Influence of low muscle activation levels on the ankle torque and muscle shear modulus during plantar flexor stretching

During stretching studies, surface electromyography (sEMG) is used to ensure the passive state of the muscle, for the characterization of passive muscle mechanical properties. Different thresholds (1%, 2% or 5% of maximal) are indifferently used to set “passive state”. This study aimed to investigate the effects of a slight activity on the joint and muscle mechanical properties during stretching.The joint torque and muscle shear modulus of the triceps surae muscles were measured in fifteen healthy volunteers during ankle dorsiflexions: (i) in a “fully relaxed” state, (ii) during active conditions where participants were asked to produce an sEMG amplitude of 1%, 2% or 5% of their maximal sEMG amplitude of the triceps surae. The 1% condition was the only that did not result in significant differences in joint torque or shear modulus compared to the relaxed condition. In the 2% condition, increases in joint torque were found at 80% of the maximal angle in dorsiflexion, and in the shear modulus of gastrocnemius medialis and gastrocnemius lateralis at the maximal angle in dorsiflexion. During the 5% condition, joint torque and the shear modulus of gastrocnemius medialis were higher than during relaxed condition at angles larger than 40% of maximal angle in dorsiflexion. The results provide new insights on the thresholds that should be considered for the design of stretching studies. A threshold of 1% seems much more appropriate than a 2% or 5% threshold in healthy participants. Further studies are required to define similar thresholds for patients.     

Estimation of individual muscle force using elastography

Background

Estimation of an individual muscle force still remains one of the main challenges in biomechanics. In this way, the present study aimed: (1) to determine whether an elastography technique called Supersonic Shear Imaging (SSI) could be used to estimate muscle force, (2) to compare this estimation to that one provided by surface electromyography (EMG), and (3) to determine the effect of the pennation of muscle fibers on the accuracy of the estimation.

Methods and Results

Eleven subjects participated in two experimental sessions; one was devoted to the shear elastic modulus measurements and the other was devoted to the EMG recordings. Each session consisted in: (1) two smooth linear torque ramps from 0 to 60% and from 0 to 30% of maximal voluntary contraction, for the first dorsal interosseous and the abductor digiti minimi, respectively (referred to as “ramp contraction”); (2) two contractions done with the instruction to freely change the torque (referred to as “random changes contraction”). Multi-channel surface EMG recordings were obtained from a linear array of eight electrodes and the shear elastic modulus was measured using SSI. For ramp contractions, significant linear relationships were reported between EMG activity level and torque (R² = 0.949±0.036), and between shear elastic modulus and torque (R² = 0.982±0.013). SSI provided significant lower RMS_deviation between measured torque and estimated torque than EMG activity level for both types of contraction (1.4±0.7 vs. 2.8±1.4% of maximal voluntary contraction for “ramp contractions”, p<0.01; 4.5±2.3 vs. 7.9±5.9% of MVC for “random changes contractions”, p<0.05). No significant difference was reported between muscles.

Conclusion

The shear elastic modulus measured using SSI can provide a more accurate estimation of individual muscle force than surface EMG. In addition, pennation of muscle fibers does not influence the accuracy of the estimation.     

Can pennation angles be predicted from EMGs for the primary ankle plantar and dorsiflexors during isometric contractions?

Ultrasonography was used to measure pennation angle and electromyography (EMG) to record muscle activity of the human tibialis anterior (TA), lateral gastrocnemius (LG), medial gastrocnemius (MG), and soleus (SOL) muscles during graded isometric ankle plantar and dorsiflexion contractions done on a Biodex dynamometer. Data from 8 male and 8 female subjects were collected in increments of approximately 25% of maximum voluntary contraction (MVC) ranging from rest to MVC. A significant positive linear relationship (p<0.05) between normalized EMG and pennation angle for all muscles was observed when subject specific pennation angles at rest and MVC were included in the analysis. These were included to account for gender differences and inter-subject variability in pennation angle. The coefficient of determination, R(2), ranged between 0.76 for the TA and 0.87 for the SOL. The EMG-pennation angle relationships have ramifications for use in EMG-driven models of muscle force. The regression equations can be used to characterize fiber pennation angle more accurately and to determine how it changes with contraction intensity, thus providing improved estimates of muscle force when using musculoskeletal models.     

Agonist muscle activity and antagonist muscle co-activity levels during standardized isotonic and isokinetic knee extensions

This study aimed to analyze the effects of the contraction mode (isotonic vs. isokinetic concentric conditions), the joint angle and the investigated muscle on agonist muscle activity and antagonist muscle co-activity during standardized knee extensions. Twelve healthy adult subjects performed three sets of isotonic knee extensions at 40% of their maximal voluntary isometric torque followed by three sets of maximal isokinetic knee extensions on an isokinetic dynamometer. For each set, the mean angular velocity and the total external amount of work performed were standardized during the two contraction modes. Surface electromyographic activity of vastus lateralis (VL), vastus medialis (VM), rectus femoris (RF), semitendinosus (ST) and biceps femoris (BF) muscles was recorded. Root mean square values were then calculated for each 10° between 85° and 45° of knee extension (0° = horizontal position). Results show that agonist muscle activity and antagonist muscle co-activity levels are significantly greater in isotonic mode compared to isokinetic mode. Quadriceps activity and hamstrings co-activity are significantly lower at knee extended position in both contraction modes. Considering agonist muscles, VL reveals a specific pattern of activity compared to VM and RF; whereas considering hamstring muscles, BF shows a significantly higher co-activity than ST in both contraction modes. Results of this study confirmed our hypothesis that higher quadriceps activity is required during isotonic movements compared to isokinetic movements leading to a higher hamstrings co-activity.     

Comparison between muscle activation measured by electromyography and muscle thickness measured using ultrasonography for effective muscle assessment

In this study, we aimed to compare the intrarater reliability and validity of muscle thickness measured using ultrasonography (US) and muscle activity via electromyography (EMG) during manual muscle testing (MMT) of the external oblique (EO) and lumbar multifidus (MF) muscles. The study subjects were 30 healthy individuals who underwent MMT at different grades. EMG was used to measure the muscle activity in terms of ratio to maximum voluntary contraction (MVC) and root mean square (RMS) metrics. US was used to measure the raw muscle thickness, the ratio of muscle thickness at MVC, and the ratio of muscle thickness at rest. One examiner performed measurements on each subject in 3 trials. The intrarater reliabilities of the % MVC RMS and raw RMS metrics for EMG and the % MVC thickness metrics for US were excellent (ICC = 0.81–0.98). There was a significant difference between all the grades measured using the % MVC thickness metric (p < 0.01). Further, this % MVC thickness metric of US showed a significantly higher correlation with the EMG measurement methods than with the others (r = 0.51–0.61). Our findings suggest that the % MVC thickness determined by US was the most sensitive of all methods for assessing the MMT grade.     

Effect of muscle model parameter scaling for isometric plantar flexion torque prediction

This paper uses a EMG-driven Hill-type muscle model to estimate individual muscle forces of the triceps surae in isometric plantar flexion contractions. A uniform group of 20 young physical-active adult males was instructed to follow a specific contraction protocol with low (20%MVC) and medium-high (60%MVC) contractions, separated by relaxing intervals. The torque calculated by summing the individual muscle forces multiplied by the respective moment arms was compared to the torque measured by a dynamometer. Musculoskeletal parameters from the literature were used. Then, three different “correction factors” or bias have been applied on some of the muscle model parameters. These factors were based on anthropometric and dynamometric measurements: moment arm scaled by bimalleolar diameter, tendon slack length by leg length and optimal force by the maximum torque. Model torque agreement with dynamometer was recalculated with the parameter scales. It was observed that the relative torque estimation error decreased slightly but significantly when all factors were applied simultaneously (12.92±4.94% without scaling to 10.12±1.73%), which resulted mainly from the correction of the maximal muscle force parameter.     

Unchanged H-reflex during a sustained isometric submaximal plantar flexion performed with an EMG biofeedback

The aim of this study was to assess H-reflex plasticity and activation pattern of the plantar flexors during a sustained contraction where voluntary EMG activity was controlled via an EMG biofeedback. Twelve healthy males (28.0 ± 4.8 yr) performed a sustained isometric plantar flexion while instructed to maintain summed EMG root mean square (RMS) of gastrocnemius lateralis (GL) and gastrocnemius medialis (GM) muscles fixed at a target corresponding to 80% maximal voluntary contraction torque via an EMG biofeedback. Transcutaneous electrical stimulation of the posterior tibial nerve was evoked during the contraction to obtain the maximal H-reflex amplitude to maximal M-wave amplitude ratio (H_sup/M_sup ratio) from GL, GM and soleus (SOL) muscles. Neuromuscular function was also assessed before and immediately after exercise. Results showed a decrease in SOL activation during sustained flexion (from 65.5 ± 6.4% to 42.3 ± 3.8% maximal EMG, p < 0.001), whereas summed EMG RMS of GL and GM remained constant (59.7 ± 4.8% of maximal EMG on average). No significant change in the H_sup/M_sup ratio was found for SOL, GL and GM muscles. Furthermore, it appears that the decrease in maximal voluntary contraction torque (?20.4 ± 2.9%, p < 0.001) was related to both neural and contractile impairment. Overall, these findings indicate that the balance between excitation and inhibition affecting the motoneuron pool remains constant during a sustained contraction where myoelectrical activity is controlled via an EMG biofeedback or let free to vary.     

Electromyogram refinement using muscle synergy based regulation of uncertain information

Electromyogram signal (EMG) measurement frequently experiences uncertainty attributed to issues caused by technical constraints such as cross talk and maximum voluntary contraction. Due to these problems, individual EMGs exhibit uncertainty in representing their corresponding muscle activations. To regulate this uncertainty, we proposed an EMG refinement, which refines EMGs with regulating the contribution redundancy of the signals from EMGs to approximating torques through EMG-driven torque estimation (EDTE) using the muscular skeletal forward dynamic model. To regulate this redundancy, we must consider the synergistic contribution redundancy of muscles, including “unmeasured” muscles, to approximating torques, which primarily causes redundancy of EDTE. To suppress this redundancy, we used the concept of muscle synergy, which is a key concept of analyzing the neurophysiological regulation of contribution redundancy of muscles to exerting torques. Based on this concept, we designed a muscle-synergy-based EDTE as a framework for EMG refinement, which regulates the abovementioned uncertainty of individual EMGs in consideration of unmeasured muscles. In achieving the proposed EMG refinement, the most considerable point is to suppress a large change such as overestimation attributed to enhancement of the contribution of particular muscles to estimating torques. Therefore it is reasonable to refine EMGs by minimizing the change in EMGs. To evaluate this model, we used a Bland-Altman plot, which quantitatively evaluates the proportional bias of refined signals to EMGs. Through this evaluation, we showed that the proposed EDTE minimizes the bias while approximating torques. Therefore this minimization optimally regulates the uncertainty of EMGs and thereby leads to optimal EMG refinement.     

Age-related changes in neuromuscular function of the quadriceps muscle in physically active adults

Substantial evidence exists for the age-related decline in maximal strength and strength development. Despite the importance of knee extensor strength for physical function and mobility in the elderly, studies focusing on the underlying neuromuscular mechanisms of the quadriceps muscle weakness are limited.The aim of this study was to investigate the contributions of age-related neural and muscular changes in the quadriceps muscle to decreases in isometric maximal voluntary torque (iMVT) and explosive voluntary strength. The interpolated twitch technique and normalized surface electromyography (EMG) signal during iMVT were analyzed to assess changes in neural drive to the muscles of 15 young and 15 elderly volunteers. The maximal rate of torque development as well as rate of torque development, impulse and neuromuscular activation in the early phase of contraction were determined. Spinal excitability was estimated using the H reflex technique. Changes at the muscle level were evaluated by analyzing the contractile properties and lean mass.The age-related decrease in iMVT was accompanied by a decline in voluntary activation and normalized surface EMG amplitude. Mechanical parameters of explosive voluntary strength were reduced while the corresponding muscle activation remained primarily unchanged. The spinal excitability of the vastus medialis was not different while M wave latency was longer. Contractile properties and lean mass were reduced.In conclusion, the age-related decline in iMVT of the quadriceps muscle might be due to a reduced neural drive and changes in skeletal muscle properties. The decrease in explosive voluntary strength seemed to be more affected by muscular than by neural changes.     

Spatial EMG potential distribution pattern of vastus lateralis muscle during isometric knee extension in young and elderly men

The aim of the present study was to compare spatial electromyographic (EMG) potential distribution during force production between elderly and young individuals using multi-channel surface EMG (SEMG). Thirteen elderly (72-79 years) and 13 young (21-27 years) healthy male volunteers performed ramp submaximal contraction during isometric knee extension from 0% to 65% of maximal voluntary contraction. During contraction, multi-channel EMG was recorded from the vastus lateralis muscle. To evaluate alteration in heterogeneity and pattern in spatial EMG potential distribution, coefficient of variation (CoV), modified entropy and correlation coefficients with initial torque level were calculated from multi-channel SEMG at 5% force increment. Increase in CoV and decrease in modified entropy of RMS with increase of exerted torque were significantly smaller in elderly group (p < 0.05) and correlation coefficients with initial torque level were significantly higher in elderly group than in young group at moderate torque levels (p < 0.05). These data suggest that the increase of heterogeneity and the change in the activation pattern are smaller in elderly individuals than in young individuals. We speculated that multi-channel SEMG pattern in elderly individual reflects neuromuscular activation strategy regulated predominantly by clustering of similar type of muscle fibers in aged muscle.     

The contribution of the supraspinatus muscle at sub-maximal contractions

During maximum effort, the supraspinatus muscle contributes approximately 50% of the torque need to elevate the arm, but this has not been examined at sub-maximal levels. The purpose of this study was to determine the contribution of the supraspinatus muscle to shoulder elevation at sub-maximal levels. Seven healthy subjects (four males, three females) performed isometric ramp contractions at the shoulder. Middle deltoid electromyography (EMG) and force applied at the wrist were collected before and after a suprascapular nerve block. For the same level of deltoid EMG, less external force will be measured after the nerve block as the supraspinatus muscle no longer contributes. The difference between the EMG/force curve was the contribution of the supraspinatus muscle. The supraspinatus contributed 40%, 95% CI [32%–48%], to shoulder elevation. The effect of angle (p = .67) and % maximal voluntary contraction (p = .13) on supraspinatus contribution were not significant. The maximum is slightly less than reported in a previous suprascapular nerve block study using maximal contractions. The results from this study can be used to assess supraspinatus contribution in rotator cuff tears, after rehabilitation interventions, and as a restraint in computation modelling.     

Spinal reflexes and coactivation of ankle muscles during a submaximal fatiguing contraction.

This study examined the involvement of spinal mechanisms in the control of coactivation during a sustained contraction of the ankle dorsiflexors at 50% of maximal voluntary contraction. Changes in the surface electromyogram (EMG) of the tibialis anterior and of two antagonist muscles, the soleus and lateral gastrocnemius, were investigated during and after the fatigue task. Concurrently, the compound action potential (M-wave) and the Hoffmann reflex of the soleus and lateral gastrocnemius were recorded. The results showed that the torque of the ankle dorsiflexors and the average EMG of the tibialis anterior during maximal voluntary contraction declined by 40.9 +/- 17.7% (mean +/- SD; P < 0.01) and 37.0 +/- 19.9% (P < 0.01), respectively, at task failure. During the submaximal fatiguing contraction, the average EMG of both the agonist and antagonist muscles increased, leading to a nearly constant ratio at the end of the contraction when normalized to postfatigue values. In contrast to the monotonic increase in average EMG of the antagonist muscles, the excitability of their spinal reflex pathways exhibited a biphasic modulation. The amplitude of the Hoffman reflexes in the soleus and lateral gastrocnemius increased to 147.5 +/- 52.9% (P < 0.05) and 166.7 +/- 74.9% (P < 0.01), respectively, during the first 20% of the contraction and then subsequently declined to 66.3 +/- 44.8 and 74.4 +/- 44.2% of their initial values. In conclusion, the results show that antagonist coactivation did not contribute to task failure. The different changes in voluntary EMG activity and spinal reflex excitability in the antagonist muscles during the fatiguing contraction support the concept that the level of coactivation is controlled by supraspinal rather than spinal mechanisms. The findings indicate, however, that antagonist coactivation cannot simply be mediated by a central descending "common drive" to the motor neuron pools of the agonist-antagonist muscle pairs. Rather, they suggest a more subtle regulation of the drive, possibly through presynaptic mechanisms, to the motoneurons that innervate the antagonist muscles.     

Feedback controlled force enhancement and activation reduction of voluntarily activated quadriceps femoris during sub-maximal muscle action

Stretch of activated muscles leads to enhanced forces compared to isometric contractions at the same muscle length and the same level of activation. This so-called residual force enhancement (RFE) is thought to be a property of all muscles and preparations. However, observations concerning the existence, amount and duration of RFE are inconsistent, especially for voluntary activated large human muscles. Therefore, physiological relevance for daily activity is still questionable and the purpose of this study was to examine whether RFE is present in voluntary sub-maximal activated quadriceps femoris (QF). Seated in a rotational dynamometer with EMG attached to superficial parts of QF, 30 subjects performed isometric and isometric-eccentric-isometric contractions (20° stretch, ω = 60° s⁻¹) at 30% and 60% of maximum voluntary activation (MVA) and contraction (MVC). To account for the complexity of the multi-headed QF, a compensation model based on physiological cross-sectional area and individual EMG-torque relations was used to interpret EMG data. For both levels of intensity and both feedback control strategies, ANOVA identified significant RFE (at the same level of activation) and reduced activation (at the same level of torque). Against expectations, RFE was independent of the level of activation.     

Limiting mechanisms of force production after repetitive dynamic contractions in human triceps surae.

The influence of repetitive dynamic fatiguing contractions on the neuromuscular characteristics of the human triceps surae was investigated in 10 subjects. The load was 50% of the torque produced during a maximal voluntary contraction, and the exercise ended when the ankle range of motion declined to 50% of control. The maximal torque of the triceps surae and the electromyographic (EMG) activities of the soleus and medial gastrocnemius were studied in response to voluntary and electrically induced contractions before and after the fatiguing task and after 5 min of recovery. Reflex activities were also tested by recording the Hoffmann reflex (H reflex) and tendon reflex (T reflex) in the soleus muscle. The results indicated that whereas the maximal voluntary contraction torque, tested in isometric conditions, was reduced to a greater extent (P < 0.05) at 20 degrees of plantar flexion (-33%) compared with the neutral position (-23%) of the ankle joint, the EMG activity of both muscles was not significantly reduced after fatigue. Muscle activation, tested by the interpolated-twitch method or the ratio of the voluntary EMG to the amplitude of the muscle action potential (M-wave), as well as the neuromuscular transmission and sarcolemmal excitation, tested by the M-wave amplitude, did not change significantly after the fatiguing exercise. Although the H and T reflexes declined slightly (10-13%; P < 0.05) after fatigue, these adjustments did not appear to have a direct deleterious effect on muscle activation. In contrast, alterations in the mechanical twitch time course and postactivation potentiation indicated that intracellular Ca(2+)-controlled excitation-contraction coupling processes most likely played a major role in the force decrease after dynamic fatiguing contractions performed for short duration.     

The influence of seat height on the mechanical function of the triceps surae muscles during steady-rate cycling

The purpose of this study was to manipulate bicycle seat height in order to perturbate muscle length, contraction velocity and excitation of soleus and medial gastrocnemius muscles. One group of female riders (n = 13) rode a stationary ergometer at 200 W and a cadence of 80 rpm. Individuals rode at a self-selected seat height, a 10% lowered and 5% raised seat position. It was hypothesized that because the two muscles would operate at decreased contraction velocities at the low seat, the integrated EMG would be less for the lowest seat position. The soleus and medial gastrocnemius muscles showed a significant decrease in integrated EMG value with decreased seat height (soleus F_2,24 = 5.4, p < 0.01, gastrocnemius F_2,24 = 51.6, p < 0.0001). The combined effect of the movement at the ankle and knee joints resulted in increased length of gastrocnemius rather than shortening at the lowered seat-height position as anticipated. This suggested that there was a greater role of knee-joint angle in determining the muscle excitation for medial gastrocnemius. The original hypothesis was accepted, confirming the importance of setting proper seat height.     

Validity of trunk extensor and flexor torque measurements using isokinetic dynamometry

This study aimed to evaluate the validity and test–retest reliability of trunk muscle strength testing performed with a latest-generation isokinetic dynamometer. Eccentric, isometric, and concentric peak torque of the trunk flexor and extensor muscles was measured in 15 healthy subjects. Muscle cross sectional area (CSA) and surface electromyographic (EMG) activity were respectively correlated to peak torque and submaximal isometric torque for erector spinae and rectus abdominis muscles. Reliability of peak torque measurements was determined during test and retest sessions. Significant correlations were consistently observed between muscle CSA and peak torque for all contraction types (r = 0.74−0.85; P < 0.001) and between EMG activity and submaximal isometric torque (r ⩾ 0.99; P < 0.05), for both extensor and flexor muscles. Intraclass correlation coefficients were comprised between 0.87 and 0.95, and standard errors of measurement were lower than 9% for all contraction modes. The mean difference in peak torque between test and retest ranged from −3.7% to 3.7% with no significant mean directional bias. Overall, our findings establish the validity of torque measurements using the tested trunk module. Also considering the excellent test–retest reliability of peak torque measurements, we conclude that this latest-generation isokinetic dynamometer could be used with confidence to evaluate trunk muscle function for clinical or athletic purposes.     

High-density surface EMG study on the time course of central nervous and peripheral neuromuscular changes during 8 weeks of bed rest with or without resistive vibration exercise

The aim of the present study was to assess the time course and the origin of adaptations in neuromuscular function as a consequence of prolonged bed rest with or without countermeasure. Twenty healthy males volunteered to participate in the present study and were randomly assigned to either an inactive control group (Ctrl) or to a resistive vibration exercise (RVE) group. Prior to, and seven times during bed rest, we recorded high-density surface electromyogram (sEMG) signals from the vastus lateralis muscle during isometric knee extension exercise at a range of contraction intensities (5–100% of maximal voluntary isometric torque). The high-density sEMG signals were analyzed for amplitude (root mean square, RMS), frequency content (median frequency, F_med) and muscle fiber conduction velocity (MFCV) in an attempt to describe bed rest-induced changes in neural activation properties at the levels of the motor control and muscle fibers. Without countermeasures, bed rest resulted in a significant progressive decline in maximal isometric knee extension strength, whereas RMS remained unaltered throughout the bed rest period. In line with observed muscle atrophy, both F_med and MFCV declined during bed rest. RVE training during bed rest resulted in maintained maximal isometric knee extension strength, and a strong increase (～30%) in maximal EMG amplitude, from 10 days of bed rest on. Exclusion of other factors led to the conclusion that the RVE training increased motor unit firing rates as a consequence of an increased excitability of motor neurons. An increased firing rate might have been essential under training sessions, but it did not affect isometric voluntary torque capacity.     

Knee angle-specific MVIC for triceps surae EMG signal normalization in weight and non weight-bearing conditions

Varying the degree of weight-bearing (WB) and/or knee flexion (KF) angle during a plantar-flexion maximal voluntary isometric contraction (MVIC) has been proposed to alter soleus and/or gastrocnemius medialis and lateralis activation. This study compared the surface EMG signals from the triceps surae of 27 men and 27 women during WB and non weight bearing (NWB) plantar-flexion MVICs performed at 0° and 45° of KF. The aim was to determine which condition was most effective at eliciting the greatest EMG signals from soleus, gastrocnemius medialis, and gastrocnemius lateralis, respectively, for subsequent use for the normalization of EMG signals. WB was more effective than NWB at eliciting the greatest signals from soleus (p = 0.0021), but there was no difference with respect to gastrocnemius medialis and lateralis (p ? 0.2482). Although the greatest EMG signals during MVICs were more frequently elicited at 0° of KF from gastrocnemius medialis and lateralis, and at 45° from soleus (p < 0.001); neither angle consistently captured peak gastrocnemius medialis, gastrocnemius lateralis or soleus activity. The present findings encourage more consistent use of WB plantar flexion MVICs for soleus normalization; confirm that both WB and NWB procedures can elicit peak gastrocnemius activity; and emphasize the fact that no single KF angle consistently evokes selective maximal activity of any individual triceps surae muscle.     

Influence of the feature space on the estimation of hand grasping force from intramuscular EMG

The study compares the performance of different combinations of nine features extracted from intramuscular electromyogram (EMG) recordings for the estimation of grasping force within the range 0–100% maximum voluntary contraction (MVC). Single-channel intramuscular EMG was recorded from the flexor digitorum profundus (FDP) muscle from 11 subjects who exerted three force profiles during power grasping. The ability of the features to estimate force with a 1st order polynomial (poly1) and an artificial neural network (ANN) model was assessed using the adjusted coefficient of determination (R²). Willison amplitude (WAMP) and root mean square (RMS) showed the highest R² (～0.88) values for poly1. The performance of all the features to predict force significantly increased (P < 0.01) when an ANN was applied. In this case, the Modified Mean Absolute Value (MMAV) demonstrated the best performance (～0.91). The results showed that a single channel intramuscular EMG recording represents the entire grasping force range (0–100% MVC) measured from the FDP muscle. The association between EMG and force depends on the features extracted and on the model.