Tutorial. Surface EMG detection, conditioning and pre-processing: Best practices.

This tutorial is aimed primarily to non-engineers, using or planning to use surface electromyography (sEMG) as an assessment tool for muscle evaluation in the prevention, monitoring, assessment and rehabilitation fields. The main purpose is to explain basic concepts related to: (a) signal detection (electrodes, electrode-skin interface, noise, ECG and power line interference), (b) basic signal properties, such as amplitude and bandwidth, (c) parameters of the front-end amplifier (input impedance, noise, CMRR, bandwidth, etc.), (d) techniques for interference and artifact reduction, (e) signal filtering, (f) sampling and (g) A/D conversion, These concepts are addressed and discussed, with examples. The second purpose is to outline best practices and provide general guidelines for proper signal detection, conditioning and A/D conversion, aimed to clinical operators and biomedical engineers. Issues related to the sEMG origin and to electrode size, interelectrode distance and location, have been discussed in a previous tutorial. Issues related to signal processing for information extraction will be discussed in a subsequent tutorial.

Tutorial. Surface EMG detection in space and time: Best practices.

This tutorial is aimed to non-engineers using, or planning to use, surface electromyography (sEMG) as an assessment tool in the prevention, monitoring and rehabilitation fields. Its first purpose is to address the issues related to the origin and nature of the signal and to its detection (electrode size, distance, location) by one-dimensional (bipolar and linear arrays) and two-dimensional (grids) electrode systems while avoiding advanced mathematical, physical or physiological issues. Its second purpose is to outline best practices and provide general guidelines for proper signal detection. Issues related to the electrode-skin interface, signal conditioning and interpretation will be discussed in subsequent tutorials.

Extraction of the brachialis muscle activity using HD-sEMG technique and canonical correlation analysis.

The Brachialis (BR) is placed under the Biceps Brachii (BB) deep in the upper arm. Therefore, the detection of the corresponding surface Electromyogram (sEMG) is a complex task. The BR is an important elbow flexor, but it is usually not considered in the sEMG based force estimation process. The aim of this study was to attempt to separate the two sEMG activities of the BR and the BB by using a High Density sEMG (HD-sEMG) grid placed at the upper arm and Canonical Component Analysis (CCA) technique. For this purpose, we recorded sEMG signals from seven subjects with two 8 × 4 electrode grids placed over BB and BR. Four isometric voluntary contraction levels were recorded (5, 10, 30 and 50 %MVC) for 90° elbow angle. Then using CCA and image processing tools the sources of each muscle activity were separated. Finally, the corresponding sEMG signals were reconstructed using the remaining canonical components in order to retrieve the activity of the BB and the BR muscles.

Power line interference attenuation in multi-channel sEMG signals: Algorithms and analysis.

Electromyogram (EMG) recordings are often corrupted by power line interference (PLI) even though the skin is prepared and well-designed instruments are used. This study focuses on the analysis of some of the recent and classical existing digital signal processing approaches have been used to attenuate, if not eliminate, the power line interference from EMG signals. A comparison of the signal to interference ratio (SIR) of the output signals is presented, for four methods: classical notch filter, spectral interpolation, adaptive noise canceller with phase locked loop (ANC-PLL) and adaptive filter, applied to simulated multichannel monopolar EMG signals with different SIR. The effect of each method on the shape of the EMG signals is also analyzed. The results show that ANC-PLL method gives the best output SIR and lowest shape distortion compared to the other methods. Classical notch filtering is the simplest method but some information might be lost as it removes both the interference and the EMG signals. Thus, it is obvious that notch filter has the lowest performance and it introduces distortion into the resulting signals.

A new method to assess skin treatments for lowering the impedance and noise of individual gelled Ag-AgCl electrodes.

A model-based new procedure for measuring the single electrode-gel-skin impedance (ZEGS) is presented. The method is suitable for monitoring the contact impedance of the electrodes of a large array with limited modifications of the hardware and without removing or disconnecting the array from the amplifier. The procedure is based on multiple measurements between electrode pairs and is particularly suitable for electrode arrays. It has been applied to study the effectiveness of three skin treatments, with respect to no treatment, for reducing the electrode-gel-skin impedance (ZEGS) and noise: (i) rubbing with alcohol; (ii) rubbing with abrasive conductive paste; (iii) stripping with adhesive tape. The complex impedances ZEGS of the individual electrodes were measured by applying this procedure to disposable commercial Ag-AgCl gelled electrode arrays (4  ×  1) with a 5 mm(2) contact area. The impedance unbalance ΔZ = ZEGS1 - ZEGS2 and the RMS noise (VRMS) were measured between pairs of electrodes. The tissue impedance ZT was also obtained, as a collateral result. Measurements were repeated at t0 = 0 min and at t30 = 30 min from the electrode application. Mixed linear models and linear regression analysis applied to ZEGS, ΔZ and noise VRMS for the skin treatment factor demonstrated (a) that skin rubbing with abrasive conductive paste is more effective in lowering ZEGS, ΔZ and VRMS (p < 0.01) than the other treatments or no treatment, and (b) a statistically significant decrement (p < 0.01), between t0 and t30, of magnitude and phase of ZEGS.Rubbing with abrasive conductive paste significantly decreased the noise VRMS with respect to other treatments or no treatment.

Characterization of the motor units of the external anal sphincter in pregnant women with multichannel surface EMG.

INTRODUCTION AND HYPOTHESIS: Locating the innervation zones (IZs) of the external anal sphincter (EAS) is helpful to obstetricians to identify areas particularly vulnerable to episiotomy in pregnant women. The aim was to investigate the motor unit (MU) properties of the EAS during voluntary contractions. METHODS: Electromyographic signals were detected, from 478 pregnant women, by means of an intra-anal cylindrical probe carrying a circumferential array of 16 electrodes. The signals were decomposed into the constituent MU action potential trains and 5,947 templates were extracted and analyzed in order to identify the IZ position. RESULTS: MUs innervated at one end are concentrated in the dorsal portion of the sphincter, while MUs innervated in the middle are distributed symmetrically in the left and right portions of the EAS. The angular propagation velocity was estimated for each MU resulting in 260 ± 45 rad/s, corresponding to 1.8 m/s on the probe surface and to about 4 m/s at a radial depth of 10 mm from the probe surface. CONCLUSIONS: A novel method for identification and classification of MUs of the EAS is proposed and applied to a large-scale study. It is possible to distinguish MUs of the EAS in a minimally invasive way and identify their IZs. This information should be used to plan episiotomies and minimize risks of EAS denervation.

Innervation zone of the vastus medialis muscle: position and effect on surface EMG variables.

The aim of this study was to investigate the position of the innervation zone (IZ) of the vastus medialis (VM) and its effect on the electromyographic (EMG) amplitude and mean frequency estimates. Eighteen healthy subjects performed maximal isometric knee extensions at three knee angles. Surface EMG signals were collected by using a 16 × 8 electrode grid placed on the VM muscle. The position of the IZ was estimated through visual analysis, and traditional bipolar signals were obtained from channels over and away from it; amplitude and mean frequency values were extracted and compared using an analysis of variance (ANOVA) with repeated measures. The IZ is shaped as a line running from the proximal-lateral to the distal-medial aspect of the VM muscle. The presence of an IZ under the electrodes lowered the EMG amplitude (P < 0.001, F = 58.11) and increased the EMG mean frequency (P < 0.001, F = 26.47); variations of these parameters due to the knee flexion angle were less frequently observed in EMG signals collected over than away from the IZ. Electrodes placed 'over the belly of the VM muscle' are likely to collect EMG signals influenced by the presence of the IZ, thus hindering the detection of changes in muscle activity.

A novel system of electrodes transparent to ultrasound for simultaneous detection of myoelectric activity and B-mode ultrasound images of skeletal muscles.

Application of two-dimensional surface electrode arrays can provide a means of mapping motor unit action potentials on the skin surface above a muscle. The resulting muscle tissue displacement can be quantified, in a single plane, using ultrasound (US) imaging. Currently, however, it is not possible to simultaneously map spatio-temporal propagation of activation and resulting tissue strain. In this paper, we developed and tested a material that will enable concurrent measurement of two-dimensional surface electromyograms (EMGs) with US images. Specific protocols were designed to test the compatibility of this new electrode material, both with EMG recording and with US analysis. Key results indicate that, for this new electrode material, 1) the electrode-skin impedance is similar to that of arrays of electrodes reported in literature; 2) the reflection of US at the electrode-skin interface is negligible; 3) the likelihood of observing missing contacts, short-circuits, and artifacts in EMGs is not affected by the US probe; 4) movement of tissues sampled by US can be tracked accurately. We, therefore, conclude this approach will facilitate multimodal imaging of muscle to provide new spatio-temporal information regarding electromechanical function of muscle. This is relevant to basic physiology-biomechanics of active and passive force transmission through and between muscles, of motor unit spatio-temporal activity patterns, of their variation with architecture and task-related function, and of their adaptation with aging, training-exercise-disuse, neurological disease, and injury.

Identification of isometric contractions based on High Density EMG maps.

Identification of motion intention and muscle activation strategy is necessary to control human-machine interfaces like prostheses or orthoses, as well as other rehabilitation devices, games and computer-based training programs. Pattern recognition from sEMG signals has been extensively investigated in the last decades, however, most of the studies did not take into account different strengths and EMG distributions associated to the intended task. The identification of such quantities could be beneficial for the training of the subject or the control of assistive devices. Recent studies have shown the need to improve pattern-recognition classification by reducing sensitivity to changes in the exerted strength, muscle-electrode shifts and bad contacts. Surface High Density EMG (HD-EMG) obtained from 2-dimensional arrays can provide much more information than electrode pairs for inferring not only motion intention but also the strategy adopted to distribute the load between muscles as well as changes in the spatial distribution of motor unit action potentials within a single muscle because of it. The objectives of this study were: (a) the automatic identification of four isometric motor tasks associated with the degrees of freedom of the forearm: flexion-extension and supination-pronation and (b) the differentiation among levels of voluntary contraction at low-medium efforts. For this purpose, monopolar HD-EMG maps were obtained from five muscles of the upper-limb in healthy subjects. An original classifier is proposed, based on: (1) Two steps linear discriminant analysis of the EMG information for each type of contraction, and (2) features extracted from HD-EMG maps and related to its intensity and distribution in the 2D space. The classifier was trained and tested with different effort levels. Spatial distribution-based features by themselves are not sufficient to classify the type of task or the effort level with an acceptable accuracy; however, when calculated with the "isolated masses" method proposed in this study and combined with intensity-base features, the performance of the classifier is improved. The classifier is capable of identifying the tasks even at 10% of Maximum Voluntary Contraction, in the range of effort level developed by patients with neuromuscular disorders, showing that intention end effort of motion can be estimated from HD-EMG maps and applied in rehabilitation.

Accuracy assessment of CKC high-density surface EMG decomposition in biceps femoris muscle.

The aim of this study was to assess the accuracy of the convolution kernel compensation (CKC) method in decomposing high-density surface EMG (HDsEMG) signals from the pennate biceps femoris long-head muscle. Although the CKC method has already been thoroughly assessed in parallel-fibered muscles, there are several factors that could hinder its performance in pennate muscles. Namely, HDsEMG signals from pennate and parallel-fibered muscles differ considerably in terms of the number of detectable motor units (MUs) and the spatial distribution of the motor-unit action potentials (MUAPs). In this study, monopolar surface EMG signals were recorded from five normal subjects during low-force voluntary isometric contractions using a 92-channel electrode grid with 8 mm inter-electrode distances. Intramuscular EMG (iEMG) signals were recorded concurrently using monopolar needles. The HDsEMG and iEMG signals were independently decomposed into MUAP trains, and the iEMG results were verified using a rigorous a posteriori statistical analysis. HDsEMG decomposition identified from 2 to 30 MUAP trains per contraction. 3 ± 2 of these trains were also reliably detected by iEMG decomposition. The measured CKC decomposition accuracy of these common trains over a selected 10 s interval was 91.5 ± 5.8%. The other trains were not assessed. The significant factors that affected CKC decomposition accuracy were the number of HDsEMG channels that were free of technical artifact and the distinguishability of the MUAPs in the HDsEMG signal (P < 0.05). These results show that the CKC method reliably identifies at least a subset of MUAP trains in HDsEMG signals from low force contractions in pennate muscles.

Robust outlier detection in high-density surface electromyographic signals.

High Density surface Electromyography (HDsEMG) has been applied in both research and clinical applications for non-invasive neuromuscular assessment in several different fields using 2-D array. Proper interpretation of HDsEMG signals requires identifying "good" channels (where there is no short-circuit or bad-contact or major power line interference problem). Recording with many channels usually implies bad-contacts (that introduces large power line interference) and short-circuits (when using gels). In addition to online monitoring the electrode-contact quality, it is necessary to identify "bad" channels, or outliers, prior to the analysis of HDsEMG signal. In this paper we introduce a robust method to identify outliers in a set of monopolar HDsEMG signals recorded from Biceps and Triceps Brachii, Anconeus, Brachioradialis and Pronator Teres. The sensitivity and precision of this method show that this approach is promising.

Effect of composition on the dielectric properties of hydrogels for biomedical applications.

The effect of composition on the dielectric properties of hydrogels for biomedical applications is experimentally investigated. The hydrogels, containing hydroxyethylcellulose (HEC), are characterized by means of the impedance spectroscopy technique. The real and imaginary parts of the electrical impedance of the hydrogels are determined for several concentrations of HEC, and of KCl, dissolved in it, for the frequency range of interest for biomedical applications. The results are compared with those corresponding to depurated water. The equivalent conductivity of the hydrogels is evaluated by performing measurements on samples of different thicknesses. With the limit of a small concentration of salt (≤ 6%), the dependence of the electrical conductivity of the hydrogel on the salt concentration is linear.

Investigation of motor unit recruitment during stimulated contractions of tibialis anterior muscle.

This work investigated motor unit (MU) recruitment during transcutaneous electrical stimulation (TES) of the tibialis anterior (TA) muscle, using experimental and simulated data. Surface electromyogram (EMG) and torque were measured during electrically-elicited contractions at different current intensities, on eight healthy subjects. EMG detected during stimulation (M-wave) was simulated selecting the elicited MUs on the basis of: (a) the simulated current density distribution in the territory of each MU and (b) the excitation threshold characteristic of the MU. Exerted force was simulated by adding the contribution of each of the elicited MUs. The effects of different fat layer thickness (between 2 and 8mm), different distributions of excitation thresholds (random excitation threshold, higher threshold for larger MUs or smaller MUs), and different MU distributions within the muscle (random distribution, larger MU deeper in the muscle, smaller MU deeper) on EMG variables and torque were tested. Increase of the current intensity led to a first rapid increase of experimental M-wave amplitude, followed by a plateau. Further increases of the stimulation current determined an increase of the exerted force, without relevant changes of the M-wave. Similar results were obtained in simulations. Rate of change of conduction velocity (CV) and leading coefficient of the second order polynomial interpolating the force vs. stimulation level curve were estimated as a function of increasing current amplitudes. Experimental data showed an increase of estimated CV with increasing levels of the stimulation current (for all subjects) and a positive leading coefficient of force vs. stimulation current curve (for five of eight subjects). Simulations matched the experimental results only when larger MUs were preferably located deeper in the TA muscle (in line with a histochemical study). Marginal effect of MU excitation thresholds was observed, suggesting that MUs closer to the stimulation electrode are recruited first during TES regardless of their excitability.

Time and frequency domain analysis of surface myoelectric signals during electrically-elicited cramps.

OBJECTIVES: To examine if different frequencies of electrical stimulation trigger different sized cramps in the abductor hallucis muscle and to analyze their surface electromyographic (EMG) behaviour in both time and frequency domains. METHODS: Fifteen subjects were studied. Stimulation trains of 150 pulses were applied to the muscle motor point. Frequency was increased (starting from 4pps with 2-pps steps) until a cramp developed. Current intensity was 30% higher than that eliciting maximal M-waves. After the first cramp ("threshold cramp"), a 30-minute rest was provided before a second cramp ("above-threshold cramp") was elicited with a frequency increased by 50% with respect to that eliciting the first cramp. RESULTS: We found greater EMG amplitude and a compression of the power spectrum for above-threshold cramps with respect to threshold cramps. M-wave changes (ranging between small decreases of M-wave amplitude to complete M-wave disappearance) occurred and progressively increased throughout stimulation trains. Significant positive correlations were found between estimates of EMG amplitude during cramps and estimated reductions of M-wave amplitude. CONCLUSIONS: Varying frequencies of electrical stimulation triggered different sized cramps. Moreover, decreases in M-wave amplitude were observed during both threshold and above-threshold stimulations. The choice of the stimulation frequency has relevance for optimizing electrical stimulation protocols for the study of muscle cramps in both healthy and pathological subjects.

Myoelectric fatigue profiles of three knee extensor muscles.

Aims of the present study were to: 1) investigate the differences between the myoelectric fatigue profiles of the vasti muscles of the quadriceps during electrically evoked contractions; 2) compare the myoelectric fatigue profiles of the vasti muscles between sedentary subjects and rowers; 3) analyze motor unit activation order during stimulation of the vasti muscles. In nine sedentary subjects and nine rowers surface EMG signals were detected during electrically elicited contractions of the following three muscles: vastus medialis obliquus (VMO), vastus lateralis (VL), and vastus medialis longus (VML). M-waves were recorded as the muscles were stimulated with both variable (increasing-decreasing) and constant stimulation intensities. Changes in M-wave conduction velocity (CV) during trains with non-constant current were adopted for the study of the motor unit recruitment order. Rates of change of myoelectric signal variables were adopted to assess myoelectric manifestations of fatigue during stimulation trains with constant current. We found that: 1) VL muscle was more fatigable than vastus medialis muscles; 2) VL and VML muscles of rowers resulted less fatigable than sedentary subjects; and 3) in the three muscles, motor units tended to be recruited in order of increasing CV and derecruited in order of decreasing CV with increasing/decreasing stimulation current.

Surface EMG: the issue of electrode location.

This paper contributes to clarifying the conditions under which electrode position for surface EMG detection is critical and leads to estimates of EMG variables that are different from those obtained in other nearby locations. Whereas a number of previous works outline the need to avoid the innervation zone (or the muscle belly), many authors place electrodes in the central part or bulge of the muscle of interest where the innervation zone is likely to be. Computer simulations are presented to explain the effect of the innervation zone on amplitude, frequency and conduction velocity estimates from the signal and the need to avoid placing electrodes near it. Experimental signals recorded from some superficial muscles of the limbs and trunk (abductor pollicis brevis, flexor pollicis brevis, biceps, upper trapezius, vastus medialis, vastus lateralis) were processed providing support for the findings obtained from simulations. The use of multichannel techniques is recommended to estimate the location of the innervation zone and to properly choose the optimal position of the detection point(s) allowing meaningful estimates of EMG variables during movement analysis.

Pulse charge and not waveform affects M-wave properties during progressive motor unit activation.

The aim of this study was to investigate changes in experimentally recorded M-waves with progressive motor unit (MU) activation induced by transcutaneous electrical stimulation with different pulse waveforms. In 10 subjects, surface electromyographic signals were detected with a linear electrode array during electrically elicited contractions of the biceps brachii muscle. Three different monophasic waveforms of 304-micros duration were applied to the stimulation electrode on the main muscle motor point: triangular, square, and sinusoidal. For each waveform, increasing stimulation current intensities were applied in 10 s (frequency: 20 Hz). It was found that: (a) the degree of MU activation, as indicated by M-wave average rectified value, was a function of the injected charge and not of the stimulation waveform, and (b) MUs tended to be recruited in order of increasing conduction velocity with increasing charge of transcutaneous stimulation. Moreover, the subjects reported lower discomfort during the contractions elicited by the triangular waveform with respect to the others. Since subject tolerance to the stimulation protocol must be considered as important as MU recruitment in determining the effectiveness of neuromuscular electrical stimulation (NMES), we suggest that both charge and waveform of the stimulation pulses should be considered relevant parameters for optimizing NMES protocols.

Effect of electrode location on EMG signal envelope in leg muscles during gait.

The aim of the study was to assess the variability of EMG signal envelope with electrode location during gait. Surface EMG signals were recorded from 10 healthy subjects from the tibialis anterior (TA), peroneus longus (PL), gastrocnemius medialis (GM), gastrocnemius lateralis (GL), and soleus (SO) muscles. From TA, PL, GL and GM, signals were acquired using a two-dimensional grid of 4 x 3 electrodes (10 x 15 mm in size, as used in most gait laboratories) with 20-mm interelectrode distance in both directions. A similar grid of 3 x 3 electrodes was used for SO. EMG envelope was characterized by its peak value, area after normalization by the peak value, and time instant corresponding to the maximum. The maximum relative change in peak value with electrode location, expressed as a percentage of the peak value in the central location, was (mean+/-SD) 31+/-18% for TA, 29+/-13% for PL, 25+/-15% for GL, 14+/-8% for GM, and 26+/-14% for SO. The maximum relative change in area was 29+/-13% for TA, 73+/-40% for PL, 31+/-23% for GL, 35+/-20% for GM, 20+/-13% for SO, and in the position of maximum, computed as distance from the maximum position in the central channel, it was 5+/-10% of the gait cycle for TA, 26+/-16% for PL, 3+/-2% for GL, 3+/-1% for GM, 3+/-3% for SO. A crosstalk index, defined on the basis of the expected intervals of muscle activation for healthy subjects, indicated that estimated crosstalk was present between TA and PL, in an amount which depended on electrode location. It was concluded that the estimate of muscle activation intensity during gait from surface EMG is variable with location of the electrodes while timing of muscle activity is more robust to electrode displacement and can be reliably extracted in those cases in which crosstalk is limited. These results are valid for healthy subjects, where the level of muscular activity during gait is much lower than maximum.

[Usefulness of surface electromyography of hand muscles in the assessment of myoelectric parameters changes due to repetitive manual tasks]. / Applicazione dell'elettromiografia di superficie dei muscoli della mano nello studio delle patologie da sovraccarico biomeccanico.

The aim of this project was to investigate the possible role of sEMG in the diagnosis of Carpal tunnel syndrome (CTS). The study group consisted of 37 subjects, of whom 14 (control group) were not employed in manual tasks and 23 (exposed) were engaged in repetitive and forceful manual tasks. Of the 23 exposed workers, 10 reported CTS symptoms, whereas all the subjects of the control group resulted asymptomatic. The surface electromyography signal was recorded from the abductor pollicis brevis muscle, using different levels of isometric contraction: 20% and 50% of the maximum voluntary contraction (MVC), respectively. The initial values and rate of change of the average rectified value (ARV), mean power spectral frequency (MNF) and conduction velocity (CV) were calculated. Moreover the study protocol included clinical evaluation and electrodiagnostic study of the median nerve. Data from the exposed and control group were compared. Statistically significant differences between the two groups were found for ARV initial value and for CV and MNF rate of change at 50% MVC. These parameters resulted lower in the exposed group, with the lowest values among symptomatic subjects. Possible explanations may be the loss of motor units, particularly affecting the fast and fatigable type II muscle fibers, involved in the myoelectric manifestation of fatigue. In conclusion, this technique was able to show different myoelectric patterns and manifestations of fatigue between subjects exposed and nonexposed to manual intensive work, suffering or not suffering from CTS.

Neuromuscular assessment in elderly workers with and without work related shoulder/neck trouble: the NEW-study design and physiological findings.

Musculoskeletal disorders in the neck and shoulder area are a major occupational concern in the European countries especially among elderly females. The aim was to assess these disorders based on quantitative EMG indicators and functional tests. 252 female computer users (45-68 years) were recruited from four European countries in two contrast groups: (1) 88 neck/shoulder (NS) cases reporting trouble in the neck and/or shoulder region for more than 30 days during the last year, and (2) 164 NS-controls reporting such trouble for no more than 7 days. Questionnaires, functional/clinical tests, and physiological recordings were performed in workplace related field studies. The results showed no differences in anthropometrics but NS-cases reported more strained head positions and more eye problems than controls. The psychosocial working factors were similar, although, NS-controls had slightly better scores on working conditions, general health, and vitality compared to cases. The NS-cases had lower maximal voluntary contraction (MVC) during shoulder elevation (mean (SD) 310 (122) N) compared to the controls (364 (122) N). During 30% MVC electromyography (EMGrms) in the trapezius muscle was lower in NS-cases (194 (105) muV) than in controls (256 (169) muV), while no differences were found regarding endurance time. Estimated conduction velocity was not different between NS-cases and -controls. Four functional computer tests were performed equally well by NS-cases and -controls, and the corresponding EMG variables also did not differ. A major finding in this large-scale epidemiological study is the significantly lower MVC in NS-cases compared with NS-controls together with lower EMGrms value at 30% MVC, while computer tasks were performed at similar relative muscle activation. The study was unable to reveal quantitative EMG indicators and functional tests that could objectively assess disorders in NS-cases.