EMG-Based Continuous and Simultaneous Estimation of Arm Kinematics in Able-Bodied Individuals and Stroke Survivors

Continuous Kalman Estimation Method for Finger Kinematics Tracking from Surface Electromyography

Deciphering hand motion intention from surface electromyography (sEMG) encounters challenges posed by the requisites of multiple degrees of freedom (DOFs) and adaptability. Unlike discrete action classification grounded in pattern recognition, the pursuit of continuous kinematics estimation is appreciated for its inherent naturalness and intuitiveness. However, prevailing estimation techniques contend with accuracy limitations and substantial computational demands. Kalman estimation technology, celebrated for its ease of implementation and real-time adaptability, finds extensive application across diverse domains. This study introduces a continuous Kalman estimation method, leveraging a system model with sEMG and joint angles as inputs and outputs. Facilitated by model parameter training methods, the approach deduces multiple DOF finger kinematics simultaneously. The method's efficacy is validated using a publicly accessible database, yielding a correlation coefficient (CC) of 0.73. With over 45,000 windows for training Kalman model parameters, the average computation time remains under 0.01 s. This pilot study amplifies its potential for further exploration and application within the realm of continuous finger motion estimation technology.

Continuous Motion Intention Prediction Using sEMG for Upper-Limb Rehabilitation: A Systematic Review of Model-Based and Model-Free Approaches

Upper limb functional impairments persisting after stroke significantly affect patients' quality of life. Precise adjustment of robotic assistance levels based on patients' motion intentions using sEMG signals is crucial for active rehabilitation. This paper systematically reviews studies on continuous prediction of upper limb single joints and multi-joint combinations motion intention using Model-Based (MB) and Model-Free (MF) approaches over the past decade, based on 186 relevant studies screened from six major electronic databases. The findings indicate ongoing challenges in terms of subject composition, algorithm robustness and generalization, and algorithm feasibility for practical applications. Moreover, it suggests integrating the strengths of both MB and MF approaches to improve existing algorithms. Therefore, future research should further explore personalized MB-MF combination methods incorporating deep learning, attention mechanisms, muscle synergy features, motor unit features, and closed-loop feedback to achieve precise, real-time, and long-duration prediction of multi-joint complex movements, while further refining the transfer learning strategy for rapid algorithm deployment across days and subjects. Overall, this review summarizes the current research status, significant findings, and challenges, aiming to inspire future research on predicting upper limb motion intentions based on sEMG.

EMG-Based Estimation of Lower Limb Joint Angles and Moments Using Long Short-Term Memory Network

One of the fundamental limitations in human biomechanics is that we cannot directly obtain joint moments during natural movements without affecting the motion. However, estimating these values is feasible with inverse dynamics computation by employing external force plates, which can cover only a small area of the plate. This work investigated the Long Short-Term Memory (LSTM) network for the kinetics and kinematics prediction of human lower limbs when performing different activities without using force plates after the learning. We measured surface electromyography (sEMG) signals from 14 lower extremities muscles to generate a 112-dimensional input vector from three sets of features: root mean square, mean absolute value, and sixth-order autoregressive model coefficient parameters for each muscle in the LSTM network. With the recorded experimental data from the motion capture system and the force plates, human motions were reconstructed in a biomechanical simulation created using OpenSim v4.1, from which the joint kinematics and kinetics from left and right knees and ankles were retrieved to serve as output for training the LSTM. The estimation results using the LSTM model deviated from labels with average R² scores (knee angle: 97.25%, knee moment: 94.9%, ankle angle: 91.44%, and ankle moment: 85.44%). These results demonstrate the feasibility of the joint angle and moment estimation based solely on sEMG signals for multiple daily activities without requiring force plates and a motion capture system once the LSTM model is trained.

A New Nonlinear Autoregressive Exogenous (NARX)-based Intra-spinal Stimulation Approach to Decode Brain Electrical Activity for Restoration of Leg Movement in Spinally-injured Rabbits

Introduction

This study aimed at investigating the stimulation by intra-spinal signals decoded from electrocorticography (ECoG) assessments to restore the movements of the leg in an animal model of spinal cord injury (SCI).

Methods

The present work is comprised of three steps. First, ECoG signals and the associated leg joint changes (hip, knee, and ankle) in sedated healthy rabbits were recorded in different trials. Second, an appropriate set of intra-spinal electric stimuli was discovered to restore natural leg movements, using the three leg joint movements under a fuzzy-controlled strategy in spinally-injured rabbits under anesthesia. Third, a nonlinear autoregressive exogenous (NARX) neural network model was developed to produce appropriate intra-spinal stimulation developed from decoded ECoG information. The model was able to correlate the ECoG signal data to the intra-spinal stimulation data and finally, induced desired leg movements. In this study, leg movements were also developed from offline ECoG signals (deciphered from rabbits that were not injured) as well as online ECoG data (extracted from the same rabbit after SCI induction).

Results

Based on our data, the correlation coefficient was 0.74±0.15 and the normalized root means square error of the brain-spine interface was 0.22±0.10.

Conclusion

Overall, we found that using NARX, appropriate information from ECoG recordings can be extracted and used for the generation of proper intra-spinal electric stimulations for restoration of natural leg movements lost due to SCI.

sEMG-Based Continuous Hand Action Prediction by Using Key State Transition and Model Pruning

Conventional classification of hand motions and continuous joint angle estimation based on sEMG have been widely studied in recent years. The classification task focuses on discrete motion recognition and shows poor real-time performance, while continuous joint angle estimation evaluates the real-time joint angles by the continuity of the limb. Few researchers have investigated continuous hand action prediction based on hand motion continuity. In our study, we propose the key state transition as a condition for continuous hand action prediction and simulate the prediction process using a sliding window with long-term memory. Firstly, the key state modeled by GMM-HMMs is set as the condition. Then, the sliding window is used to dynamically look for the key state transition. The prediction results are given while finding the key state transition. To extend continuous multigesture action prediction, we use model pruning to improve reusability. Eight subjects participated in the experiment, and the results show that the average accuracy of continuous two-hand actions is 97% with a 70 ms time delay, which is better than LSTM (94.15%, 308 ms) and GRU (93.83%, 300 ms). In supplementary experiments with continuous four-hand actions, over 85% prediction accuracy is achieved with an average time delay of 90 ms.

Periodic event-triggered sliding mode control for lower limb exoskeleton based on human-robot cooperation

This paper presents a periodic event-triggered sliding mode control (SMC) scheme based on human-robot cooperation for lower limb exoskeletons. Firstly, a Genetic Algorithm-Back propagation (GA-BP) neural network is proposed to estimate the motion intention of the wearer through electromyography (EMG) signals. Secondly, the periodic event-triggered SMC strategy based on tanh function is designed to ensure the asymptotic convergence of the exoskeleton system and save communication resources, where the detailed expressions of sampling period and control gain are designed. Finally, comparative simulation and experimental analysis is presented to verify the effectiveness of the proposed control method.

Estimation of Joint Angle From sEMG and Inertial Measurements Based on Deep Learning Approach

Continuous kinematics estimation from surface electromyography (sEMG) allows more natural and intuitive human-machine collaboration. Recent research has suggested the use of multimodal inputs (sEMG signals and inertial measurements) to improve estimation performance. This work focused on assessing the use of angular velocity in combination with myoelectric signals to simultaneously and continuously predict 12 joint angles in the hand. Estimation performance was evaluated for five functional and grasping movements in 20 subjects. The proposed method is based on convolutional and recurrent neural networks using transfer learning (TL). A novel aspect was the use of a pretrained deep network model from basic joint hand movements to learn new patterns present in functional motions. A comparison was carried out with the traditional method based solely on sEMG. Although the performance of the algorithm slightly improved with the use of the multimodal combination, both strategies had similar behavior. The results indicated a significant improvement for a single task: opening a bottle with a tripod grasp.

Disrupted cortico-peripheral interactions in motor disorders

Motor disorders may arise from neurological damage or diseases at different levels of the hierarchical motor control system and side-loops. Altered cortico-peripheral interactions might be essential characteristics indicating motor dysfunctions. By integrating cortical and peripheral responses, top-down and bottom-up cortico-peripheral coupling measures could provide new insights into the motor control and recovery process. This review first discusses the neural bases of cortico-peripheral interactions, and corticomuscular coupling and corticokinematic coupling measures are addressed. Subsequently, methodological efforts are summarized to enhance the modeling reliability of neural coupling measures, both linear and nonlinear approaches are introduced. The latest progress, limitations, and future directions are discussed. Finally, we emphasize clinical applications of cortico-peripheral interactions in different motor disorders, including stroke, neurodegenerative diseases, tremor, and other motor-related disorders. The modified interaction patterns and potential changes following rehabilitation interventions are illustrated. Altered coupling strength, modified coupling directionality, and reorganized cortico-peripheral activation patterns are pivotal attributes after motor dysfunction. More robust coupling estimation methodologies and combination with other neurophysiological modalities might more efficiently shed light on motor control and recovery mechanisms. Future studies with large sample sizes might be necessary to determine the reliabilities of cortico-peripheral interaction measures in clinical practice.

Intra-Subject and Inter-Subject Movement Variability Quantified with Muscle Synergies in Upper-Limb Reaching Movements

Quantifying movement variability is a crucial aspect for clinical and laboratory investigations in several contexts. However, very few studies have assessed, in detail, the intra-subject variability across movements and the inter-subject variability. Muscle synergies are a valuable method that can be used to assess such variability. In this study, we assess, in detail, intra-subject and inter-subject variability in a scenario based on a comprehensive dataset, including multiple repetitions of multi-directional reaching movements. The results show that muscle synergies are a valuable tool for quantifying variability at the muscle level and reveal that intra-subject variability is lower than inter-subject variability in synergy modules and related temporal coefficients, and both intra-subject and inter-subject similarity are higher than random synergy matching, confirming shared underlying control structures. The study deepens the available knowledge on muscle synergy-based motor function assessment and rehabilitation applications, discussing their applicability to real scenarios.

Continuous Multi-DoF Wrist Kinematics Estimation Based on a Human-Machine Interface With Electrical-Impedance-Tomography

This study proposed a multiple degree-of-freedom (DoF) continuous wrist angle estimation approach based on an electrical impedance tomography (EIT) interface. The interface can inspect the spatial information of deep muscles with a soft elastic fabric sensing band, extending the measurement scope of the existing muscle-signal-based sensors. The designed estimation algorithm first extracted the mutual correlation of the EIT regions with a kernel function, and second used a regularization procedure to select the optimal coefficients. We evaluated the method with different features and regression models on 12 healthy subjects when they performed six basic wrist joint motions. The average root-mean-square error of the 3-DoF estimation task was 7.62°, and the average R² was 0.92. The results are comparable to state-of-the-art with sEMG signals in multi-DoF tasks. Future endeavors will be paid in this new direction to get more promising results.

Continuous Myoelectric Prediction of Future Ankle Angle and Moment Across Ambulation Conditions and Their Transitions

A hallmark of human locomotion is that it continuously adapts to changes in the environment and predictively adjusts to changes in the terrain, both of which are major challenges to lower limb amputees due to the limitations in prostheses and control algorithms. Here, the ability of a single-network nonlinear autoregressive model to continuously predict future ankle kinematics and kinetics simultaneously across ambulation conditions using lower limb surface electromyography (EMG) signals was examined. Ankle plantarflexor and dorsiflexor EMG from ten healthy young adults were mapped to normal ranges of ankle angle and ankle moment during level overground walking, stair ascent, and stair descent, including transitions between terrains (i.e., transitions to/from staircase). Prediction performance was characterized as a function of the time between current EMG/angle/moment inputs and future angle/moment model predictions (prediction interval), the number of past EMG/angle/moment input values over time (sampling window), and the number of units in the network hidden layer that minimized error between experimentally measured values (targets) and model predictions of ankle angle and moment. Ankle angle and moment predictions were robust across ambulation conditions with root mean squared errors less than 1° and 0.04 Nm/kg, respectively, and cross-correlations (R²) greater than 0.99 for prediction intervals of 58 ms. Model predictions at critical points of trip-related fall risk fell within the variability of the ankle angle and moment targets (Benjamini-Hochberg adjusted p > 0.065). EMG contribution to ankle angle and moment predictions occurred consistently across ambulation conditions and model outputs. EMG signals had the greatest impact on noncyclic regions of gait such as double limb support, transitions between terrains, and around plantarflexion and moment peaks. The use of natural muscle activation patterns to continuously predict variations in normal gait and the model's predictive capabilities to counteract electromechanical inherent delays suggest that this approach could provide robust and intuitive user-driven real-time control of a wide variety of lower limb robotic devices, including active powered ankle-foot prostheses.

Gaussian Process Autoregression for Joint Angle Prediction Based on sEMG Signals

There is uncertainty in the neuromusculoskeletal system, and deterministic models cannot describe this significant presence of uncertainty, affecting the accuracy of model predictions. In this paper, a knee joint angle prediction model based on surface electromyography (sEMG) signals is proposed. To address the instability of EMG signals and the uncertainty of the neuromusculoskeletal system, a non-parametric probabilistic model is developed using a Gaussian process model combined with the physiological properties of muscle activation. Since the neuromusculoskeletal system is a dynamic system, the Gaussian process model is further combined with a non-linear autoregressive with eXogenous inputs (NARX) model to create a Gaussian process autoregression model. In this paper, the normalized root mean square error (NRMSE) and the correlation coefficient (CC) are compared between the joint angle prediction results of the Gaussian process autoregressive model prediction and the actual joint angle under three test scenarios: speed-dependent, multi-speed and speed-independent. The mean of NRMSE and the mean of CC for all test scenarios in the healthy subjects dataset and the hemiplegic patients dataset outperform the results of the Gaussian process model, with significant differences (p < 0.05 and p < 0.05, p < 0.05 and p < 0.05). From the perspective of uncertainty, a non-parametric probabilistic model for joint angle prediction is established by using Gaussian process autoregressive model to achieve accurate prediction of human movement.

EMG-Based 3D Hand Motor Intention Prediction for Information Transfer from Human to Robot

(1) Background: Three-dimensional (3-D) hand position is one of the kinematic parameters that can be inferred from Electromyography (EMG) signals. The inferred parameter is used as a communication channel in human–robot collaboration applications. Although its application from the perspective of rehabilitation and assistive technologies are widely studied, there are few papers on its application involving healthy subjects such as intelligent manufacturing and skill transfer. In this regard, for tasks associated with complex hand trajectories without the consideration of the degree of freedom (DOF), the prediction of 3-D hand position from EMG signal alone has not been addressed. (2) Objective: The primary aim of this study is to propose a model to predict human motor intention that can be used as information from human to robot. Therefore, the prediction of a 3-D hand position directly from the EMG signal for complex trajectories of hand movement, without the direct consideration of joint movements, is studied. In addition, the effects of slow and fast motions on the accuracy of the prediction model are analyzed. (3) Methods: This study used the EMG signal that is collected from the upper limb of healthy subjects, and the position signal of the hand while the subjects manipulate complex trajectories. We considered and analyzed two types of tasks with complex trajectories, each with quick and slow motions. A recurrent fuzzy neural network (RFNN) model was constructed to predict the 3-D position of the hand from the features of EMG signals alone. We used the Pearson correlation coefficient (CC) and normalized root mean square error (NRMSE) as performance metrics. (4) Results: We found that 3-D hand positions of the complex movement can be predicted with the mean performance of CC = 0.85 and NRMSE = 0.105. The 3-D hand position can be predicted well within a future time of 250 ms, from the EMG signal alone. Even though tasks performed under quick motion had a better prediction performance; the statistical difference in the accuracy of prediction between quick and slow motion was insignificant. Concerning the prediction model, we found that RFNN has a good performance in decoding for the time-varying system. (5) Conclusions: In this paper, irrespective of the speed of the motion, the 3-D hand position is predicted from the EMG signal alone. The proposed approach can be used in human–robot collaboration applications to enhance the natural interaction between a human and a robot.

A Nonlinear Latching Filter to Remove Jitter From Movement Estimates for Prostheses

Continuous movement intent decoders are critical for precise control of hand and wrist prostheses. Noise in biological signals (e.g., myoelectric or neural signals) can lead to undesirable jitter in the output of these types of decoders. A low-pass filter (LPF) at the output of the decoder effectively reduces jitter, but also substantially slows intended movements. This paper introduces an alternative, the latching filter (LF), a recursive, nonlinear filter that provides smoothing of small-amplitude jitter but allows quick changes to its output in response to large input changes. The performance of a Kalman filter (KF) decoder smoothed with an LF is compared with that of both an KF decoder without an additional smoother and a KF decoder smoothed with a LPF. These three algorithms were tested in real-time on target holding and target reaching tasks using surface electromyographic signals recorded from 5 non-amputee subjects, and intramuscular electromyographic and peripheral neural signals recorded from an amputee subject. When compared with the LPF, the LF provided a statistically significant improvement in amputee and non-amputee subjects' ability to hold the hand steady at requested positions and achieve movement goals faster. The KF decoder with LF provided a statistically significant improvement in all subjects' ability to hold the prosthetic hand steady, with only slightly lower speeds, when compared to the unsmoothed KF.

Development of a neural network based control algorithm for powered ankle prosthesis

Lower limb amputation is partial or complete removal of the limb due to disease, accident or trauma. Surface electromyograms (sEMG) of a large number of muscles and force sensors have been used to develop control algorithms for lower limb powered prostheses, but there are no commercial sEMG controlled prostheses available to date. Unlike ankle disarticulation, transtibial amputation yields less intact lower leg muscle mass. Therefore, minimizing the use of sEMG muscle sources utilized will make powered prosthesis controller economic, and limiting the use of specifically the lower leg muscles will make it flexible. Presently, we have used healthy population data to (1) test the feasibility of the neural network (NN) approach for developing a powered ankle prosthesis control algorithm that successfully predicts sagittal ankle angle and moment during walking using exclusively sEMG, and (2) rank all muscle combination variations according to their success to determine the economic and flexible NN's. sEMG amplitudes of five lower extremity muscles were used as inputs: the tibialis anterior (TA), medial gastrocnemius (MG), rectus femoris (RF), biceps femoris (BF) and gluteus maximus (GM). A time-delay feed-forward-multilayer-architecture NN algorithm was developed. Muscle combination variations were ranked using Pearson's correlation coefficient (r > 0.95 indicates successful correlations) and root-mean-square error between actual vs. estimated ankle position and moment. The trained NN TA + MG was successful (r_position = 0.952, r_moment = 0.997) whereas, TA + MG + BF (r_position = 0.981, r_moment = 0.996) and MG + BF + GM (r_position = 0.955, r_moment = 0.995) were distinguished as the economic and flexible variations, respectively. The algorithms developed should be trained and tested for data acquired from amputees in new studies.

Prediction of the performance of artificial neural networks in mapping sEMG to finger joint angles via signal pre-investigation techniques

The inputs to the outputs of nonlinear systems can be modeled using machine and deep learning approaches, among which artificial neural networks (ANNs) are a promising option. However, noisy signals affect ANN modeling negatively; hence, it is important to investigate these signals prior to the modeling. Herein, two customized and simple approaches, visual inspection and absolute correlation, are proposed to examine the relationship between the inputs and outputs of a nonlinear system. The system under consideration uses biosignals from surface electromyography as inputs and human finger joint angles as outputs, acquired from eight intact participants performing movements and grasping tasks in dynamic conditions. Furthermore, the results of these approaches are tested using the standard mutual information measure. Hence, the system dimensionality is reduced, and the ANN learning (convergence) is accelerated, where the most informative inputs are selected for the next phase. Subsequently, four ANN types, i.e., feedforward, cascade-forward, radial basis function, and generalized regression ANNs, are used to perform the modeling. Finally, the performance of the ANNs is compared with findings from the signal analysis. Results indicate a high level of consistency among all the aforementioned signal pre-analysis techniques from one side, and they also indicate that these techniques match the ANN performances from the other side. As an example, for a certain movement set, the ANN models resulted in the rotation estimation accuracy of the joints in the following descending order: carpometacarpal, metacarpophalangeal, proximal interphalangeal, and distal interphalangeal. This information has been indicated in the signal pre-analysis step. Therefore, this step is crucial in input–output variable selections prior to machine-/deep-learning-based modeling approaches.

sEMG-Based Trunk Compensation Detection in Rehabilitation Training

Stroke patients often use trunk to compensate for impaired upper limb motor function during upper limb rehabilitation training, which results in a reduced rehabilitation training effect. Detecting trunk compensations can improve the effect of rehabilitation training. This study investigates the feasibility of a surface electromyography-based trunk compensation detection (sEMG-bTCD) method. Five healthy subjects and nine stroke subjects with cognitive and comprehension skills were recruited to participate in the experiments. The sEMG signals from nine superficial trunk muscles were collected during three rehabilitation training tasks (reach-forward-back, reach-side-to-side, and reach-up-to-down motions) without compensation and with three common trunk compensations [lean-forward (LF), trunk rotation (TR), and shoulder elevation (SE)]. Preprocessing like filtering, active segment detection was performed and five time domain features (root mean square, variance, mean absolute value (MAV), waveform length, and the fourth order autoregressive model coefficient) were extracted from the collected sEMG signals. Excellent TCD performance was achieved in healthy participants by using support vector machine (SVM) classifier (LF: accuracy = 94.0%, AUC = 0.97, F1 = 0.94; TR: accuracy = 95.8%, AUC = 0.99, F1 = 0.96; SE: accuracy = 100.0%, AUC = 1.00, F1 = 1.00). By using SVM classifier, TCD performance in stroke participants was also obtained (LF: accuracy = 74.8%, AUC = 0.90, F1 = 0.73; TR: accuracy = 67.1%, AUC = 0.85, F1 = 0.71; SE: accuracy = 91.3%, AUC = 0.98, F1 = 0.90). Compared with the methods based on cameras or inertial sensors, better detection performance was obtained in both healthy and stroke participants. The results demonstrated the feasibility of the sEMG-bTCD method, and it helps to prompt the stroke patients to correct their incorrect posture, thereby improving the effectiveness of rehabilitation training.

EMG-Based Real-Time Linear-Nonlinear Cascade Regression Decoding of Shoulder, Elbow, and Wrist Movements in Able-Bodied Persons and Stroke Survivors

OBJECTIVE

This study aimed to decode shoulder, elbow and wrist dynamic movements continuously and simultaneously based on multi-channel surface electromyography signals, useful for electromyography controlled exoskeleton robots for upper-limb rehabilitation.

METHODS

Ten able-bodied subjects and ten stroke subjects were instructed to voluntarily move the shoulder, elbow and wrist joints back and forth in a horizontal plane with an exoskeleton robot. The shoulder, elbow and wrist movements and surface electromyography signals from six muscles crossing the joints were recorded. A set of three parallel linear-nonlinear cascade decoders was developed to continuously estimate the selected shoulder, elbow and wrist movements based on a generalized linear model using the anterior deltoid, posterior deltoid, biceps brachii, long head triceps brachii, flexor carpi radialis, and extensor carpi radialis muscle electromyography signals as the model inputs.

RESULTS

The decoder performed well for both healthy and stroke populations. As movement smoothness decreased, decoding performance decreased for the stroke population.

CONCLUSION

The proposed method is capable of simultaneously and continuously estimating multi-joint movements of the human arm in real-time by characterizing the nonlinear mappings between muscle activity and kinematic signals based on linear regression.

SIGNIFICANCE

This may prove useful in developing myoelectric controlled exoskeletons for motor rehabilitation of neurological disorders.

Design and effectiveness evaluation of mirror myoelectric interfaces: a novel method to restore movement in hemiplegic patients

The motor impairment occurring after a stroke is characterized by pathological muscle activation patterns or synergies. However, while robot-aided myoelectric interfaces have been proposed for stroke rehabilitation, they do not address this issue, which might result in inefficient interventions. Here, we present a novel paradigm that relies on the correction of the pathological muscle activity as a way to elicit rehabilitation, even in patients with complete paralysis. Previous studies demonstrated that there are no substantial inter-limb differences in the muscle synergy organization of healthy individuals. We propose building a subject-specific model of muscle activity from the healthy limb and mirroring it to use it as a learning tool for the patient to reproduce the same healthy myoelectric patterns on the paretic limb during functional task training. Here, we aim at understanding how this myoelectric model, which translates muscle activity into continuous movements of a 7-degree of freedom upper limb exoskeleton, could transfer between sessions, arms and tasks. The experiments with 8 healthy individuals and 2 chronic stroke patients proved the feasibility and effectiveness of such myoelectric interface. We anticipate the proposed method to become an efficient strategy for the correction of maladaptive muscle activity and the rehabilitation of stroke patients.