Electromyographic adjustments during continuous and intermittent incremental fatiguing cycling

Alpha band oscillations in common synaptic input are explanatory of the complexity of isometric knee extensor muscle torque signals

We investigated whether the strength of oscillations in common synaptic input was explanatory of knee extensor (KE) torque signal complexity during fresh and fatigued submaximal isometric contractions, in adults aged from 18 to 90 years. The discharge times of motor units were derived from the vastus lateralis muscle of 60 participants using high-density surface EMG, during 20 s isometric KE contractions at 20% of maximal voluntary contraction, performed before and after a fatiguing repeated isometric KE contraction protocol at 60% of maximal voluntary contraction. Within-muscle coherence Z-scores were estimated using frequency-domain coherence analysis, and muscle torque complexity was assessed using multiscale entropy analysis and detrended fluctuation analysis. Alpha band (5-15 Hz) coherence was found to predict 23.1% and 31.4% of the variance in the complexity index under 28-scales (CI-28) and detrended fluctuation analysis α complexity metrics, respectively, during the fresh contractions. Delta, alpha and low beta band coherence were significantly increased due to fatigue. Fatigue-related changes in alpha coherence were significantly predictive of the fatigue-related changes in CI-28 and detrended fluctuation analysis α. The fatigue-related increase in sample entropy from scales 11 to 28 of the multiscale entropy analysis curves was significantly predicted by the increase in the alpha band coherence. Age was not a contributory factor to the fatigue-related changes in within-muscle coherence and torque signal complexity. These findings indicate that the strength of alpha band oscillations in common synaptic input can explain, in part, isometric KE torque signal complexity and the fatigue-related changes in torque signal complexity.

Alterations in magnitude and spatial distribution of erector spinae muscle activity in cyclists with a recent history of low back pain

PURPOSE

While cycling offers several health benefits, repetitive loading and maintenance of static postures for prolonged periods expose cyclists to low back pain (LBP). Despite high LBP prevalence in cyclists, underlying pathomechanics and specific lumbar region muscle activation patterns during cycling are unclear. Here, we compared lumbar erector spinae (ES) muscles activation and spatial distribution activity in cyclists with and without recent LBP history.

METHODS

Ten cyclists with recent LBP history (LBPG; Oswestry Disability Index score ~ 17.8%) and 11 healthy cyclists (CG) were recruited. After assessing the Functional Threshold Power (FTP), participants underwent an incremental cycling test with 4 × 3 min steps at 70%, 80%, 90%, and 100% of their FTP. High-density surface electromyography (HDsEMG) signals were recorded from both lumbar ES using two 64-channel grids. Information about ES activation levels (root-mean-square, RMS), degree of homogeneity (entropy), and cranio-caudal displacement of muscle activity (Y-axis coordinate of the barycenter of RMS maps) was extracted from each grid separately and then grand-averaged across both grids.

RESULTS

Repeated-measure 2-way ANOVAs showed a significant intensity by group interaction for RMS amplitude (p = 0.003), entropy (p = 0.038), and Y-bar displacement (p = 0.033). LBPG increased RMS amplitude between 70-100% (+ 19%, p = 0.010) and 80-100% FTP (+ 21%, p = 0.004) and decreased entropy between 70-100% FTP (- 8.4%, p = 0.003) and 80-100% FTP (- 8.5%, p = 0.002). Between-group differences emerged only at 100% FTP (+ 9.6%, p = 0.049) for RMS amplitude.

CONCLUSION

Our findings suggest that cyclists with recent LBP history exhibit higher ES muscles activation and less homogeneous activity compared to healthy controls, suggesting potential inefficient muscle recruitment strategy.

TRIAL REGISTRATION NUMBER

HEC-DSB/09-2023.

Exploration and Application of a Muscle Fatigue Assessment Model Based on NMF for Multi-Muscle Synergistic Movements

Muscle fatigue significantly impacts coordination, stability, and speed in daily activities. Accurate assessment of muscle fatigue is vital for effective exercise programs, injury prevention, and sports performance enhancement. Current methods mostly focus on individual muscles and strength evaluation, overlooking overall fatigue in multi-muscle movements. This study introduces a comprehensive muscle fatigue model using non-negative matrix factorization (NMF) weighting. NMF is employed to analyze the duration multi-muscle weight coefficient matrix (DMWCM) during synergistic movements, and four electromyographic (EMG) signal features in time, frequency, and complexity domains are selected. Particle Swarm Optimization (PSO) optimizes feature weights. The DMWCM and weighted features combine to calculate the Comprehensive Muscle Fatigue Index (CMFI) for multi-muscle synergistic movements. Experimental results show that CMFI correlates with perceived exertion (RPE) and Speed Dynamic Score (SDS), confirming its accuracy and real-time tracking in assessing multi-muscle synergistic movements. This model offers a more comprehensive approach to muscle fatigue assessment, with potential benefits for exercise training, injury prevention, and sports medicine.

A Muscle Fatigue Classification Model Based on LSTM and Improved Wavelet Packet Threshold

Previous studies have used the anaerobic threshold (AT) to non-invasively predict muscle fatigue. This study proposes a novel method for the automatic classification of muscle fatigue based on surface electromyography (sEMG). The sEMG data were acquired from 20 participants during an incremental test on a cycle ergometer using sEMG sensors placed on the vastus rectus femoris (RF), vastus lateralis (VL), vastus medialis (VM), and gastrocnemius (GA) muscles of the left leg. The ventilation volume (VE), oxygen uptake (VO₂), and carbon dioxide production (VCO₂) data of each participant were collected during the test. Then, we extracted the time-domain and frequency-domain features of the sEMG signal denoised by the improved wavelet packet threshold denoising algorithm. In this study, we propose a new muscle fatigue recognition model based on the long short-term memory (LSTM) network. The LSTM network was trained to classify muscle fatigue using sEMG signal features. The results showed that the improved wavelet packet threshold function has better performance in denoising sEMG signals than hard threshold and soft threshold functions. The classification performance of the muscle fatigue recognition model proposed in this paper is better than that of CNN (convolutional neural network), SVM (support vector machine), and the classification models proposed by other scholars. The best performance of the LSTM network was achieved with 70% training, 10% validation, and 20% testing rates. Generally, the proposed model can be used to monitor muscle fatigue.

Understanding the effect of window length and overlap for assessing sEMG in dynamic fatiguing contractions: A non-linear dimensionality reduction and clustering

The Short-Time Fourier transform (STFT) is a helpful tool to identify muscle fatigue with clinical and sports applications. However, the choice of STFT parameters may affect the estimation of myoelectrical manifestations of fatigue. Here, we determine the effect of window length and overlap selections on the frequency slope and the coefficient of variation from EMG spectrum features in fatiguing contractions. We also determine whether STFT parameters affect the relationship between frequency slopes and task failure. Eighty-eight healthy adult men performed one-leg heel-rise until exhaustion. A factorial design with a window length of 50, 100, 250, 500, and 1000 ms with 0, 25, 50, 75, and 90% of overlap was used. The frequency slope was non-linearly fitted as a task failure function, followed by a dimensionality reduction and clustering analysis. The STFT parameters elicited five patterns. A small window length produced a higher slope frequency for the peak frequency (p < 0.001). The contrary was found for the mean and median frequency (p < 0.001). A larger window length elicited a higher slope frequency for the mean and peak frequencies. The largest frequency slope and dispersion was found for a window length of 50 ms without overlap using peak frequency. A combination of 250 ms with 50% of overlap reduced the dispersion both for peak, median, and mean frequency, but decreased the slope frequency. Therefore, the selection of STFT parameters during dynamic contractions should be accompanied by a mechanical measure of the task failure, and its parameters should be adjusted according to the experiment's requirements.

Impact of Power Output on Muscle Activation and 3D Kinematics During an Incremental Test to Exhaustion in Professional Cyclists

This study aimed to quantify the influence of an increase in power output (PO) on joint kinematics and electromyographic (EMG) activity during an incremental test to exhaustion for a population of professional cyclists. The hip flexion/extension and internal/external rotation as well as knee abduction/adduction ranges of motion were significantly decreased at 100% of the maximal aerobic power (MAP). EMG analysis revealed a significant increase in the root mean square (RMS) for all muscles from 70% of the MAP. Gastrocnemius muscles [lateralis gastrocnemius (GasL) and medialis gastrocnemius (GasM)] were the less affected by the increase of PO. Cross-correlation method showed a significant increase in the lag angle values for VM in the last stage compared to the first stage, meaning that the onset of the activation started earlier during the pedaling cycle. Statistical Parametric Mapping (SPM) demonstrated that from 70% MAP, biceps femoris (BF), tibialis anterior (TA), gluteus maximus (GM), and rectus femoris (RF) yielded larger ranges of the crank cycle on which the level of recruitment was significantly increased. This study revealed specific muscular and kinematic coordination for professional cyclists in response to PO increase.

Trunk control during repetitive sagittal movements following a real-time tracking task in people with chronic low back pain

Precision of trunk movement has commonly been examined by testing relocation accuracy rather than evaluating accuracy of tracking dynamic movement. In this study we used a 3-D motion capture system to provide a novel real-time tracking task to assess trunk motor control at varying movement speeds between people with and without chronic non-specific low back pain (LBP). Eleven asymptomatic volunteers and 15 participants with chronic non-specific LBP performed 12 continuous cycles of trunk flexion-extension following real time visual feedback, during which, trunk motion was measured using eight optoelectronic infrared cameras. Significant time differences between the feedback and actual trunk motion were found between groups (P = 0.001). Both groups had similar variability of tracking accuracy when following the feedback (P > 0.05). However, tracking variability at a slow speed correlated (P = 0.03; r = 0.55) with the Fear-Avoidance Beliefs Questionnaire (FABQ) scores in those with LBP. This study shows that both asymptomatic people and individuals with LBP displayed anticipatory behaviour, however, the response of those with LBP was consistently delayed in tracking the visual feedback compared to the asymptomatic group. Additionally, the extent of variability of tracking accuracy over repeated tracking cycles was associated with the degree of fear of movement in people with LBP.

Spatial distribution of lumbar erector spinae muscle activity in individuals with and without chronic low back pain during a dynamic isokinetic fatiguing task

BACKGROUND

Individuals with chronic low back pain (CLBP) commonly present with increased trunk muscle fatigability; typically assessed as reduced time to task failure during non-functional isometric contractions. Less is known about the specific neuromuscular responses of individuals with CLBP during dynamic fatiguing tasks. We investigate the regional alteration in muscle activation and peak torque exertion during a dynamic isokinetic fatiguing task in individuals with and without CLBP.

METHODS

Electromyography (EMG) was acquired from the lumbar erector spinae unilaterally of 11 asymptomatic controls and 12 individuals with CLBP, using high-density EMG (13 × 5 grid of electrodes). Seated in an isokinetic dynamometer, participants performed continuous cyclic trunk flexion-extension at 60^o/s until volitional exhaustion.

FINDINGS

Similar levels of muscle activation and number of repetitions were observed for both groups (p > 0.05). However, the CLBP group exerted lower levels of peak torque for both flexion and extension moments (p < 0.05). The centre of lumbar erector spinae activity was shifted cranially in the CLBP group throughout the task (p < 0.05), while the control participants showed a more homogenous distribution of muscle activity.

INTERPRETATION

People with CLBP displayed altered and potentially less efficient activation of their lumbar erector spinae during a dynamic fatiguing task. Future studies should consider using high-density EMG biofeedback to optimise the spatial activation of the paraspinal musculature in people with low back pain (LBP).

Inability to increase the neural drive to muscle is associated with task failure during submaximal contractions

Motor unit firing and contractile properties during a submaximal contraction until failure were assessed with a new tracking technique. Two distinct phases in firing behavior were observed, which compensated for changes in twitch area and predicted time to failure. However, the late increase in firing rate was below the rates attained in the absence of fatigue, which points to an inability of the central nervous system to sufficiently increase the neural drive to muscle with fatigue.

Can visual feedback on upper trapezius high-density surface electromyography increase time to task failure of an endurance task?

We investigate whether visual feedback on the spatial distribution of upper trapezius muscle activity can prolong time to task failure of sustained shoulder abduction. Surface electromyographic signals were acquired with a 13x5 grid of high-density electromyography (HDEMG) electrodes from the right upper trapezius muscle of 12 healthy volunteers as they performed sustained isometric shoulder abduction at 20% of their maximum voluntary contraction torque (MVC) until volitional exhaustion. Data were collected in two sessions; one with HDEMG visual feedback on the spatial distribution of upper trapezius activity and one without feedback. Although the HDEMG amplitude maps could be voluntarily modified by the participants during the feedback condition (significant shift in the barycenter of activity towards the cranial direction, P = 0.038), this did not influence endurance time (total endurance time with HDEMG feedback: 149.01 ± 77.07 s, no feedback 141.74 ± 60.93 s, P = 0.532). Future studies should assess whether endurance performance can be enhanced by allowing changes in arm position during the task (changing fiber tension-length relationships), by providing a more individual motor strategy, and/or by manipulating the colours used for the HDEMG maps (lighter colours for higher contraction intensities).

Reliability of muscular activation patterns and their alterations during incremental handcycling in able-bodied participants

The aim of this study was to assess muscle activity patterns (MAPs) in handcycling in terms of reliability and their alterations due to increasing workload. A total of 12 able-bodied triathletes performed an incremental step test until subjective exhaustion in a racing handcycle that was mounted on an ergometer. During the test, muscular activity of 10 muscles of the upper extremity and trunk was measured using surface electromyography (sEMG). MAPs were examined by calculating integrated EMG (iEMG), the onset, offset and range of activation (RoA). Parameters of MAPs were analysed using intraclass correlation coefficient (ICC) and two-way ANOVA with repeated measures. ICCs ranged from 0.775 to 0.999 indicating 'good' to 'excellent' reliability. All muscles increased their iEMG from low to high intensity with differing effect-sizes. Several muscles showed an earlier onset and increased RoA. MAPs in handcycling are highly reliable and alterated due to increasing workload in able-bodied participants. Whereas muscular effort can be examined in a single cycle, muscle activation characteristics require at least six to ten consecutive revolutions to achieve 'good' or 'excellent' reliability. At high intensity, many muscles demonstrated an earlier onset and larger RoA. Future studies should validate these findings in several elite handcyclists and investigate all-out sprint exercises.

Feasibility of Muscle Activity Assessment With Surface Electromyography During Bed Cycling Exercise In Intensive Care Unit Patients

Introduction: Intensive care unit (ICU) patients often develop weakness. Rehabilitation is initiated early to prevent physical deterioration, but knowledge of optimal training schedules is lacking. A reliable method to assess muscle activity during exercise is needed. In this study we explored the feasibility of electrical activity measurement by surface electromyography (sEMG) during bed cycling in ICU patients. Methods: SEMG was performed in 9 ICU patients and 6 healthy controls. A standardized 1‐minute incremental resistance bedside cycle ergometer protocol was used. Results: The median cycle time was 5.3 minutes in patients and 12.0 minutes in controls. The maximum sEMG increased in both groups; the minimal sEMG activity remained the same in patients, whereas an increase in the control group was found. Discussion: sEMG is feasible and can detect muscle activity during bed cycling in ICU patients. It may be a useful monitoring tool. Repeated measurements could possibly provide information on the effects of training. Muscle Nerve58: 688–693, 2018

Does vibration superimposed on low-level isometric contraction alter motor unit recruitment strategy?

OBJECTIVE

Beneficial effects, including improved muscle strength and power performance, have been observed during vibration exercise (VE) and partially ascribed to a specific reflex mechanism referred to as Tonic vibration reflex (TVR). TVR involves motor unit (MU) activation synchronized and un-synchronized with the vibration cycle; this suggests VE to alter the temporal MU recruitment strategy. However, the effects of VE on MU recruitment remain poorly understood. This study aims to elucidate the influence of VE on MU recruitment indirectly, by investigating the effects of low-intensity VE on muscle activation.

APPROACH

Twenty volunteers performed isometric contractions on the biceps brachii of the right arm at a baseline (low) force equal to 30% of the maximum voluntary contraction without vibration (control) and with vibration at 20, 30, 40, and 55 Hz. Three vibration amplitudes were employed at 12.5%, 25%, and 50% of the baseline. Mean muscle-fiber conduction velocity (mCV), mean frequency (MF), and root mean square (RMS) value were estimated from surface electromyography as indicators of the alteration in MU recruitment strategies.

MAIN RESULTS

The mCV estimates during VE were significantly (p  <  0.05) higher compared to the control condition. Furthermore, six VE conditions produced significantly larger RMS values compared to control condition. The estimated MF did not show any consistent trend.

SIGNIFICANCE

These results suggest that vibration superimposed on low-level isometric contraction alters the MU recruitment strategy, activating larger and faster MUs.

Surface electromyographic amplitude does not identify differences in neural drive to synergistic muscles

Surface electromyographic (EMG) signal amplitude is typically used to compare the neural drive to muscles. We experimentally investigated this association by studying the motor unit (MU) behavior and action potentials in the vastus medialis (VM) and vastus lateralis (VL) muscles. Eighteen participants performed isometric knee extensions at four target torques [10, 30, 50, and 70% of the maximum torque (MVC)] while high-density EMG signals were recorded from the VM and VL. The absolute EMG amplitude was greater for VM than VL ( P < 0.001), whereas the EMG amplitude normalized with respect to MVC was greater for VL than VM ( P < 0.04). Because differences in EMG amplitude can be due to both differences in the neural drive and in the size of the MU action potentials, we indirectly inferred the neural drives received by the two muscles by estimating the synaptic inputs received by the corresponding motor neuron pools. For this purpose, we analyzed the increase in discharge rate from recruitment to target torque for motor units matched by recruitment threshold in the two muscles. This analysis indicated that the two muscles received similar levels of neural drive. Nonetheless, the size of the MU action potentials was greater for VM than VL ( P < 0.001), and this difference explained most of the differences in EMG amplitude between the two muscles (~63% of explained variance). These results indicate that EMG amplitude, even following normalization, does not reflect the neural drive to synergistic muscles. Moreover, absolute EMG amplitude is mainly explained by the size of MU action potentials.

NEW & NOTEWORTHY Electromyographic (EMG) amplitude is widely used to compare indirectly the strength of neural drive received by synergistic muscles. However, there are no studies validating this approach with motor unit data. Here, we compared between-muscles differences in surface EMG amplitude and motor unit behavior. The results clarify the limitations of surface EMG to interpret differences in neural drive between muscles.

Differential Motor Unit Changes after Endurance or High-Intensity Interval Training

PURPOSE

Using a novel technique of high-density surface EMG decomposition and motor unit (MU) tracking, we compared changes in the properties of vastus medialis and vastus lateralis MU after endurance (END) and high-intensity interval training (HIIT).

METHODS

Sixteen men were assigned to the END or the HIIT group (n = 8 each) and performed six training sessions for 14 d. Each session consisted of 8-12 × 60-s intervals at 100% peak power output separated by 75 s of recovery (HIIT) or 90-120 min continuous cycling at ~65% V˙O2peak (END). Pre- and postintervention, participants performed 1) incremental cycling to determine V˙O2peak and peak power output and 2) maximal, submaximal (10%, 30%, 50%, and 70% maximum voluntary contraction [MVC]), and sustained (until task failure at 30% MVC) isometric knee extensions while high-density surface EMG signals were recorded from the vastus medialis and vastus lateralis. EMG signals were decomposed (submaximal contractions) into individual MU by convolutive blind source separation. Finally, MU were tracked across sessions by semiblind source separation.

RESULTS

After training, END and HIIT improved V˙O2peak similarly (by 5.0% and 6.7%, respectively). The HIIT group showed enhanced maximal knee extension torque by ~7% (P = 0.02) and was accompanied by an increase in discharge rate for high-threshold MU (≥50% knee extension MVC) (P < 0.05). By contrast, the END group increased their time to task failure by ~17% but showed no change in MU discharge rates (P > 0.05).

CONCLUSIONS

HIIT and END induce different adjustments in MU discharge rate despite similar improvements in cardiopulmonary fitness. Moreover, the changes induced by HIIT are specific for high-threshold MU. For the first time, we show that HIIT and END induce specific neuromuscular adaptations, possibly related to differences in exercise load intensity and training volume.