of EMG energy of the motion pattern is 56.85%±14.22% (see Table 2). The point diagram of the percentage proportion of the EMG energy share of the test motion pattern for the muscle m. rectus femoris and the muscle group m. gastrocnemius lateralis + m. soleus was drawn (Fig. 5). Figure 6 shows the histograms of the percentage ratio of the EMG energy shares of the motion pattern for the studied muscles, corresponding to the above groups. The estimation of distribution of muscle efforts in the phases of motion. The analysis of the distribution of muscle efforts in the phases of the test motion was carried out in order to identify the types of motion skills. 1. The analysis of the distribution of muscle efforts in the motion phases was carried out according to the following algorithm: 2. For each subject, the EMG pattern of the test motion was created in accordance with the above method. 3. The summary energy of the EMG motion pattern ? E was calculated as the sum of the EMG signal energies of all studied muscles (see formula 4). # The summary energy of the EMG signals of each phase f E was calculated as the sum of the EMG signal energies of the muscles, concentrated in the phase f: ? = = M m f m f E E 1 , ,(6) 5. For each phase, the percentage of energy from the summary EMG energy of the motion pattern % f E was determined: % 100 * % ? = E E E f f ,(7) 6. For each phase, theme an value and standard deviation (RMS) of the energy fraction from the summary EMG energy of the motion pattern were calculated. The percentage of the energy from the summary EMG energy of the motion pattern for each phase is presented in Table 3. It was found that the largest variation of the EMG energy share of the test motion pattern is observed in the squatting phase and the push phase. The deviation of the EMG energy share of the motion pattern for the squatting phase is ±18.15%, for the push phase -±19.80%. For the other phases (flight, 1 22 Global Journal of Researches in Engineering ( ) Volume XIX Issue V Version I F touchdown and rise), this parameter does not exceed 6% of the summary EMG energy of the motion pattern. In addition, the sum of the EMG energy of the test motion pattern for the squatting and push phases is 71.14%±7.95%. Thus, these phases are determinative in the formation of test motion. Based on the above, the squatting and push phases were chosen as the basis for the identification of jump motion types. The point diagram of the percentage proportion of the EMG energy share of the test motion pattern in the squatting phase and the push phase was drawn (Fig. 7). for which the maximum muscle effort is performed in the push phase: the share of EMG energy of the motion pattern in the squatting phase is 6.12%± 5. 45%, in the push phase -65. 07%± 9. 82%. 2. The group (6 people -14. 29% of the whole sampling), for which the maximum muscle effort is performed in the squatting phase: the share of EMG energy of the motion pattern in the squatting phase is 51. 47%± 6. 44%, in the push phase -17. 90%± 4. 59%. 3. The group (13 people -30, 95% of the whole group), for which the muscle efforts are equally distributed between squatting and push phases: the share of EMG energy of the motion pattern in the squatting phase is 29, 40%±7, 78%, in the push phase -38, 74%±8, 29%. 4. Figure 8 shows the histograms of the percentage ratio of the EMG energy shares of the motion pattern by the phases, corresponding to the above groups. IV. # Global # Discussing The results of the research prove the effectiveness of the proposed complex analysis of human motion, which allows to classify the types of motion skills of the subjects on the following bases: ? by the energy contribution of the muscles in the process of motion; ? by the distribution of muscle efforts in the phases of motion. The research of the energy contribution of the studied muscles during the test motion "jump up" was carried out. As a result of the analysis of the percentage ratio of EMG signal energy of the muscles during the test movement, three types of motion skills were identified: a) 40 The research of the distribution of muscle efforts in the phases of the test motion "jump up" was carried out. The percentage ratio of EMG energy of the motion pattern in the squatting and push phases is different and allows to identify three types of motion skills: a) 54, 76% of all subjects: the maximum muscle effort is performed in the push phase (the share of EMG energy of the motion pattern in the push phase -65. 07%± 9. 82%); b) 14, 29% of all subjects: the maximum muscle effort is performed in the squatting phase (the share of EMG energy of the motion pattern in the squatting phase is 51. 47%± 6.44%,); c) 30, 95% of all subjects: the muscle efforts are equally distributed between squatting and push phases (the share of EMG energy of the motion pattern in the squatting phase is 29, 40%±7, 78%, in the push phase -38, 74%±8, 29%). V. # Conclusion In the paper, the method of multichannel interference electromyography was chosen as a method of investigation of electrophysiological parameters of motion. The method of multichannel interference electromyography allows to estimate the participation of muscles in various movements, the sequence of their on and off, the integrated level of muscle activation, the correlation of muscle activity in different periods of motion skills formation. The method of podography was chosen to study the biomechanical motion parameters. The method of podography (PDG) allows to record the time moments of contact of various parts of the foot with the ground, and to identify motion phases. The complex application of these methods (multichannel interference electromyography and podography) allows to create an electromyographic pattern of motion, which is an electrophysiological analogue of motion and is a combination of time, spatial and amplitude electromyographic characteristics of motion. In the paper, the algorithm of creation of an electromyographic motion pattern based on analysis of multichannel electrophysiological and biomechanical signals was proposed. The algorithm includes: 1) the overlay of electrophysiological and biomechanical signals in the time domain; 2) the filtration of multichannel electromyographic signals to remove motion artifacts and network interference from the spectrum of the useful signal; 3) the time rationing of multichannel electromyogrphic signals for unification of the EMG pattern of the same movement in different people; 4) the detection of motion phases. In the paper, the method of complex estimation of human motion skills based on the analysis of electromyographic motion pattern was proposed. The analysis of multichannel electromyogrphic signals of muscles is made in the time domain, due to the need for comparison of the innervation motion structure with the spatial and time characteristics of the studied motion. The proposed method allows to calculate the energy contribution of the muscles in the process of motion and to calculate the distribution of muscle efforts in the phases of motion. In the paper, the research of the motion skills was carried out on the example of the test movement "jump up". The research of the energy contribution of the muscles during the test motion allowed to identify three types of motion skills: 1) the muscles m. gastrocnemius lateralis and m. soleus mainly provide test motion (40.48% of all subjects); the muscles m. rectus femoris, m. gastrocnemius lateralis and m. soleus are equal in the degree of involvement in the process of motion (50% of all subjects); the muscle m. rectus femoris dominates in the process of the motion (9.52% of all subjects).The research of the distribution of muscle efforts in the phases of the test motion allowed to identify three types of motion skills: the maximum muscle effort is performed in the push phase (54,76% of all subjects);the maximum muscle effort is performed in the squatting phase (14,29% of all subjects); the muscle efforts are equally distributed between squatting and push phases (30,95% of all subjects).The results of the research demonstrate the effectiveness of the proposed complex analysis of human motion. The proposed approach to the analysis of human movements can be used to study human motion abilities in clinical and sports medicine (motion rehabilitation, clinical and sports biomechanics, prosthetics), training process (control of the sportsman's motion stereotype, prognosis of the children's motion talent) and special activity (effective training of special skills, professional selection). ![Global Journal of Researches in Engineering ( ) Volume XIX Issue V Version I F (time resolution of 3,125 ms), and data transmission via a wireless interface to a distance of up to 100 m [34-35].](image-2.png "") 1![Figure 1: The software of the system "Mio Sport"](image-3.png "Figure 1 :") 2![Figure 2: View of synchronously recorded EMG and PDG signals during the test movement](image-4.png "Figure 2 :") 3![Figure 3: The time structure of the motion "jump up"](image-5.png "Figure 3 :") 1![The detection of motion phases.](image-6.png "1 .") 4![Figure 4 shows an example of the EMG pattern of the motion "jump up". The figure shows the normalized electromyograms of muscles m. rectus femoris, m. biceps femoris, m. gastrocnemius lateralis and m. soleus of the right leg, and markers of the movement phases borders corresponding to the time moments "start of the movement '', "heel separation from the ground'', ''toe separation from the ground'', "toe touching with the ground'', "heel touching with the](image-7.png "Figure 4") 12051![Figure 5: The point diagram of the percentage proportion of the EMG energy share of the test motion pattern for the muscle m. rectus femoris and the muscle group m. gastrocnemius lateralis + m. soleus It was found that the data are clustered into 3 groups: 1. The group (17 people -40.48% of the whole group), for which the muscles that mainly provide test motion are m. gastrocnemius lateralis and m. soleus. The share of EMG energy of the motion pattern for the muscular group m. gastrocnemius lateralis + m. soleus is 71.17%± 5.63%, for the muscle m. rectus femoris -25.09%± 5.30%. 2. The group (21 persons -50% of the whole group), for which the muscles m. rectus femoris, m. gastrocnemius lateralis and m. soleus are equal in the degree of involvement in the process of motion. The share of EMG energy of the motion pattern for the muscular group m. gastrocnemius lateralis + m. soleus is 52.78%± 4.69%, for the muscle m. femoris -43.81%± 5.58%. 3. The group (4 persons -9.52% of the whole group), for which the muscle m. rectus femoris dominates in the process of the movement. The energy share of the EMG motion patternfor muscle groupm. gastrocnemius lateralis + m. soleus is 25.79%±1.32%, for the muscle m. rectus femoris -72.07%±2.94%.Figure6shows the histograms of the percentage ratio of the EMG energy shares of the motion pattern for the studied muscles, corresponding to the above groups.](image-8.png "1 20 GlobalFFigure 5 : 1 .") 6![The histograms of the percentage ratio of the EMG energy shares of the motion pattern for the studied muscles: the muscles m. gastrocnemius lateralis and m. soleus mainly provide test motion (?); the muscles m. rectus femoris, m. gastrocnemius lateralis and m. soleus are equal in the degree of involvement in the process of motion (b); the muscle m. rectus femoris dominates in the process of the motion (c) Thus, the percentage ratio of the EMG energy shares of the motion pattern for the studied muscles is different and allows to identify three types of motion skills for the jump: a) the muscles m. gastrocnemius lateralis and m. soleus mainly provide test motion (Fig. 6 a); b) the muscles m. rectus femoris, m. gastrocnemius lateralis and m. soleus are equal in the degree of involvement in the process of motion (Fig. 6 b); c) the muscle m. rectus femoris dominates in the process of the motion (Fig. 6 c). The EMG energy share of the motion pattern for the muscle m. biceps femoris does not exceed 5%.](image-9.png "Figure 6 :") 7![Figure 7: The point diagram of the percentage proportion of the EMG energy share of the test motion pattern inthe squatting phase and the push phaseIt was found that the data are grouped into 3 groups (Figure7):1. The group (23 people -54. 76% of the whole group), for which the maximum muscle effort is performed in the push phase: the share of EMG energy of the motion pattern in the squatting phase is 6.12%± 5. 45%, in the push phase -65. 07%± 9. 82%. 2. The group (6 people -14. 29% of the whole sampling), for which the maximum muscle effort is performed in the squatting phase: the share of EMG energy of the motion pattern in the squatting phase is 51. 47%± 6. 44%, in the push phase -17. 90%± 4. 59%. 3. The group (13 people -30, 95% of the whole group), for which the muscle efforts are equally distributed between squatting and push phases: the share of EMG energy of the motion pattern in the squatting phase is 29, 40%±7, 78%, in the push phase -38, 74%±8, 29%. 4. Figure8shows the histograms of the percentage ratio of the EMG energy shares of the motion pattern by the phases, corresponding to the above groups.](image-10.png "Figure 7 :") 8![The histograms of the percentage ratio of the EMG energy shares of the motion pattern by the phases: the maximum muscle effort is performed in the push phase (?); the maximum muscle effort is performed in the squatting phase (b); the muscle efforts are equally distributed between squatting and push phases (c) Thus, the percentage ratio of the EMG energy shares of the motion pattern for the squatting and push phase is different and allows to identify three types of motion skills for the jump: a) the maximum muscle effort is in the push phase (Figure8a); b) the maximum muscle effort is in the squatting phase (Figure 8 b); c) the muscle efforts are equally distributed between squatting and push phases (Figure 8 c). The distribution of muscle efforts in the flight, touch-down, and rise phases does not change (the deviation of the EMG energy share of the motion pattern for these phases is less than 6%).](image-11.png "Figure 8 :") 2m.rectusfemorism. bicepsfemorism. gastrocnemiuslater alism. soleusm. gastrocnemius + m. soleusMeanvalue of% m E , %38,824,3328,6228,2356,85RMS of % m E , %14,752,8612,3313,4314,22 3The squatting phaseThe pushphaseThe flight phaseThe touchdown phaseThe risephaseMeanvalue of% f E , %20,2050,9415,914,148,81RMS of% f E , %18,1519,806,003,054,70 © 2019 Global Journals ## Acknowledgements The authors express their appreciation to the patients and subjects for participating this study. * JHWilmore DLCostill WLKenney Physiology of sport and exercise 524 1994 Human kinetics * ?????????? ???????? ? ?????????? ??????????? ???????????? 1990 ????? * Biomedical engineering and design handbook MKutz 2009 McGraw-Hill Professional * Biomedical Technology and Devices Handbook JMoore 2009 CRC Press * Electrodiagnosis in Clinical Neurology MJAminoff 2012 Elsevier Health Sciences * Electrodiagnosis in Diseases of Nerve and Muscle: Principles and Practice JKimura 2001 Oxford University Press * Electro diagnostic tests in polyneuropathy and radiculo pathy BCCallaghan JFBurke ELFeldman Journal of the American Medical Association 315 3 2016 * EMG signals analysis of BF and RF muscles in autism spectrum disorder (ASD) during walking MMNor RJailani NMTahir IMYassin ZIRizman RHidayat International Journal on Advanced Science, Engineering and Information Technology 6 5 2016 * Clinical decision making and differential diagnosis in a cyclist with upper quarter pain, numbness, and weakness: a case report MSBriggs KKRethman MTLopez International journal of sports physical therapy 13 2 255 2018 * A multichannel surface EMG system for hand motion recognition YFang HLiu GLi XZhu International Journal of Humanoid Robotics 12 02 1550011 2015 * ?????????? ??????????: ??????? ??????? ??? ????????? ????? ??????????? 2005 ???????? * Biomechanics: principles, trends and applications JHLevy 2010 Nova Science Publishers Incorporated * CJPayton Burden, A. * Biomechanical evaluation of movement in sport and exercise: the British Association of Sport and Exercise Sciences guide Routledge * Surface electromyography in sports and exercise HTurker HSözen Electro diagnosis in new frontiers of clinical research 2013 Intech Open * Peripheral nerve injuries attributable to sport and recreation. Physical medicine and rehabilitation clinics of North America CToth 2009 20 * Expert opinion and controversies in musculoskeletal and sports medicine: stingers. Archives of physical medicine and rehabilitation CJStandaert SAHerring 2009 90 * ?????? ?????????????? ???????? ?????????? ??????????????????. ?.: ????????? 2007 * Mobile recording system for sport applications PKugler DSchuldhaus UJensen BEskofier Proceedings of the 8th international symposium on computer science in sport (IACSS 2011) the 8th international symposium on computer science in sport (IACSS 2011)Liverpool 2011 * Development of a wireless system for the remote monitoring of muscular activity GLCerone Gait Posture 57 3 2017 * The use of surface electromyography in biomechanics CJDe Luca Journal of applied biomechanics 13 2 1997 * ?????????? ???????? ????????? ???????? ?????. ??????????. ??????????. Litres 2018 * EMG and muscle force: an introduction ALHof Human Movement Science 3 1-2 1984 * RMerletti PAParker PJParker Electromyography: physiology, engineering, and non-invasive applications John Wiley & Sons 2004 11 * Robust EMG pattern recognition in the presence of confounding factors: features, classifiers and adaptive learning YGu DYang QHuang WYang HLiu Expert Systems with Applications 96 2018 * ?????? ?????? ? ????? ????? ? ???????? ????????????? ?? ?????????? ? ??????? ? ???? ? ????????? ????????? ??????? ?????????? ????? 2 7 2005 * ?????????? ??? ???????????-??????????????? ???????? ?Ð?"????????? ?????????? ?????????? ????????: ??????????, ???????????, ?????????? 2000 3 * DLMount Atlas of Anatomy. Plasticand Reconstructive Surgery 2017 139 1490 * Atlas of electromyography AALeis VCTrapani 2000 Oxford University Press * Introduction to surface electromyography ECriswell 2010 Jones & Bartlett Publishers * Electromyography in Clinical Practice E-Book: A Case Study Approach BKatirji 2007 Elsevier Health Sciences * A Computer System for Biomechanical Studies of the Foot by Podography SMJatsun ASJatsun ANRukavitsyn Biomedical Engineering 52 3 2018 * ?????????????-??????????? ? ????????????? ??????????: ????????? ? ??????????? ?????????? ?????????? 2004 * ?????????-??????????? ???????? ?????????????? ???????????????? ??? ??????????? ???????????? ??????? ???????? (Doctorald is sertation) ??Ð?"??????? Ð?"???????????????????????????????????????????? ????????????? ? ???????????????? BSUIR. 34. 34. Ð?"???????, ?. ?. 4 66 2012. 2012 ??????????? ??????? ??????????? ??????????-?????????????? ??????????? ???????? ???????? * ??Ð?"??????? ???????? ??Ð?"?????? ?????????? ??????????? ??????? ?????????????? ?????? ?????????? ???????????????? ????????. In?????????????? -2014. ???????? ??????????? ??????????? ? ????? ??????????? ?????????? 2014 * European recommendations for surface electromyography HJHermens BFreriks RMerletti DStegeman JBlok GRau GHägg Roessingh research and development 8 2 1999 * Filtering the surface EMG signal:Movement artifact and baseline noise contamination CJDe Luca LDGilmore MKuznetsov SHRoy Journal of biomechanics 43 8 2010 * ????????-??????????? ??????????????? ????????? ??????-????????? ???????? ?????????? ?? ?????? ?????????? ? ??????? ?????????????????????? ???????? ?????? ??Ð?"??????? ?????????? ??Ð?"?????? ???????? ????????????????? 2 30 * MKMandal AAsif Continuous and discrete time signals and systems Cambridge Cambridge University Press 2007 1 * SKMitra YKuo Digital signal processing: a computer-based approach New York Mc Graw-Hill 2006 2