Review Article - Biomedical Research (2017) Complex World of Neuroscience

Review of surface electrode placement for recording electromyography signals

Department of Electrical and Electronic Engineering, Faculty of Engineering, UPM Serdang, 43400, Selangor, Malaysia

*Corresponding Author:

Hossein Ghapanchizadeh
Department of Electrical and Electronic Engineering Faculty of Engineering, UPM Serdang, Malaysia

Accepted date: July 13, 2016

Abstract

Background: Surface Electromyography (SEMG) signal has used in monitoring muscle activities. It has been widely applied in many areas, such as body member prosthesis, noise cancellation for braincomputer interface, and robotics. The SEMG acquisition method for collecting the signal with low-noise has extensively investigated in the last decade. The objective of this study is to review the recent works on electrode position and identify avenues for future research. Methods: A review of the relevant literature published between 1986 and 2015. This study commences with the basics of SEMG and recent methods for electrode position. Result: The different noises affecting SEMG signal include the spread of the innervation zone, cross-talk from neighbour muscles, electrode size, and location of electrode placement. Moreover, electrode placement or displacement effect SEMG signal in both time and frequency domain. Conclusion: Although several SEMG studies examined the effects of electrode position and internal electrode distance on forearm muscles, only a few studies addressed the methodological difficulties of the electrode position. In the majority of studies, electrodes were placed without the specific symptoms of the points along the length or shape of the muscle. Moreover, IED varied in different studies.

Keywords

Electrode placement, Inter electrode distance, SEMG electrode, Surface electromyography.

Introduction

Many techniques have been developed to monitor muscle behavior and movements; these methods include electromyography (EMG) [1,2], mechanomyography [3,4], and electroencephalography [5,6]. Surface Electromyography (SEMG) has recently applied in other areas, such as in hand prosthesis [7,8]. SEMG signals are also involved in monitoring the muscle activities of artificial body members [9], removal of noise from the brain-computer interface [10,11], and robotics exoskeleton [12]. Furthermore, SEMG can detect motor unit (MU) functions when these cells are active [13].

EMG signals can be recorded using invasive and non-invasive methods, with the latter referred to as SEMG [14,15]. Invasive techniques utilize needle electrodes to monitor EMG signals directly from the muscles. By contrast, SEMG collects data using surface electrodes placed on the skin [16,17]. SEMG technique has considerable advantages over the invasive method, including the easier detection of SEMG signal over the skin and more comfort for subjects [18,19]. Thus, recognizing muscle behavior through SEMG detection is more feasible. A raw SEMG signal has peak-to-peak amplitudes of 0-2 mV with a band frequency ranging from 0 Hz to 1000 Hz [19]. However, the band frequency of SEMG, which includes significant information, ranges from 20 Hz to 500 Hz [20]. Raw SEMG signals have low amplitude; thus, the SEMG signal can be affected by many types of noises such as ambition noise or combines with them [21].

The impedance of the body skin reduces the amplitude of SEMG signals and induces noise. The noise generated from the skin is caused by fat between the muscle and skin and the blood flow in minuscule vessels under the skin [22]. Other powerful noises include environmental noise that mainly comes from the environment, as well as electromagnetic radiation sources, electrical power wires, and fluorescent lamps during recording [23]. Various significant noises that affect SEMG are from the subcutaneous tissue layers [24], the spread of the innervation zone (IZ) [25], crosstalk from neighboring muscles [26], electrode size, and electrode position [27,28]. Electrode position can significantly mislead the description of a statistical and spectral factor of SEMG, thus affecting SEMG evaluation [29,30].

Muscle physiology

The muscles of the human body are divided into three categories, namely, skeletal, cardiac, and smooth muscles; each muscle group has its own characteristics [31]. Smooth muscles are non-striated and self-acting because humans cannot control their movements. Smooth muscles are tissues that basically form supporting blood vessels and the walls of hollow organs, such as the stomach [32]. Cardiac muscle or heart muscle is one of the major muscle groups, which is independent of the neural network [33]. The heart muscle is contracted automatically in the walls and histological foundation [34]. The most important muscle category consists of skeletal muscles that are found in the majority of muscle tissues [35]. These muscle groups are attached by the tendon to the bone [36]. In contrast to cardiac and smooth muscles, skeletal muscles are known as the voluntary muscle group. Humans can control skeletal muscles to make movements for a daily living because these muscles are under the control of the nervous system [37].

Skeletal muscles generate force and movement. Their structural unit is the muscle fiber [38]. The construction group of muscle fibers is the muscle myofiber or muscle cell. A myofiber has an approximately cylindrical shape with a diameter a few microns meters (10-100 μm) and a few millimeters to a few centimeters in length (1.5 mm to 30 cm) [39]. Many myofiber bundles or fascicules are spliced by tissue in muscle. The fascicule arrangement in a muscle is associated with muscle power and its motion limitation. The muscle fibers contract to move or produce a force.

Muscle origin is the attachment of the muscle to the bone, which is fixed and does not move during contraction. Muscle origin typically has more mass, and it is more stable when the muscle is contracted compared with the other muscle end. The skeletal muscles are connected to the bone by the tendon in an area called insertion zone. Movements occur at the joint muscles; thus, the insertion zone is usually distant from the distal portion of the muscle about the muscle origin to facilitate the movement [40]. Tendon is a strong band tissue connection, which usually links one end of the skeletal muscle to a joint or bone and is capable of tolerating the tension. Furthermore, the tendon has been considered to transmit forces. This link allows tendons to modulate forces reactively during movements and provide stability when muscles are in the rest mode [41].

Tendons are nonessential in performing the same practical role, with some tendons typically positioned in the limbs. Moreover, tendons can save and regain energy during the high performance. For example, when a human walks, the Achilles tendon stretches and the joint ankle flexes. After this action, the stored elastic energy will be released when the foot plantar flexes. The stretching of the tendon allows the muscle to increase its force when the muscle acts with less or without any change in length [42]. The nervous system controls each muscle function in the muscle IZ. Involuntary muscles, such as heart muscles, have IZ by the autonomic nervous system and skeletal muscles by the external. The point at the end of the nervous system (innervation point of voluntary muscles) is called the motor unit (MU) [43]. MU includes a fiber of the motor nerve, and all of the fibers of the muscle have their innervate point.

Motor unit action potential

When a human decides to move his body parts, such as leg, hand, and neck, the motor cortex, which is a part of the brain, produces a signal [44]. The generated signal of the motor cortex leads to the particular skeletal muscle through the neural network. The neural network is a functional entity of interconnected motor neurons. The motor neurons at the end of the neural network are connected to the muscle. Motor neurons or MUs are attached to the myofibrils. The power of muscle contraction must determine the number of activated MUs [43]. For example, to perform the same function, a few numbers of MUs are required to lift a piece of paper from a table. However, large amounts of MUs would be required to lift a hardback book. The size of the MU population depends on muscle size [45]. Small muscles have small MUs. Some minor muscles have a number ratio of muscle per number of MUs of 1:1. Larger muscles are required for harmonious and fine motor operations that require less control. For example, the large gluteus maximus has a ratio of 1:20, which indicates that each MU is responsible for activating 20 muscle fibers. During the activity of a large muscle, which is appropriated for power and force, a motor fiber is requested to fire a muscle fiber group together. However, fine tuning is necessary for controlling the motions of a smaller group [45].

All of the muscle fibers from the same MU contract or relax approximately at the same time [46]. Furthermore, if the MUs of muscle fibers are activated to contract the muscle, the fibers will be in the maximum contraction [47]. This condition is called the all-or-none law. MU is basically a functional unit in the myofibrils that induce muscle contraction.

Surface electromyography

The contraction of the muscle is fundamentally driven by the electrochemical process. The degree of muscle contraction is controlled by the frequency of nervous impulses. The pulse travels through the spinal cord to activate the muscle fiber. Sodium and potassium channels open in response to a stimulus, thereby triggering the active response of excitable membranes in the nerve and muscle fibers. The polarizing and depolarizing actions in opening sodium and potassium channels produce the myoelectric signal [48]. This process is called motor unit action potential (MUAP). MUAP is the final and fastest event in the myofibril, which rapidly increases and decreases the electrical potential through a cell membrane. MUAP activation can be monitored under electrodes. The surface electromyography (SEMG) is a technique for detecting, collecting, and monitoring the myoelectric signal over the muscles using electrodes [49].

Methodology and Materials

This paper investigates significant diversified studies performed through 1995 to 2016, and it emphasizes mostly on the newly published articles including various effect of electrode placement/ displacement on signal processing, methods of finding exactly electrodes location as well as a place of electrode over different muscles.

Electrode placement

Various studies demonstrate that electrode placement over a muscle exhibits significant efficacy on the specification of the recorded SEMG signals [29,30,50-60]. The changing distances between the electrodes showed different complicated shapes of action potential from both the intra and extracellular action potentials of isolated frog muscle fibers. However, there is no significant difference between various recording placements for intracellular action potentials [61]. This process is related to the presence of the end of the muscle fiber based on a further correspondence of the recorded data. The muscle fibers can be considered an invariant system for the time shifter because of the distance between the muscle fiber and electrode site [61]. Moreover, the response to the signal during recording the SEMG could be altered. Accordingly, a surface electrode is more responsive to the primary signal compared with the micro electrode.

Effect of electrode placement on SEMG amplitude

The amplitude of the SEMG signal can be affected by electrode placement or displacement. For example, Chris Jensen et al. described the efficacy of the electrode position on the amplitude of the SEMG signal over the trapezius muscle during arm flexion and abduction. This study indicated that the maximum amplitude was provided in the midpoint between the acromion and spine of the seventh cervical vertebra, and also the slight displacement decreased the SEMG amplitude [62]. The surface electromyography for the non-invasive assessment of muscles (SENIAM) project, which was presented for 22 various muscles, proposed to place electrode between IZ and TZ over studied muscles [63]. However, the SENIAM project excludes the forearm muscles for wrist movements, which are used for daily life activities.

After SENIAM project, the works of Hermens et al., and Farina et al. concluded that the IZ and Tendon Zone (TZ) are unsuitable for electrode placement. Because the SEMG signals, which were collected from both IZ and TZ, were unstable and unsubstantial when they were estimated in terms of magnitude [64,65]. Amplitude and Magnitude of SEMG signal calculated as average rectified value (ARV) and the mean and median frequency (MNF and MDF). Studies show that the SEMG signal has different amplitudes and values of spectra feature for various electrode positions and inter-electrode distances [66,67]. One example study is that Wong et al. examined the SEMG activities of different muscles from various electrode positions [68]. They normalized the collected the SEMG signals and compared amplitudes using root mean square and ANOVA. The results indicated that the electrode area of SEMG significantly affected the SEMG amplitude.

Effect of IZ and TZ location on SEMG signal

SEMG signals are extremely sensitive to small electrode position, particularly when the detection area is close to the IZ or TZ [65]. Nishihara et al. also demonstrated that the IZ shifted during the activities [69]. Spearing MUAPs is another difficulty of electrode placement, and can be detected over superficial muscles using the linear electrode configuration. Shifting an electrode 15 mm over the muscle may define differences in the variable estimation of SEMG. Current studies investigated the effect of distance between electrode position and IZ or TZ [70]. Electrode placement and IZ location on the torque-related patterns of responses for normalized and absolute SEMG amplitudes and mean power frequency can be affected [70].

A recent study analyzed the effect of electrode displacement in detecting SEMG signal using wavelet methods [71]. Wavelet technique was adopted to verify the differences between electrode placement over IZ and far away from IZ. Studies showed that all levels of isometric torque from the distal electrode configuration, which were selected far away from IZ, exhibited more intensity values than the SEMG signals that were acquired over IZ in 2 to 110 Hz frequency band. However, the results indicated that the lack of a significant difference in the frequency of 110 Hz. Thus, the electrode placement over IZ can affect the SEMG signals in low frequency [71].

Effect of electrode types on SEMG signal

The square or circular shape of the electrode did not significantly change the result of SEMG recording [72]. However, this study demonstrated that the mean frequency and peak-to-peak amplitude depend on the internal electrode distance and the depth of the fiber over the electrode site. Also, mathematical simulation indicates that the tissue of muscle could not function as a low pass frequency filter using either a point or rectangular electrode [73]. The electrode species, linear electrode position, and terminal phases (reflecting the excitation extinction) in MUPs, the high frequencies in the power spectrum of MUP, and the distance between MU and the electrode can change the value of cross-talk during SEMG signal recording. Hence, using high-pass filtering or differential detecting techniques could remove the in-depth cross-talk produced by MUPs. Furthermore, electrode position should correspond to the muscle anatomy to reduce the effects of cross-talk. Thus, the amount of cross-talk is significantly smaller during signal detecting using a bipolar electrode above the end-plate or beyond deep muscles.

The various electrode configurations, such as regular double differentiation, longitudinal double differentiation, transversal double differentiation, and 2-D multi-electrode shows the different value of cross-talk. Studies show that demonstrated that 2-D multi-electrode configuration exhibited higher signal and lower cross-talk compared with other types [74]. Moreover, using multi-electrode shows the cross-talk effect off neighboring muscles through both 1-D and 2-D multielectrodes during recording SEMG signals from a single MU that were selected as a convolution of intercellular AP, and the realization area could not define electrode specification depending on the uptake on the source properties.

Several studies investigated the multi- electrode configuration for the human gait. Campanini et al. recorded the SEMG signal during gait by using 2-D, 4 × 3 grid electrodes were placed over the different muscles [75]. The SEMG was specified by its peak value, and time instant corresponded to the maximum value. The results demonstrate that the SEMG intensity of muscle activities during gait depended on the electrode position. Furthermore, the findings indicated that the best electrode position could reduce the cross-talk values while detecting the activated muscle where the IED is 20 mm in both directions.

Although several SEMG studies examined the effects of electrode position and internal electrode distance on forearm muscles, only a few studies addressed the methodological difficulties of the electrode position. In the majority of studies, electrodes were placed over a bully area without the specific symptoms of the points along the length or shape of the muscle. Moreover, the inter-electrode distance varied in different studies. The reviewed publications summarized in Table 1 indicate the electrode position, inter-electrode distance, and a number of subjects.

Table 1: Summary of some reviewed studies on electrode placement or displacement of SEMG.

Quantitative studies on the sensitivity of the signal feature extracted from the SEMG signal on the recording type, including electrode position and inter-electrode-distance, for forearm muscles related to wrist movements are scarce. This limitation is significantly crucial for the repeatability of the results and the feasibility of comparing the data from various studies.

Discussion and Conclusion

Although several SEMG studies examined the effects of electrode position and internal electrode distance on forearm muscles, only a few studies addressed the methodological difficulties of the electrode position. In the majority of studies, electrodes were placed over a bully area without the specific symptoms of the points along the length or shape of the muscle. Moreover, the inter-electrode distance varied in different studies. The reviewed publications summarized in Table 1 indicate the electrode position, inter-electrode distance, and a number of subjects.

Quantitative studies on the sensitivity of the signal feature extracted from the SEMG signal on the recording type, including electrode position and inter-electrode-distance, for forearm muscles related to wrist movements are scarce. This limitation is significantly crucial for the repeatability of the results and the feasibility of comparing the data from various studies.

This literature survey was conducted to provide necessary information about SEMG signals and electrode position. The results indicated the critical importance of electrode placement or displacement in performing the variable estimation of SEMG. Several studies estimated the differences in frequency, amplitude, and velocity conduction over various electrode positions and IED. This work focused on surveying methods to demonstrate the effect of electrode position and IED. At the end of the discussion on the electrode position, a review of the most significant effects of the belly, IZ, and TZ area in the frequency and time domain of SEMG signal detecting was conducted.

Despite the abundant SEMG studies on the electrode position and internal electrode distance on forearm muscles, only a few studies have addressed the methodological difficulties of the electrode position. In the majority of studies, the electrodes were placed over a bully area without the specific symptoms of the points along the length or shape of the muscle. Moreover, the inter-electrode distance varied in different studies. Hence, finding IZ and TZ over the skin is the main difficulty of presented methods. Guideline for electrode placement needs to be developed independently of IZ and TZ locations in the future.

References

1. Zhou P, Lock B, Kuiken TA. Real time ECG artifact removal for myoelectric prosthesis control. Physiol Meas 2007; 28: 397-413.
2. Zecca M, Micera S, Carrozza MC, Dario P. Control of multifunctional prosthetic hands by processing the electromyographic signal. Crit Rev Biomed Eng 2002; 30: 459-485.
3. Yoshitake Y. Assessment of lower-back muscle fatigue using electromyography, mechanomyography, and near-infrared spectroscopy. European J Appl Physiol 2001; 84: 174-179.
4. Scheeren E. Wrist Movement Characterization by Mechanomyography Technique. J Med Biol Eng 2010; 30: 373-380.
5. Chang PF, Arendt-Nielsen L, Graven-Nielsen T, Chen AC. Psychophysical and EEG responses to repeated experimental muscle pain in humans: pain intensity encodes EEG activity. Brain Res Bull 2003; 59: 533-543.
6. Svoboda J, Sovka P, Stancák A. Intra-and inter-hemispheric coupling of electroencephalographic 8-13 Hz rhythm in humans and force of static finger extension. Neurosci lett 2002; 334: 191-195.
7. Coburn JW. Mechanomyographic and electromyographic responses of the vastus medialis muscle during isometric and concentric muscle actions. J Strength Condition Res 2005; 19: 412-420.
8. Weir JP. Mechanomyographic and electromyographic responses during fatigue in humans: influence of muscle length. European J Applied Physiol 2000; 81: 352-359.
9. Mitani H, Mushimoto E. Device for displaying masticatory muscle activities. 1982, Google Patents.
10. Fatourechi M, Bashashati A, Ward RK, Birch GE. EMG and EOG artifacts in brain computer interface systems: A survey. Clin Neurophysiol 2007; 118: 480-494.
11. Ghapanchizadeh H, Ahmad S, Ishak AJ. Investigate the transcendent adapted of wavelet threshold algorithms for elbow movement by surface EMG signal. IEEE Conference on Biomedical Engineering and Sciences (IECBES), 2014.
12. Kiguchi K, Tanaka T, Fukuda T. Neuro-fuzzy control of a robotic exoskeleton with EMG signals. IEEE Transact Fuzzy Syst 2004; 12: 481-490.
13. Cescon C, Raimondi EE, Zaest V, Drusany-Stari K, Martsidis K. Characterization of the motor units of the external anal sphincter in pregnant women with multichannel surface EMG. Int Urogynecol J 2014; 25: 1097-1103.
14. Filho EB, da Silva EA, de Carvalho MB. On EMG signal compression with recurrent patterns. IEEE Trans Biomed Eng 2008; 55: 1920-1923.
15. McKeown MJ, Torpey DC, Gehm WC. Non-invasive monitoring of functionally distinct muscle activations during swallowing. Clin Neurophysiol 2002; 113: 354-366.
16. Fattah SA. Identifying the motor neuron disease in EMG signal using time and frequency domain features with comparison. Signal Image Process 2012.
17. Ghapanchizadeh H, Ahmad SA, Ishak AJ. Developing multichannel surface EMG acquisition system by using instrument opamp INA2141. IEEE Region 10 Symposium 2014.
18. Criswell E. Cram's introduction to surface electromyography. 2010: Jones & Bartlett Publishers.
19. Soderberg GL, Cook TM. Electromyography in biomechanics. Phys Ther 1984; 64: 1813-1820.
20. van Boxtel A. Optimal signal bandwidth for the recording of surface EMG activity of facial, jaw, oral, and neck muscles. Psychophysiol 2001; 38: 22-34.
21. Day S. Important factors in surface EMG measurement. Bortec Biomedical Ltd publishers, 2002: 1-17.
22. De Luca CJ. Surface electromyography: Detection and recording. DelSys Incorporated, 2002.
23. Cutmore TR, James DA. Identifying and reducing noise in psychophysiological recordings. Int J Psychophysiol 1999; 32: 129-150.
24. Nordander C, Willner J, Hansson GA, Larsson B, Unge J. Influence of the subcutaneous fat layer, as measured by ultrasound, skinfold calipers and BMI, on the EMG amplitude. Eur J Appl Physiol 2003; 89: 514-519.
25. Masuda T, Miyano H, Sadoyama T. The position of innervation zones in the biceps brachii investigated by surface electromyography. IEEE Transact Biomed Eng 1985.
26. Koh TJ, Grabiner MD. Cross talk in surface electromyograms of human hamstring muscles. J Orthop Res 1992; 10: 701-709.
27. Young AJ, Hargrove LJ, Kuiken TA. The effects of electrode size and orientation on the sensitivity of myoelectric pattern recognition systems to electrode shift. IEEE Trans Biomed Eng 2011; 58: 2537-2544.
28. Kleine BU, Stegeman DF, Mund D, Anders C. Influence of motoneuron firing synchronization on SEMG characteristics in dependence of electrode position. J Appl Physiol (1985) 2001; 91: 1588-1599.
29. Erik SP, Englehart PK. Electromyogram pattern recognition for control of powered upper-limb prostheses: State of the art and challenges for clinical use. J Rehab Res Development 2011; 48: 643.
30. Rodriguez-Falces J, Neyroud, Place N. Influence of inter-electrode distance, contraction type, and muscle on the relationship between the sEMG power spectrum and contraction force. European J Applied Physiol 2015; 115: 627-638.
31. Saladin KS, McFarland RK. Human anatomy. 2008: McGraw-Hill New York.
32. Rüegg JC. Smooth muscle tone. Physiol Rev 1971; 51: 201-248.
33. Olivetti G. Aging, cardiac hypertrophy and ischemic cardiomyopathy do not affect the proportion of mononucleated and multinucleated myocytes in the human heart. J Mol Cell Cardiol 1996; 28: 1463-1477.
34. Pollard TD, Earnshaw WC, Lippincott-Schwartz J. Cell biology. 2007: Elsevier Health Sciences.
35. Kamen G, Gabriel D. Essentials of electromyography. 2010: Human Kinetics.
36. Lieber RL. Skeletal muscle structure, function, and plasticity. 2009: Lippincott Williams & Wilkins Baltimore.
37. MacIntosh BR, Gardiner PF, McComas AJ. Skeletal muscle: form and function. 2006: Human kinetics.
38. Frontera WR, Ochala J. Skeletal Muscle: A Brief Review of Structure and Function. Calcified Tissue Int 2014.
39. Wang WZ, Fang XH, Stephenson LL, Khiabani KT, Zamboni WA. Ischemia/reperfusion-induced necrosis and apoptosis in the cells isolated from rat skeletal muscle. J Orthop Res 2008; 26: 351-356.
40. Lu HH, Jiang J. Interface tissue engineering and the formulation of multiple-tissue systems. Adv Biochem Eng Biotechnol 2006; 102: 91-111.
41. Thorpe CT, Birch HL, Clegg PD, Screen HR. The role of the non-collagenous matrix in tendon function. Int J Exp Pathol 2013; 94: 248-259.
42. Chleboun GS. Fascicle length change of the human tibialis anterior and vastus lateralis during walking. J Orthopaedic Sport Physic Ther 2007; 37: 372-379.
43. Beretta Piccoli M, Rainoldi A, Heitz C, Wüthrich M, Boccia G. Innervation zone locations in 43 superficial muscles: toward a standardization of electrode positioning. Muscle Nerve 2014; 49: 413-421.
44. Sidhu SK, Cresswell AG, Carroll TJ. Corticospinal responses to sustained locomotor exercises: moving beyond single-joint studies of central fatigue. Sport Med 2013; 43: 437-449.
45. Dideriksen JL. Motor unit recruitment strategies and muscle properties determine the influence of synaptic noise on force steadiness. J Neurophysiol 2012; 107: 3357-3369.
46. Harris AJ, Duxson MJ, Butler JE, Hodges PW, Taylor JL. Muscle fiber and motor unit behavior in the longest human skeletal muscle. J Neurosci 2005; 25: 8528-8533.
47. Drenthen J, Jacobs BC, Maathuis EM, van Doorn PA, Visser GH. Residual fatigue in Guillain-Barre syndrome is related to axonal loss. Neurology 2013; 81: 1827-1831.
48. Lovely DF. The Origins and Nature of the Myoelectric Signal, in Powered Upper Limb Prostheses. 2004, Springer, Berlin.
49. Hogrel JY. Clinical applications of surface electromyography in neuromuscular disorders. Neurophysiologie Clinique/Clin Neurophysiol 2005; 35: 59-71.
50. Castroflorio T, Farina D, Bottin A, Debernardi C, Bracco P. Non-invasive assessment of motor unit anatomy in jaw-elevator muscles. J Oral Rehabil 2005; 32: 708-713.
51. Côté J, Mathieu PA. Mapping of the human upper arm muscle activity with an electrode matrix. Electromyogr Clin Neurophysiol 2000; 40: 215-223.
52. Falla D. Location of innervation zones of sternocleidomastoid and scalene muscles-a basis for clinical and research electromyography applications. Clin Neurophysiol 2002; 113: 57-63.
53. Fuglevand AJ, Winter DA, Patla AE, Stashuk D. Detection of motor unit action potentials with surface electrodes: influence of electrode size and spacing. Biol Cybern 1992; 67: 143-153.
54. Hogrel JY, Duchêne J, Marini JF. Variability of some SEMG parameter estimates with electrode location. J Electromyogr Kinesiol 1998; 8: 305-315.
55. Kaneko H, Kiryu T, Saitoh Y. Compensation for the distortion of bipolar surface EMG signals caused by innervation zone movement. IEICE Transact Informa Syst 1996; 79: 373-381.
56. Li W, Sakamoto K. The influence of location of electrode on muscle fiber conduction velocity and EMG power spectrum during voluntary isometric contraction measured with surface array electrodes. Appl Hum Sci 1996; 15: 25-32.
57. Lynn PA, Bettles ND, Hughes AD, Johnson SW. Influences of electrode geometry on bipolar recordings of the surface electromyogram. Med Biol Eng Comput 1978; 16: 651-660.
58. Mercer JA, Bezodis N, DeLion D, Zachry T, Rubley MD. EMG sensor location: Does it influence the ability to detect differences in muscle contraction conditions? J Electromyogr Kinesiol 2006; 16: 198-204.
59. Roy SH, De Luca CJ, Schneider J. Effects of electrode location on myoelectric conduction velocity and median frequency estimates. J Appl Physiol (1985) 1986; 61: 1510-1517.
60. Zipp P. Effect of electrode parameters on the bandwidth of the surface e.m.g. power-density spectrum. Med Biol Eng Comput 1978; 16: 537-541.
61. Gydikov A, Gerilovsky L, Radicheva N, Trayanova N. Influence of the muscle fibre end geometry on the extracellular potentials. Biol Cybern 1986; 54: 1-8.
62. Jensen C, Vasseljen O, Westgaard RH. The influence of electrode position on bipolar surface electromyogram recordings of the upper trapezius muscle. European J Appl Physiol Occupat Physiol 1993; 67: 266-273.
63. Hermens HJ. European recommendations for surface electromyography. Roessingh Res Develop 1999; 8: 13-54.
64. Hermens HJ, Freriks B, Disselhorst-Klug C, Rau G. Development of recommendations for SEMG sensors and sensor placement procedures. J Electromyogr Kinesiol 2000; 10: 361-374.
65. Farina D, Merletti R, Nazzaro M, Caruso I. Effect of joint angle on EMG variables in leg and thigh muscles. IEEE Eng Med Biol Mag 2001; 20: 62-71.
66. Castroflorio T, Farina D, Bottin A, Piancino MG, Bracco P. Surface EMG of jaw elevator muscles: effect of electrode location and inter-electrode distance. J Oral Rehabil 2005; 32: 411-417.
67. Ghapanchizadeh H, Ahmad SA, Ishak AJ. Effect of Surface Electromyography Electrode Position during Wrist Extension and Flexion Based on Time and Frequency Domain Analyses. Int J Control Theor Appl 2016; 9: 7.
68. Wong YM, Ng GY. Surface electrode placement affects the EMG recordings of the quadriceps muscles. Physic Ther Sport 2006; 7: 122-127.
69. Nishihara K, Kawai H, Chiba Y, Kanemura N, Gomi T. Investigation of innervation zone shift with continuous dynamic muscle contraction. Comput Math Methods Med 2013; 2013: 174342.
70. Beck TW, Housh TJ, Cramer JT, Malek MH, Mielke M. Electrode shift and normalization reduce the innervation zone's influence on EMG. Med Sci Sports Exerc 2008; 40: 1314-1322.
71. Beck TW, Housh TJ, Cramer JT, Stout JR, Ryan ED. Electrode placement over the innervation zone affects the low-, not the high-frequency portion of the EMG frequency spectrum. J Electromyogr Kinesiol 2009; 19: 660-666.
72. Farina D, Cescon C, Merletti R. Influence of anatomical, physical, and detection-system parameters on surface EMG. Biol Cybern 2002; 86: 445-456.
73. Dimitrova NA, Dimitrov GV, Nikitin OA. Neither high-pass filtering nor mathematical differentiation of the EMG signals can considerably reduce cross-talk. J Electromyogr Kinesiol 2002; 12: 235-246.
74. Dimitrov GV. Simulation analysis of the ability of different types of multi-electrodes to increase selectivity of detection and to reduce cross-talk. J Electromyograph Kinesiol 2003; 13: 125-138.
75. Campanini I, Merlo A, Degola P, Merletti R, Vezzosi G. Effect of electrode location on EMG signal envelope in leg muscles during gait. J Electromyogr Kinesiol 2007; 17: 515-526.