High-density Electromyography Research Articles

Supraspinal, spinal, and motor unit adjustments to fatiguing isometric contractions of the knee extensors at low and high submaximal intensities in males.

Contraction intensity is a key factor determining the development of muscle fatigue and it has been shown to induce distinct changes along the motor pathway. The role of cortical and spinal inputs that regulate motor unit (MU) behaviour during fatiguing contractions is poorly understood. We studied the cortical, spinal, and neuromuscular response to sustained fatiguing isometric tasks performed at 20 and 70% of the maximum isometric voluntary contraction (MVC), together with MUs behaviour of knee extensors in healthy active males. Neuromuscular function was assessed before and after performing both tasks. Cortical and spinal responses during exercise were measured via stimulation of the motor cortex and spinal cord. High density electromyography was used to record individual MUs from the vastus lateralis (VL). Exercise at 70% MVC induced greater decline in MVC (p = 0.023), and potentiated twitch force compared to 20%MVC (p < .001), with no difference in voluntary activation (p = 0.514). Throughout exercise, corticospinal responses were greater during the 20%MVC task (p < 0.001), and spinal responses increased over time in both tasks (p ≤ 0.042). MU discharge rate increased similarly following both tasks (p ≤ 0.043) while recruitment and de-recruitment thresholds were unaffected (p ≥ 0.295). These results suggest that increased excitability of cortical and spinal inputs might be responsible for the increase in MU discharge rate. The increase in evoked responses together with the higher MUs discharge rate might be required to compensate for peripheral adjustments to sustain fatiguing contractions at different intensities.

Graph-Driven Simultaneous and Proportional Estimation of Wrist Angle and Grasp Force via High-Density EMG.

Myoelectric prostheses are generally unable to accurately control the position and force simultaneously, prohibiting natural and intuitive human-machine interaction. This issue is attributed to the limitations of myoelectric interfaces in effectively decoding multi-degree-of-freedom (multi-DoF) kinematic and kinetic information. We thus propose a novel multi-task, spatial-temporal model driven by graphical high-density electromyography (HD-EMG) for simultaneous and proportional control of wrist angle and grasp force. Twelve subjects were recruited to perform three multi-DoF movements, including wrist pronation/supination, wrist flexion/extension, and wrist abduction/adduction while varying grasp force. Experimental results demonstrated that the proposed model outperformed five baseline models, with the normalized root mean square error of 13.2% and 9.7% and the correlation coefficient of 89.6% and 91.9% for wrist angle and grasp force estimation, respectively. In addition, the proposed model still maintained comparable accuracy even with a significant reduction in the number of HD-EMG electrodes. To the best of our knowledge, this is the first study to achieve simultaneous and proportional wrist angle and grasp force control via HD-EMG and has the potential to empower prostheses users to perform a broader range of tasks with greater precision and control, ultimately enhancing their independence and quality of life.

A Deep Learning Sequential Decoder for Transient High-Density Electromyography in Hand Gesture Recognition Using Subject-Embedded Transfer Learning

A Deep Learning Sequential Decoder for Transient High-Density Electromyography in Hand Gesture Recognition Using Subject-Embedded Transfer Learning

Developing RPC-Net: Leveraging High-Density Electromyography and Machine Learning for Improved Hand Position Estimation.

The purpose of this study was to develop and evaluate the performance of RPC-Net (Recursive Prosthetic Control Network), a novel method using simple neural network architectures to translate electromyographic activity into hand position with high accuracy and computational efficiency. RPC-Net uses a regression-based approach to convert forearm electromyographic signals into hand kinematics. We tested the adaptability of the algorithm to different conditions and compared its performance with that of solutions from the academic literature. RPC-Net demonstrated a high degree of accuracy in predicting hand position from electromyographic activity, outperforming other solutions with the same computational cost. Including previous position data consistently improved results across subjects and conditions. RPC-Net showed robustness against a reduction in the number of electromyography electrodes used and shorter input signals, indicating potential for further reduction in computational cost. The results demonstrate that RPC-Net is capable of accurately translating forearm electromyographic activity into hand position, offering a practical and adaptable tool that may be accessible in clinical settings. The development of RPC-Net represents a significant advancement. In clinical settings, its application could enable prosthetic devices to be controlled in a way that feels more natural, improving the quality of life for individuals with limb loss.

High-Density Electromyography Excitation in Front vs. Back Lat Pull-Down Prime Movers.

The current study compared the spatial excitation of the primary muscles during the lat pull-down exercise with the bar passing in front (front-LPD) or behind the neck (back-LPD) using high-density electromyography. Fourteen resistance trained men performed a front-LPD or a back-LPD within a non-fatiguing set with 8-RM as the external load. The muscle excitation centroid of latissimus dorsi, middle trapezius, pectoralis major, biceps brachii, triceps brachii and posterior deltoid muscles were recorded during the ascending and the descending phase. During the descending phase, the front-LPD showed superior excitation of the latissimus dorsi (ES = 0.97) and the pectoralis major (ES = 1.17), while in the ascending phase, the back-LPD exhibited superior excitation of the latissimus dorsi (ES = 0.63), and the front-LPD showed superior excitation of the biceps brachii (ES = 0.41) and the posterior deltoid (ES = 1.77). During the descending phase, the front-LPD showed a more lateral centroid of the latissimus dorsi (ES = 0.60), the biceps brachii (ES = 0.63) and the triceps brachii (ES = 0.98), while the centroid was more medial for the middle trapezius (ES = 0.58). The centroid of the middle trapezius was also more medial in the front-LPD during the ascending phase (ES = 0.85). The pectoralis major centroid was more cranial in the front-LPD for both the descending (ES = 1.58) and the ascending phase (ES = 0.88). The front-LPD appears to provide overall greater excitation in the prime movers. However, distinct spatial excitation patterns were observed, making exercise suitable for the training routine.

Leveraging High-Density EMG to Investigate Bipolar Electrode Placement for Gait Prediction Models

Leveraging High-Density EMG to Investigate Bipolar Electrode Placement for Gait Prediction Models

Influence of spatio-temporal filtering on hand kinematics estimation from high-density EMG signals *

Abstract Objective. Surface electromyography (sEMG) is a non-invasive technique that records the electrical signals generated by muscles through electrodes placed on the skin. sEMG is the state-of-the-art method used to control active upper limb prostheses because of the association between its amplitude and the neural drive sent from the spinal cord to muscles. However, accurately estimating the kinematics of a freely moving human hand using sEMG from extrinsic hand muscles remains a challenge. Deep learning has been recently successfully applied to this problem by mapping raw sEMG signals into kinematics. Nonetheless, the optimal number of EMG signals and the type of pre-processing that would maximize performance have not been investigated yet. Approach. Here, we analyze the impact of these factors on the accuracy in kinematics estimates. For this purpose, we processed monopolar sEMG signals that were originally recorded from 320 electrodes over the forearm muscles of 13 subjects. We used a previously published deep learning method that can map the kinematics of the human hand with real-time resolution. Main results. While myocontrol algorithms essentially use the temporal envelope of the EMG signal as the only EMG feature, we show that our approach requires the full bandwidth of the signal in the temporal domain for accurate estimates. Spatial filtering however, had a smaller impact and low-order spatial filters may be suitable. Moreover, reducing the number of channels by ablation resulted in large performance losses. The highest accuracy was reached with the highest number of available sensors (n = 320). Importantly and unexpected, our results suggest that increasing the number of channels above those used in this study may further enhance the accuracy in predicting the kinematics of the human hand. Significance. We conclude that full bandwidth high-density EMG systems of hundreds of electrodes are needed for accurate kinematic estimates of the human hand.

Design, Fabrication and Evaluation of a Stretchable High-Density Electromyography Array.

The adoption of high-density electrode systems for human-machine interfaces in real-life applications has been impeded by practical and technical challenges, including noise interference, motion artefacts and the lack of compact electrode interfaces. To overcome some of these challenges, we introduce a wearable and stretchable electromyography (EMG) array, and present its design, fabrication methodology, characterisation, and comprehensive evaluation. Our proposed solution comprises dry-electrodes on flexible printed circuit board (PCB) substrates, eliminating the need for time-consuming skin preparation. The proposed fabrication method allows the manufacturing of stretchable sleeves, with consistent and standardised coverage across subjects. We thoroughly tested our developed prototype, evaluating its potential for application in both research and real-world environments. The results of our study showed that the developed stretchable array matches or outperforms traditional EMG grids and holds promise in furthering the real-world translation of high-density EMG for human-machine interfaces.

More Variability in Tibialis Anterior Function during the Adduction of the Foot than Dorsiflexion of the Ankle.

The aim of the study was to compare maximal force, force steadiness, and the discharge characteristics of motor units in the tibialis anterior (TA) muscle during submaximal isometric contractions for ankle dorsiflexion and adduction of the foot. Nineteen active young adults performed maximal and submaximal isometric dorsiflexion and adduction contractions at five target forces (5%, 10%, 20%, 40%, and 60% maximal voluntary contraction [MVC]). The activity of motor units in TA was recorded by high-density EMG. The maximal force was similar between dorsiflexion and adduction, despite EMG amplitude for TA being greater ( P < 0.05) during dorsiflexion than adduction. Τhe coefficient of variation (CV) for force (force steadiness) during dorsiflexion was always less ( P < 0.05) than during adduction, except of 5% MVC force. No differences were observed for mean discharge rate; however, the regression between the changes in discharge rate relative to the change of force was significant for dorsiflexion ( R2 = 0.25, P < 0.05) but not for adduction. Discharge variability, however, was usually less during dorsiflexion. The CV for interspike interval was less ( P < 0.05) at 10%, 20%, and 40% MVC but greater at 60% MVC during dorsiflexion than adduction. Similarly, the SD values of the filtered cumulative spike train of the motor units in TA were less ( P < 0.05) at 5%, 10%, 20%, and 40% MVC during dorsiflexion than adduction. Although the mean discharge rate of motor units in TA was similar during foot adduction and ankle dorsiflexion, discharge variability was less during dorsiflexion resulting in less accurate performance of the steady adduction contractions. The neural drive to bifunctional muscles differs during their accessory function, which must be considered for training and rehabilitation interventions.

Simultaneous Estimation of Digit Tip Forces and Hand Postures in a Simulated Real-life Condition with High-density Electromyography and Deep Learning

Simultaneous Estimation of Digit Tip Forces and Hand Postures in a Simulated Real-life Condition with High-density Electromyography and Deep Learning

Generalizing Upper Limb Force Modeling With Transfer Learning: A Multimodal Approach Using EMG and IMU for New Users and Conditions.

In the field of EMG-based force modeling, the ability to generalize models across individuals could play a significant role in its adoption across a range of applications, including assistive devices, robotic and rehabilitation devices. However, current studies have predominately focused on intra-subject modeling, largely neglecting the burden of end-user data acquisition. In this work, we propose the use of transfer learning (TL) to generalize force modeling to a new user by first establishing a baseline model trained using other users' data, and then adapting to the end-user using a small amount of new data (only 10% , 20% , and 40% of the new user data). Using a deep multimodal convolutional neural network, consisting of two CNN models, one with high-density (HD) EMG and one with motion data recorded by an Inertial Measurement Unit (IMU), our proposed TL technique significantly improved force modeling compared to leave-one-subject-out (LOSO) and even intra-subject scenarios. The TL approach increased the average R squared values of the force modeling task by 60.81%, 190.53%, and 199.79% compared to the LOSO case, and by 13.4%, 36.88%, and 45.51% compared to the intra-subject case for isotonic, isokinetic and dynamic conditions, respectively. These results show that it is possible to adapt to a new user with minimal data while improving performance significantly compared to the intra-subject scenario. We also show that TL can be used to generalize on a new experimental condition for a new user.

Towards Efficient Neural Decoder for Dexterous Finger Force Predictions.

Dexterous control of robot hands requires a robust neural-machine interface capable of accurately decoding multiple finger movements. Existing studies primarily focus on single-finger movement or rely heavily on multi-finger data for decoder training, which requires large datasets and high computation demand. In this study, we investigated the feasibility of using limited single-finger surface electromyogram (sEMG) data to train a neural decoder capable of predicting the forces of unseen multi-finger combinations. We developed a deep forest-based neural decoder to concurrently predict the extension and flexion forces of three fingers (index, middle, and ring-pinky). We trained the model using varying amounts of high-density EMG data in a limited condition (i.e., single-finger data). We showed that the deep forest decoder could achieve consistently commendable performance with 7.0% of force prediction errors and R2 value of 0.874, significantly surpassing the conventional EMG amplitude method and convolutional neural network approach. However, the deep forest decoder accuracy degraded when a smaller amount of data was used for training and when the testing data became noisy. The deep forest decoder shows accurate performance in multi-finger force prediction tasks. The efficiency aspect of the deep forest lies in the short training time and small volume of training data, which are two critical factors in current neural decoding applications. This study offers insights into efficient and accurate neural decoder training for advanced robotic hand control, which has the potential for real-life applications during human-machine interactions.

Multi-view fusion network-based gesture recognition using sEMG data.

sEMG(surface electromyography) signals have been widely used in rehabilitation medicine in the past decades because of their non-invasive, convenient and informative features, especially in human action recognition, which has developed rapidly. However, the research on sparse EMG in multi-view fusion has made less progress compared to high-density EMG signals, and for the problem of how to enrich sparse EMG feature information, a method that can effectively reduce the information loss of feature signals in the channel dimension is needed. In this paper, a novel IMSE (Inception-MaxPooling-Squeeze- Excitation) network module is proposed to reduce the loss of feature information during deep learning. Then, multiple feature encoders are constructed to enrich the information of sparse sEMG feature maps based on the multi-core parallel processing method in multi-view fusion networks, while SwT (Swin Transformer) is used as the classification backbone network. By comparing the feature fusion effects of different decision layers of the multi-view fusion network, it is experimentally obtained that the fusion of decision layers can better improve the classification performance of the network. In NinaPro DB1, the proposed network achieves 93.96% average accuracy in gesture action classification with the feature maps obtained in 300ms time window, and the maximum variation range of action recognition rate of individuals is less than 11.2%. The results show that the proposed framework of multi-view learning plays a good role in reducing individuality differences and augmenting channel feature information, which provides a certain reference for non-dense biosignal pattern recognition.

Motoneuron-driven computational muscle modelling with motor unit resolution and subject-specific musculoskeletal anatomy.

The computational simulation of human voluntary muscle contraction is possible with EMG-driven Hill-type models of whole muscles. Despite impactful applications in numerous fields, the neuromechanical information and the physiological accuracy such models provide remain limited because of multiscale simplifications that limit comprehensive description of muscle internal dynamics during contraction. We addressed this limitation by developing a novel motoneuron-driven neuromuscular model, that describes the force-generating dynamics of a population of individual motor units, each of which was described with a Hill-type actuator and controlled by a dedicated experimentally derived motoneuronal control. In forward simulation of human voluntary muscle contraction, the model transforms a vector of motoneuron spike trains decoded from high-density EMG signals into a vector of motor unit forces that sum into the predicted whole muscle force. The motoneuronal control provides comprehensive and separate descriptions of the dynamics of motor unit recruitment and discharge and decodes the subject's intention. The neuromuscular model is subject-specific, muscle-specific, includes an advanced and physiological description of motor unit activation dynamics, and is validated against an experimental muscle force. Accurate force predictions were obtained when the vector of experimental neural controls was representative of the discharge activity of the complete motor unit pool. This was achieved with large and dense grids of EMG electrodes during medium-force contractions or with computational methods that physiologically estimate the discharge activity of the motor units that were not identified experimentally. This neuromuscular model advances the state-of-the-art of neuromuscular modelling, bringing together the fields of motor control and musculoskeletal modelling, and finding applications in neuromuscular control and human-machine interfacing research.

Can intermittent changes in trunk extensor muscle length delay muscle fatigue development?

Muscle length changes may evoke alternating activity and consequently reduce local fatigue and pain during prolonged static bending. The aim of this study was to assess whether a postural intervention involving intermittent trunk extensor muscle length changes (INTERMITTENT) can delay muscle fatigue during prolonged static bending when compared to a near-isometric condition (ISOMETRIC) or when participants were allowed to voluntarily vary muscle length (VOLUNTARY). These three conditions were completed by 11 healthy fit male participants, in three separate sessions of standing with 30 ± 3 degrees trunk inclination until exhaustion. Conventional and high-density electromyography (convEMG and HDsEMG, respectively) were measured on the left and right side of the spine, respectively. The endurance time for INTERMITTENT was 33.6% greater than ISOMETRIC (95% CI: [3.8, 63.5]; p = 0.027) and 29.4% greater than VOLUNTARY (95% CI: [7.0, 51.7]; p = 0.010), but not different between ISOMETRIC and VOLUNTARY. The convEMG and HDsEMG amplitude coefficient of variation was significantly greater for INTERMITTENT versus ISOMETRIC. The rate of change in convEMG and HDsEMG spectral content did not reveal significant differences between conditions as found in endurance time. Additional regression analyses between endurance time and rate of change in convEMG (p > 0.05) and HDsEMG (R2 = 0.39–0.65, p = 0.005–0.039) spectral content indicated that HDsEMG better reflects fatigue development in low-level contractions. In conclusion, imposed intermittent trunk extensor muscle length changes delayed muscle fatigue development when compared to a near-isometric condition or when participants were allowed to voluntarily vary muscle length, possibly due to evoking alternating activity between/within trunk extensor muscles.

High-density EMG, IMU, kinetic, and kinematic open-source data for comprehensive locomotion activities

Novel sensor technology enables new insights in the neuromechanics of human locomotion that were previously not possible. Here, we provide a dataset of high-density surface electromyography (HDsEMG) and high-resolution inertial measurement unit (IMU) signals, along with motion capture and force data for the lower limb of 10 healthy adults during multiple locomotion modes. The participants performed level-ground and slope walking, as well as stairs ascent/descent, side stepping gait, and stand-to-walk and sit-to-stand-to-walk, at multiple walking speeds. These data can be used for the development and validation of locomotion mode recognition and control algorithms for prosthetics, exoskeletons, and bipedal robots, and for motor control investigations.

Fatigue task-dependent effect on spatial distribution of lumbar muscles activity

This study aims to identify how spatial distribution of lumbar muscle activity is modulated by different fatigue tasks. Twenty healthy adults performed two different isometric trunk extension endurance tasks (the modified Sorensen test and the inverted modified Sorensen test) until exhaustion. During these tasks, bilateral superficial lumbar muscle activity was recorded using high-density electromyography. The spatial distribution of activation within these muscles was obtained using the centroid coordinates in the medio-lateral and cranio-caudal directions. The effects of task and endurance time (left and right sides) were investigated using repeated measures ANOVA. Results revealed a significant lateral shift of the centroid throughout the fatigue tasks on both sides and no difference between tasks. Significant task × time interaction effects were found for the cranio-caudal direction on both sides showing a significantly more caudal location of the centroid in the modified Sorensen test compared to the inverted test at the beginning of the tasks. Our findings suggest that spatial distribution of lumbar muscle activity is task-dependent in a pre-fatigue stage while an alternative but similar muscle recruitment strategy is used in both tasks to maintain performance in the later stages of muscle fatigue.

Trunk extensor muscle endurance and its relationship to action potential conduction velocity and spectral parameters estimated using high-density electromyography

Trunk extensor muscle fatigue typically manifests as a decline in spectral content of surface electromyography. However, previous research on the relationship of this decline with trunk extensor muscle endurance have shown inconsistent results. The decline of spectral content mainly reflects the decrease in average motor unit action potential conduction velocity (CV). We evaluated whether the rate of change in CV, as well as two approaches employing the change in spectral content, are related to trunk extensor muscle endurance. Fourteen healthy male participants without a low-back pain history performed a non-strictly controlled static forward trunk bending trial until exhaustion while standing. For 13 participants, physiologically plausible CV estimates were obtained from high-density surface electromyography bilaterally from T6 to L5. Laterally between L1 and L2, the linear rate of CV change was strongly correlated to endurance time (R2 = 0.79), whereas analyses involving the linear rate of change in spectral measures showed a lower (R2 = 0.38) or no correlation. For medial electrode locations, estimating CV and its relationship with endurance time was less successful, while the linear rate of change in spectral measures correlated moderately to endurance time (R2 = 0.44; R2 = 0.56). This study provides guidance on monitoring trunk extensor muscle fatigue development using electromyography.

Clinical utility of trans-sacral magnetic stimulation-evoked sphincter potentials and high-density electromyography in pelvic floor assessment: Technical evaluation.

Faecal incontinence is common and of multifactorial aetiologies, yet current diagnostic tools are unable to assess nerve and sphincter function objectively. We developed an anorectal high-density electromyography (HD-EMG) probe to evaluate motor-evoked potentials induced via trans-sacral magnetic stimulation (TSMS). Anorectal probes with an 8 × 8 array of electrodes spaced 1 cm apart were developed for recording HD-EMG of the external anal sphincter. These HD-EMG probes were used to map MEP amplitudes and latencies evoked via TSMS delivered through the Magstim Rapid2 (MagStim Company). Patients undergoing pelvic floor investigations were recruited for this IDEAL Stage 2a pilot study. Eight participants (median age 49 years; five female) were recruited. Methodological viability, safety and diagnostic workflow were established. The test was well tolerated with median discomfort scores ≤2.5/10, median pain scores ≤1/10 and no adverse events. Higher Faecal Incontinence Severity Index scores correlated with longer MEP latencies (r = 0.58, p < 0.001) and lower MEP amplitudes (r = -0.32, p = 0.046), as did St. Mark's Incontinence Scores with both MEP latencies (r = 0.49, p = 0.001) and MEP amplitudes (r = -0.47, p = 0.002). This HD-EMG probe in conjunction with TSMS presents a novel diagnostic tool for anorectal function assessment. Spatiotemporal assessment of magnetically stimulated MEPs correlated well with symptoms and offers a feasible, safe and patient-tolerable method of evaluating pudendal nerve and external anal sphincter function. Further clinical development and evaluation of these techniques is justified.

Novel Wearable HD-EMG Sensor With Shift-Robust Gesture Recognition Using Deep Learning.

In this work, we present a hardware-software solution to improve the robustness of hand gesture recognition to confounding factors in myoelectric control. The solution includes a novel, full-circumference, flexible, 64-channel high-density electromyography (HD-EMG) sensor called EMaGer. The stretchable, wearable sensor adapts to different forearm sizes while maintaining uniform electrode density around the limb. Leveraging this uniformity, we propose novel array barrel-shifting data augmentation (ABSDA) approach used with a convolutional neural network (CNN), and an anti-aliased CNN (AA-CNN), that provides shift invariance around the limb for improved classification robustness to electrode movement, forearm orientation, and inter-session variability. Signals are sampled from a 4×16 HD-EMG array of electrodes at a frequency of 1 kHz and 16-bit resolution. Using data from 12 non-amputated participants, the approach is tested in response to sensor rotation, forearm rotation, and inter-session scenarios. The proposed ABSDA-CNN method improves inter-session accuracy by 25.67% on average across users for 6 gesture classes compared to conventional CNN classification. A comparison with other devices shows that this benefit is enabled by the unique design of the EMaGer array. The AA-CNN yields improvements of up to 63.05% accuracy over non-augmented methods when tested with electrode displacements ranging from -45 ° to +45 ° around the limb. Overall, this article demonstrates the benefits of co-designing sensor systems, processing methods, and inference algorithms to leverage synergistic and interdependent properties to solve state-of-the-art problems.