Neural Data-Driven Musculoskeletal Modeling for Personalized Neurorehabilitation Technologies

A calibrated EMG-informed neuromusculoskeletal model can estimate hip and knee joint contact forces in cycling better than static optimisation

Cycling is a popular competitive and recreational exercise and is recommended as safe to perform following hip or knee surgery. During cycling, joint contact forces (JCF) have been recorded in-vivo and estimated via neuromusculoskeletal models, but model estimates are yet to be validated. In this study, motion data, crank force, and electromyograms for a range of cadences (40 and 60 revolutions per minute (rpm)) and power outputs (25, 35, 50, 60, 79, 75, 85, 95, 120 W) were collected from 7 healthy people cycling on a powered stationary ergometer. A (1) calibrated electromyogram-informed neuromusculoskeletal model and an (2) uncalibrated model that utilised static optimisation were used to estimate hip and knee JCF. Hip and knee JCF estimates were compared against in-vivo measurements of hip and knee JCF from literature. Peak hip and knee JCF were overestimated by both electromyogram-informed and static optimisation solutions, however, the magnitude and gradients of JCF as a function of cadence and power estimated by the electromyogram-informed solution more closely matched in-vivo measurement than those computed by static optimisation. Similarly, the profile of knee JCF as a function of crank angle estimated by the electromyogram-informed solution more closely matched in-vivo knee JCF than the static optimisation solution. Results indicate electromyogram-informed modelling is a valid computational approach to estimate knee and hip biomechanics during standard seated ergometer cycling.

Biomechanical models in the lower-limb exoskeletons development: a review

Lower limb exoskeletons serve multiple purposes, like supporting and augmenting movement. Biomechanical models are practical tools to understand human movement, and motor control. This paper provides an overview of these models and a comprehensive review of the current applications of them in assistive device development. It also critically analyzes the existing literature to identify research gaps and suggest future directions. Biomechanical models can be broadly classified as conceptual and detailed models and can be used for the design, control, and assessment of exoskeletons. Also, these models can estimate unmeasurable or hard-to-measure variables, which is also useful within the aforementioned applications. We identified the validation of simulation studies and the enhancement of the accuracy and fidelity of biomechanical models as key future research areas for advancing the development of assistive devices. Additionally, we suggest using exoskeletons as a tool to validate and refine these models. We also emphasize the exploration of model-based design and control approaches for exoskeletons targeting pathological gait, and utilizing biomechanical models for diverse design objectives of exoskeletons. In addition, increasing the availability of open source resources accelerates the advancement of the exoskeleton and biomechanical models. Although biomechanical models are widely applied to improve movement assistance and rehabilitation, their full potential in developing human-compatible exoskeletons remains underexplored and requires further investigation. This review aims to reveal existing needs and cranks new perspectives for developing more effective exoskeletons based on biomechanical models.

Neural-driven activation of 3D muscle within a finite element framework: exploring applications in healthy and neurodegenerative simulations

This paper presents a novel computational framework for neural-driven finite element muscle models, with an application to amyotrophic lateral sclerosis (ALS). The multiscale neuromusculoskeletal (NMS) model incorporates physiologically accurate motor neurons, 3D muscle geometry, and muscle fiber recruitment. It successfully predicts healthy muscle force and tendon elongation and demonstrates a progressive decline in muscle force due to ALS, dropping from 203 N (healthy) to 155 N (120 days after ALS onset). This approach represents a preliminary step towards developing integrated neural and musculoskeletal simulations to enhance our understanding of neurodegenerative and neurodevelopmental conditions through predictive NMS models.

Neuro-Musculoskeletal Modeling for Online Estimation of Continuous Wrist Movements from Motor Unit Activities

Decoding movement intentions from motor unit (MU) activities remains an ongoing challenge, which restricts our comprehension of the intricate transition mechanism from microscopic neural drive to macroscopic movements. This study presents an innovative neuro-musculoskeletal (NMS) model driven by MU activities for online estimation of continuous wrist movements. The proposed model employs a physiological and comprehensive utilization of MU firings and waveforms, thus facilitating the localization of MUs to muscle-tendon units (MTU) as well as the computation of MU-specific neural excitation. Subsequently, the MU-specific neural excitation was integrated to form the MTU-specific neural excitation, which were then inputted into a musculoskeletal model to accomplish the joint angle estimation. To assess the effectiveness of this model, high-density surface electromyography and angular data were collected from the forearms of eight subjects during their performance of wrist flexion-extension task. Two pieces of 8×8 electrode arrays and a motion capture system were employed for data acquisition. Following offline model calibration with a global optimization algorithm, online angle estimation results demonstrated a significant superiority of the proposed model over the state-of-the-art NMS models (p < 0.05), yielding the lowest normalized root mean square error ( 0.10 ± 0.02 ) and the highest determination coefficient ( 0.87 ± 0.06 ). This study provides a novel idea for the decoding of joint movements from MU activities. The research findings hold the potential to advance the development of NMS models towards the control of multiple degrees of freedom, with promising applications in the fields of motor control, biomechanics, and neuro-rehabilitation engineering.

Integrating Upper-Limb Prostheses with the Human Body: Technology Advances, Readiness, and Roles in Human-Prosthesis Interaction

Significant advances in bionic prosthetics have occurred in the past two decades. The field's rapid expansion has yielded many exciting technologies that can enhance the physical, functional, and cognitive integration of a prosthetic limb with a human. We review advances in the engineering of prosthetic devices and their interfaces with the human nervous system, as well as various surgical techniques for altering human neuromusculoskeletal systems for seamless human-prosthesis integration. We discuss significant advancements in research and clinical translation, focusing on upper limbprosthetics since they heavily rely on user intent for daily operation, although many discussed technologies have been extended to lower limb prostheses as well. In addition, our review emphasizes the roles of advanced prosthetics technologies in complex interactions with humans and the technology readiness levels (TRLs) of individual research advances. Finally, we discuss current gaps and controversies in the field and point out future research directions, guided by TRLs.

A Novel Motor Unit-Driven Neuro-Musculoskeletal Model for Estimating Continuous Wrist Motion

Deciphering the intentions underlying movement based on motor unit (MU) activities poses an enduring challenge, thereby constraining our understanding of the intricate transition mechanism from microscopic neural commands to macroscopic movements. This investigation introduces an innovative MU-driven neuro-musculoskeletal (NMS) model, aimed at the continuous estimation of wrist movements. The presented model utilizes both MU firing and waveform information, incorporating a linear model and physiological twitch model for the computation of MU-specific neural excitations. Sequentially, these MU-specific neural excitations were amalgamated to generate muscle-tendon unit (MTU)-specific neural excitations, which serves as inputs for the subsequent musculoskeletal model to accomplish joint angle estimation. To evaluate the efficacy of this model, high-density surface electromyography data and angular information were procured from the forearms of four subjects engaged in wrist flexion-extension tasks. Data acquisition was performed with two high-density electrodes and a motion capture system. The angle estimation results revealed a marked superiority of the proposed model over two conventional NMS models, showcasing the lowest normalized root mean square error and the highest determination coefficient. This study propounds a groundbreaking approach for decoding joint movements from MU activities, holding the potential to propel the advancement of NMS models towards the control of multiple degrees of freedom.

EMG-Constrained and Ultrasound-Informed Muscle-Tendon Parameter Estimation in Post-Stroke Hemiparesis

Secondary morphological and mechanical property changes in the muscle-tendon unit at the ankle joint are often observed in post-stroke individuals. These changes may alter the force generation capacity and affect daily activities such as locomotion. This work aimed to estimate subject-specific muscle-tendon parameters in individuals after stroke by solving the muscle redundancy problem using direct collocation optimal control methods based on experimental electromyography (EMG) signals and measured muscle fiber length. Subject-specific muscle-tendon parameters of the gastrocnemius, soleus, and tibialis anterior were estimated in seven post-stroke individuals and seven healthy controls. We found that the maximum isometric force, tendon stiffness and optimal fiber length in the post-stroke group were considerably lower than in the control group. We also computed the root mean square error between estimated and experimental values of muscle excitation and fiber length. The musculoskeletal model with estimated subject-specific muscle tendon parameters (from the muscle redundancy solver), yielded better muscle excitation and fiber length estimations than did scaled generic parameters. Our findings also showed that the muscle redundancy solver can estimate muscle-tendon parameters that produce force behavior in better accordance with the experimentally-measured value. These muscle-tendon parameters in the post-stroke individuals were physiologically meaningful and may shed light on treatment and/or rehabilitation planning.

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.

The Simultaneous Model-Based Estimation of Joint, Muscle, and Tendon Stiffness is Highly Sensitive to the Tendon Force-Strain Relationship

OBJECTIVE

Accurate estimation of stiffness across anatomical levels (i.e., joint, muscle, and tendon) in vivo has long been a challenge in biomechanics. Recent advances in electromyography (EMG)-driven musculoskeletal modeling have allowed the non-invasive estimation of stiffness during dynamic joint rotations. Nevertheless, validation has been limited to the joint level due to a lack of simultaneous in vivo experimental measurements of muscle and tendon stiffness.

METHODS

With a focus on the triceps surae, we employed a novel perturbation-based experimental technique informed by dynamometry and ultrasonography to derive reference stiffness at the joint, muscle, and tendon levels simultaneously. Here, we propose a new EMG-driven model-based approach that does not require external joint perturbation, nor ultrasonography, to estimate multi-level stiffness. We present a novel set of closed-form equations that enables the person-specific tuning of musculoskeletal parameters dictating biological stiffness, including passive force-length relationships in modeled muscles and tendons.

RESULTS

Calibrated EMG-driven musculoskeletal models estimated the reference data with average normalized root-mean-square error ≈ 20%. Moreover, only when calibrated tendons were approximately four times more compliant than typically modeled, our approach could estimate multi-level reference stiffness.

CONCLUSION

EMG-driven musculoskeletal models can be calibrated on a larger set of reference data to provide more realistic values for the biomechanical variables across multiple anatomical levels. Moreover, the tendon models that are typically used in musculoskeletal modeling are too stiff.

SIGNIFICANCE

Calibrated musculoskeletal models informed by experimental measurements give access to an augmented range of biomechanical variables that might not be easily measured with sensors alone.

Gait Intention Prediction Using a Lower-Limb Musculoskeletal Model and Long Short-Term Memory Neural Networks

The prediction of gait motion intention is essential for achieving intuitive control of assistive devices and diagnosing gait disorders. To reduce the cost associated with using multimodal signals and signal processing, we proposed a novel method that integrates machine learning with musculoskeletal modelling techniques for the prediction of time-series joint angles, using only kinematic signals. Additionally, we hypothesised that a stacked long short-term memory (LSTM) neural network architecture can perform the task without relying on any ahead-of-motion features typically provided by electromyography signals. Optical cameras and inertial measurement unit (IMU) sensors were used to track level gait kinematics. Joint angles were modelled using the musculoskeletal model. The optimal LSTM architecture in fulfilling the prediction task was determined. Joint angle predictions were performed for joints on the sagittal plane, benefiting from joint angle modelling using signals from optical cameras and IMU sensors. Our proposed method predicted the upcoming joint angles in the prediction time of 10 ms, with an averaged root mean square error of 5.3° and a coefficient of determination of 0.81. Moreover, in support of our hypothesis, the recurrent stacked LSTM network demonstrated its ability to predict intended motion accurately and efficiently in gait, outperforming two other neural network architectures: a feedforward MLP and a hybrid LSTM-MLP. The method paves the way for the development of a cost-effective, single-modal control system for assistive devices in gait rehabilitation.

Design and assessment of a reconfigurable behavioral assistive robot: a pilot study

Introduction

For patients with functional motor disorders of the lower limbs due to brain damage or accidental injury, restoring the ability to stand and walk plays an important role in clinical rehabilitation. Lower limb exoskeleton robots generally require patients to convert themselves to a standing position for use, while being a wearable device with limited movement distance.

Methods

This paper proposes a reconfigurable behavioral assistive robot that integrates the functions of an exoskeleton robot and an assistive standing wheelchair through a novel mechanism. The new mechanism is based on a four-bar linkage, and through simple and stable conformal transformations, the robot can switch between exoskeleton state, sit-to-stand support state, and wheelchair state. This enables the robot to achieve the functions of assisted walking, assisted standing up, supported standing and wheelchair mobility, respectively, thereby meeting the daily activity needs of sit-to-stand transitions and gait training. The configuration transformation module controls seamless switching between different configurations through an industrial computer. Experimental protocols have been developed for wearable testing of robotic prototypes not only for healthy subjects but also for simulated hemiplegic patients.

Results

The experimental results indicate that the gait tracking effect during robot-assisted walking is satisfactory, and there are no sudden speed changes during the assisted standing up process, providing smooth support to the wearer. Meanwhile, the activation of the main force-generating muscles of the legs and the plantar pressure decreases significantly in healthy subjects and simulated hemiplegic patients wearing the robot for assisted walking and assisted standing-up compared to the situation when the robot is not worn.

Discussion

These experimental findings demonstrate that the reconfigurable behavioral assistive robot prototype of this study is effective, reducing the muscular burden on the wearer during walking and standing up, and provide effective support for the subject's body. The experimental results objectively and comprehensively showcase the effectiveness and potential of the reconfigurable behavioral assistive robot in the realms of behavioral assistance and rehabilitation training.

Electromyography-driven musculoskeletal models with time-varying fatigue dynamics improve lumbosacral joint moments during lifting

Muscle fatigue is prevalent across different aspects of daily life. Tracking muscle fatigue is useful to understand muscle overuse and possible risk of injury leading to musculoskeletal disorders. Current fatigue models are not suitable for real-world settings as they are either validated using simulations or non-functional tasks. Moreover, models that capture the changes to muscle activity due to fatigue either assume a linear relationship between muscle activity and muscle force or utilize a simple muscle model. Personalised electromygraphy (EMG)-driven musculoskeletal models (pEMS) offer person-specific approaches to model muscle and joint kinetics during a wide repertoire of daily life tasks. These models utilize EMG, thus capturing central fatigue-dependent changes in multi-muscle bio-electrical activity. However, the peripheral muscle force decay is missing in these models. Thus, we studied the influence of fatigue on a large scale pEMS of the trunk. Eleven healthy participants performed functional asymmetric lifting task. Average peak body-weight normalized lumbosacral moments (BW-LM) were estimated to be 2.55 ± 0.26 Nm/kg by reference inverse dynamics. After complete exhaustion of the lower back, the pEMS overestimated the peak BW-LM by 0.64 ± 0.37 Nm/kg. Then, we developed a time-varying muscle force decay model resulting in a time-varying pEMS (t-pEMS). This reduced the difference between BW-LM estimated by the t-pEMS and reference to 0.49 ± 0.14 Nm/kg. We also showed that five fatiguing contractions are sufficient to calibrate the t-pEMS. Thus, this study presents a person and muscle specific model to track fatigue during functional tasks.

Learning to Walk With Deep Reinforcement Learning: Forward Dynamic Simulation of a Physics-Based Musculoskeletal Model of an Osseointegrated Transfemoral Amputee

This paper leverages the OpenSim physics-based simulation environment for the forward dynamic simulation of an osseointegrated transfemoral amputee musculoskeletal model, wearing a generic prosthesis. A deep reinforcement learning architecture, which combines the proximal policy optimization algorithm with imitation learning, is designed to enable the model to walk by using three different observation states. The first is a complete state that includes the agent's kinematics, ground reaction forces, and muscle data; the second is a reduced state that only includes the kinematics and ground reaction forces; the third is an augmented state that combines the kinematics and ground reaction forces with a prediction of the muscle data generated by a fully-connected feed-forward neural network. The empirical results demonstrate that the model trained with the augmented observation state can achieve walking patterns with rewards and gait symmetry ratings comparable to those of the model trained with the complete observation state, while there are no symmetric walking patterns when using the reduced observation state. This paper shows the importance of including muscle data in a deep reinforcement learning architecture for the forward dynamic simulation of musculoskeletal models of transfemoral amputees.

Multi-Action Knee Contact Force Prediction by Domain Adaptation

Most recent musculoskeletal dynamics estimation methods are designed for predefined actions, such as gait, and don't generalize to various tasks. In this work, we address the problem of estimating internal biomechanical forces during more than one actions by introducing unsupervised domain adaptation into a deep learning model. More specifically, we developed a Bidirectional Long Short-Term Memory network for knee contact force prediction, enhanced with correlation alignment layers, in order to minimize the domain shift between kinematic data from different actions. Furthermore, we used the novel Neural State Machine (NSM) as a simulation platform to test and visualize our model predictions in a wide range of trajectories adapted to different 3D scene geometries in real-time. We conducted multiple experiments, including comparison with previous models, model alignment across action classes and real-to-synthetic data alignment. The results showed that the proposed deep learning architecture with domain adaptation performs better than the benchmark in terms of NRMSE and t-test. Overall, our method is capable of predicting knee contact forces for more than one action classes using a single architecture and thereby opens the path for estimating internal forces for intermediate actions, while the knowledge of the hidden state of motion may be used to support personalized rehabilitation. Moreover, our model can be easily integrated into any human motion simulation environment, which shows its potential in enabling biomechanical analysis in an automated and computationally efficient way.

The History and Future of Neuromusculoskeletal Biomechanics

The Executive Council of the International Society of Biomechanics has initiated and overseen the commemorations of the Society's 50th Anniversary in 2023. This included multiple series of lectures at the ninth World Congress of Biomechanics in 2022 and XXIXth Congress of the International Society of Biomechanics in 2023, all linked to special issues of International Society of Biomechanics' affiliated journals. This special issue of the Journal of Applied Biomechanics is dedicated to the biomechanics of the neuromusculoskeletal system. The reader is encouraged to explore this special issue which comprises 6 papers exploring the current state-of the-art, and future directions and roles for neuromusculoskeletal biomechanics. This editorial presents a very brief history of the science of the neuromusculoskeletal system's 4 main components: the central nervous system, musculotendon units, the musculoskeletal system, and joints, and how they biomechanically integrate to enable an understanding of the generation and control of human movement. This also entails a quick exploration of contemporary neuromusculoskeletal biomechanics and its future with new fields of application.

Tapping Into Skeletal Muscle Biomechanics for Design and Control of Lower Limb Exoskeletons: A Narrative Review

Lower limb exoskeletons and exosuits ("exos") are traditionally designed with a strong focus on mechatronics and actuation, whereas the "human side" is often disregarded or minimally modeled. Muscle biomechanics principles and skeletal muscle response to robot-delivered loads should be incorporated in design/control of exos. In this narrative review, we summarize the advances in literature with respect to the fusion of muscle biomechanics and lower limb exoskeletons. We report methods to measure muscle biomechanics directly and indirectly and summarize the studies that have incorporated muscle measures for improved design and control of intuitive lower limb exos. Finally, we delve into articles that have studied how the human-exo interaction influences muscle biomechanics during locomotion. To support neurorehabilitation and facilitate everyday use of wearable assistive technologies, we believe that future studies should investigate and predict how exoskeleton assistance strategies would structurally remodel skeletal muscle over time. Real-time mapping of the neuromechanical origin and generation of muscle force resulting in joint torques should be combined with musculoskeletal models to address time-varying parameters such as adaptation to exos and fatigue. Development of smarter predictive controllers that steer rather than assist biological components could result in a synchronized human-machine system that optimizes the biological and electromechanical performance of the combined system.

Estimation of Joint Torque by EMG-Driven Neuromusculoskeletal Models and LSTM Networks

Accurately predicting joint torque using wearable sensors is crucial for designing assist-as-needed exoskeleton controllers to assist muscle-generated torque and ensure successful task performance. In this paper, we estimated ankle dorsiflexion/plantarflexion, knee flexion/extension, hip flexion/extension, and hip abduction/adduction torques from electromyography (EMG) and kinematics during daily activities using neuromusculoskeletal (NMS) models and long short-term memory (LSTM) networks. The joint torque ground truth for model calibrating and training was obtained through inverse dynamics of captured motion data. A cluster approach that grouped movements based on characteristic similarity was implemented, and its ability to improve the estimation accuracy of both NMS and LSTM models was evaluated. We compared torque estimation accuracy of NMS and LSTM models in three cases: Pooled, Individual, and Clustered models. Pooled models used data from all 10 movements to calibrate or train one model, Individual models used data from each individual movement, and Clustered models used data from each cluster. Individual, Clustered and Pooled LSTM models all had relatively high joint torque estimation accuracy. Individual and Clustered NMS models had similarly good estimation performance whereas the Pooled model may be too generic to satisfy all movement patterns. While the cluster approach improved the estimation accuracy in NMS models in some movements, it made relatively little difference in the LSTM neural networks, which already had high estimation accuracy. Our study provides practical implications for designing assist-as-needed exoskeleton controllers by offering guidelines for selecting the appropriate model for different scenarios, and has potential to enhance the functionality of wearable exoskeletons and improve rehabilitation and assistance for individuals with motor disorders.

EMG-Informed Neuromusculoskeletal Modelling Estimates Muscle Forces and Joint Moments During Electrical Stimulation

This study implemented an electromyogram (EMG)-informed neuromusculoskeletal (NMS) model evaluating the volitional contributions to muscle forces and joint moments during functional electrical stimulation (FES). The NMS model was calibrated using motion and EMG (biceps brachii and triceps brachii) data recorded from able-bodied participants (n=3) performing weighted elbow flexion and extension cycling movements while equipped with an EMG-controlled closed-loop FES system. Models were executed using three computational approaches (i) EMG-driven, (ii) EMG-hybrid and (iii) EMG-assisted to estimate muscle forces and joint moments. Both EMG-hybrid and EMG-assisted modes were able estimate the elbow moment (root mean squared error and coefficient of determination), but the EMG-hybrid method also enabled quantifying the volitional contributions to muscle forces and elbow moments during FES. The proposed modelling method allows for assessing volitional contributions of patients to muscle force during FES rehabilitation, and could be used as biomarkers of recovery, biofeedback, and for real-time control of combined FES and robotic systems.

Neuromusculoskeletal model-informed machine learning-based control of a knee exoskeleton with uncertainties quantification

Introduction

Research interest in exoskeleton assistance strategies that incorporate the user's torque capacity is growing rapidly. However, the predicted torque capacity from users often includes uncertainty from various sources, which can have a significant impact on the safety of the exoskeleton-user interface.

Methods

To address this challenge, this paper proposes an adaptive control framework for a knee exoskeleton that uses muscle electromyography (EMG) signals and joint kinematics. The framework predicted the user's knee flexion/extension torque with confidence bounds to quantify the uncertainty based on a neuromusculoskeletal (NMS) solver-informed Bayesian Neural Network (NMS-BNN). The predicted torque, with a specified confidence level, controlled the assistive torque provided by the exoskeleton through a TCP/IP stream. The performance of the NMS-BNN model was also compared to that of the Gaussian process (NMS-GP) model.

Results

Our findings showed that both the NMS-BNN and NMS-GP models accurately predicted knee joint torque with low error, surpassing traditional NMS models. High uncertainties were observed at the beginning of each movement, and at terminal stance and terminal swing in self-selected speed walking in both NMS-BNN and NMS-GP models. The knee exoskeleton provided the desired assistive torque with a low error, although lower torque was observed during terminal stance of fast walking compared to self-selected walking speed.

Discussion

The framework developed in this study was able to predict knee flexion/extension torque with quantifiable uncertainty and to provide adaptive assistive torque to the user. This holds significant potential for the development of exoskeletons that provide assistance as needed, with a focus on the safety of the exoskeleton-user interface.

A Review of Myoelectric Control for Prosthetic Hand Manipulation

Myoelectric control for prosthetic hands is an important topic in the field of rehabilitation. Intuitive and intelligent myoelectric control can help amputees to regain upper limb function. However, current research efforts are primarily focused on developing rich myoelectric classifiers and biomimetic control methods, limiting prosthetic hand manipulation to simple grasping and releasing tasks, while rarely exploring complex daily tasks. In this article, we conduct a systematic review of recent achievements in two areas, namely, intention recognition research and control strategy research. Specifically, we focus on advanced methods for motion intention types, discrete motion classification, continuous motion estimation, unidirectional control, feedback control, and shared control. In addition, based on the above review, we analyze the challenges and opportunities for research directions of functionality-augmented prosthetic hands and user burden reduction, which can help overcome the limitations of current myoelectric control research and provide development prospects for future research.

BioMAT: An Open-Source Biomechanics Multi-Activity Transformer for Joint Kinematic Predictions Using Wearable Sensors

Through wearable sensors and deep learning techniques, biomechanical analysis can reach beyond the lab for clinical and sporting applications. Transformers, a class of recent deep learning models, have become widely used in state-of-the-art artificial intelligence research due to their superior performance in various natural language processing and computer vision tasks. The performance of transformer models has not yet been investigated in biomechanics applications. In this study, we introduce a Biomechanical Multi-activity Transformer-based model, BioMAT, for the estimation of joint kinematics from streaming signals of multiple inertia measurement units (IMUs) using a publicly available dataset. This dataset includes IMU signals and the corresponding sagittal plane kinematics of the hip, knee, and ankle joints during multiple activities of daily living. We evaluated the model's performance and generalizability and compared it against a convolutional neural network long short-term model, a bidirectional long short-term model, and multi-linear regression across different ambulation tasks including level ground walking (LW), ramp ascent (RA), ramp descent (RD), stair ascent (SA), and stair descent (SD). To investigate the effect of different activity datasets on prediction accuracy, we compared the performance of a universal model trained on all activities against task-specific models trained on individual tasks. When the models were tested on three unseen subjects' data, BioMAT outperformed the benchmark models with an average root mean square error (RMSE) of 5.5 ± 0.5°, and normalized RMSE of 6.8 ± 0.3° across all three joints and all activities. A unified BioMAT model demonstrated superior performance compared to individual task-specific models across four of five activities. The RMSE values from the universal model for LW, RA, RD, SA, and SD activities were 5.0 ± 1.5°, 6.2 ± 1.1°, 5.8 ± 1.1°, 5.3 ± 1.6°, and 5.2 ± 0.7° while these values for task-specific models were, 5.3 ± 2.1°, 6.7 ± 2.0°, 6.9 ± 2.2°, 4.9 ± 1.4°, and 5.6 ± 1.3°, respectively. Overall, BioMAT accurately estimated joint kinematics relative to previous machine learning algorithms across different activities directly from the sequence of IMUs signals instead of time-normalized gait cycle data.

Real-time myoelectric control of wrist/hand motion in Duchenne muscular dystrophy: A case study

Introduction: Duchenne muscular dystrophy (DMD) is a genetic disorder that induces progressive muscular degeneration. Currently, the increase in DMD individuals' life expectancy is not being matched by an increase in quality of life. The functioning of the hand and wrist is central for performing daily activities and for providing a higher degree of independence. Active exoskeletons can assist this functioning but require the accurate decoding of the users' motor intention. These methods have, however, never been systematically analyzed in the context of DMD. Methods: This case study evaluated direct control (DC) and pattern recognition (PR), combined with an admittance model. This enabled customization of myoelectric controllers to one DMD individual and to a control population of ten healthy participants during a target-reaching task in 1- and 2- degrees of freedom (DOF). We quantified real-time myocontrol performance using target reaching times and compared the differences between the healthy individuals and the DMD individual. Results and Discussion: Our findings suggest that despite the muscle tissue degeneration, the myocontrol performance of the DMD individual was comparable to that of the healthy individuals in both DOFs and with both control approaches. It was also evident that PR control performed better for the 2-DOF tasks for both DMD and healthy participants, while DC performed better for the 1-DOF tasks. The insights gained from this study can lead to further developments for the intuitive multi-DOF myoelectric control of active hand exoskeletons for individuals with DMD.

Toward higher-performance bionic limbs for wider clinical use

Most prosthetic limbs can autonomously move with dexterity, yet they are not perceived by the user as belonging to their own body. Robotic limbs can convey information about the environment with higher precision than biological limbs, but their actual performance is substantially limited by current technologies for the interfacing of the robotic devices with the body and for transferring motor and sensory information bidirectionally between the prosthesis and the user. In this Perspective, we argue that direct skeletal attachment of bionic devices via osseointegration, the amplification of neural signals by targeted muscle innervation, improved prosthesis control via implanted muscle sensors and advanced algorithms, and the provision of sensory feedback by means of electrodes implanted in peripheral nerves, should all be leveraged towards the creation of a new generation of high-performance bionic limbs. These technologies have been clinically tested in humans, and alongside mechanical redesigns and adequate rehabilitation training should facilitate the wider clinical use of bionic limbs.

Symbiotic electroneural and musculoskeletal framework to encode proprioception via neurostimulation: ProprioStim

Peripheral nerve stimulation in amputees achieved the restoration of touch, but not proprioception, which is critical in locomotion. A plausible reason is the lack of means to artificially replicate the complex activity of proprioceptors. To uncover this, we coupled neuromuscular models from ten subjects and nerve histologies from two implanted amputees to develop ProprioStim: a framework to encode proprioception by electrical evoking neural activity in close agreement with natural proprioceptive activity. We demonstrated its feasibility through non-invasive stimulation on seven healthy subjects comparing it with standard linear charge encoding. Results showed that ProprioStim multichannel stimulation was felt more natural, and hold promises for increasing accuracy in knee angle tracking, especially in future implantable solutions. Additionally, we quantified the importance of realistic 3D-nerve models against extruded models previously adopted for further design and validation of novel neurostimulation encoding strategies. ProprioStim provides clear guidelines for the development of neurostimulation policies restoring natural proprioception.

Benchmarking the Effects on Human-Exoskeleton Interaction of Trajectory, Admittance and EMG-Triggered Exoskeleton Movement Control

Nowadays, robotic technology for gait training is becoming a common tool in rehabilitation hospitals. However, its effectiveness is still controversial. Traditional control strategies do not adequately integrate human intention and interaction and little is known regarding the impact of exoskeleton control strategies on muscle coordination, physical effort, and user acceptance. In this article, we benchmarked three types of exoskeleton control strategies in a sample of seven healthy volunteers: trajectory assistance (TC), compliant assistance (AC), and compliant assistance with EMG-Onset stepping control (OC), which allows the user to decide when to take a step during the walking cycle. This exploratory study was conducted within the EUROBENCH project facility. Experimental procedures and data analysis were conducted following EUROBENCH's protocols. Specifically, exoskeleton kinematics, muscle activation, heart and breathing rates, skin conductance, as well as user-perceived effort were analyzed. Our results show that the OC controller showed robust performance in detecting stepping intention even using a corrupt EMG acquisition channel. The AC and OC controllers resulted in similar kinematic alterations compared to the TC controller. Muscle synergies remained similar to the synergies found in the literature, although some changes in muscle contribution were found, as well as an overall increase in agonist-antagonist co-contraction. The OC condition led to the decreased mean duration of activation of synergies. These differences were not reflected in the overall physiological impact of walking or subjective perception. We conclude that, although the AC and OC walking conditions allowed the users to modulate their walking pattern, the application of these two controllers did not translate into significant changes in the overall physiological cost of walking nor the perceived experience of use. Nonetheless, results suggest that both AC and OC controllers are potentially interesting approaches that can be explored as gait rehabilitation tools. Furthermore, the INTENTION project is, to our knowledge, the first study to benchmark the effects on human-exoskeleton interaction of three different exoskeleton controllers, including a new EMG-based controller designed by us and never tested in previous studies, which has made it possible to provide valuable third-party feedback on the use of the EUROBENCH facility and testbed, enriching the apprenticeship of the project consortium and contributing to the scientific community.

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.

Multi-Day EMG-Based Knee Joint Torque Estimation Using Hybrid Neuromusculoskeletal Modelling and Convolutional Neural Networks

Proportional control using surface electromyography (EMG) enables more intuitive control of a transfemoral prosthesis. However, EMG is a noisy signal which can vary over time, giving rise to the question what approach for knee torque estimation is most suitable for multi-day control. In this study we compared three different modelling frameworks to estimate knee torque in non-weight-bearing situations. The first model contained a convolutional neural network (CNN) which mapped EMG to knee torque directly. The second used a neuromusculoskeletal model (NMS) which used EMG, muscle tendon unit lengths and moment arms to compute knee torque. The third model (Hybrid) used a CNN to map EMG to specific muscle activation, which was used together with NMS components to compute knee torque. Multi-day measurements were conducted on ten able-bodied participants who performed non-weight bearing activities. CNN had the best performance in general and on each day (Normalized Root Mean Squared Error (NRMSE) 9.2 ± 4.4%). The Hybrid model (NRMSE 12.4 ± 3.4%) was able to outperform NMS (NRMSE 14.3 ± 4.2%). The NMS model showed no significant difference between measurement days. The CNN model and Hybrid models had significant performance differences between the first day and all other days. CNNs are suited for multi-day torque estimation in terms of error rate, outperforming the other two model types. NMS was the only model type which was robust over all days. This study investigated the behavior of three model types over multiple days, giving insight in the most suited modelling approach for multi-day torque estimation to be used in prosthetic control.

Lower-limb Joint Torque Prediction using LSTM Neural Networks and Transfer Learning

Estimation of joint torque during movement provides important information in several settings, such as effect of athletes' training or of a medical intervention, or analysis of the remaining muscle strength in a wearer of an assistive device. The ability to estimate joint torque during daily activities using wearable sensors is increasingly relevant in such settings. In this study, lower limb joint torques during ten daily activities were predicted by long short-term memory (LSTM) neural networks and transfer learning. LSTM models were trained with muscle electromyography signals and lower limb joint angles. Hip flexion/extension, hip abduction/adduction, knee flexion/extension and ankle dorsiflexion/plantarflexion torques were predicted. The LSTM models' performance in predicting torque was investigated in both intra-subject and inter-subject scenarios. Each scenario was further divided into intra-task and inter-task tests. We observed that LSTM models could predict lower limb joint torques during various activities accurately with relatively low error (root mean square error ≤ 0.14 Nm/kg, normalized root mean square error ≤8.7%) either through a uniform model or through ten separate models in intra-subject tests. Furthermore, a transfer learning technique was adopted in the inter-task and inter-subject tests to further improve the generalizability of LSTM models by pre-training a model on multiple subjects and/or tasks and transferring the learned knowledge to a target task/subject. Particularly in the inter-subject tests, we could predict joint torques accurately in several movements after training from only a few movements from new subjects.

Adaptation Strategies for Personalized Gait Neuroprosthetics

Personalization of gait neuroprosthetics is paramount to ensure their efficacy for users, who experience severe limitations in mobility without an assistive device. Our goal is to develop assistive devices that collaborate with and are tailored to their users, while allowing them to use as much of their existing capabilities as possible. Currently, personalization of devices is challenging, and technological advances are required to achieve this goal. Therefore, this paper presents an overview of challenges and research directions regarding an interface with the peripheral nervous system, an interface with the central nervous system, and the requirements of interface computing architectures. The interface should be modular and adaptable, such that it can provide assistance where it is needed. Novel data processing technology should be developed to allow for real-time processing while accounting for signal variations in the human. Personalized biomechanical models and simulation techniques should be developed to predict assisted walking motions and interactions between the user and the device. Furthermore, the advantages of interfacing with both the brain and the spinal cord or the periphery should be further explored. Technological advances of interface computing architecture should focus on learning on the chip to achieve further personalization. Furthermore, energy consumption should be low to allow for longer use of the neuroprosthesis. In-memory processing combined with resistive random access memory is a promising technology for both. This paper discusses the aforementioned aspects to highlight new directions for future research in gait neuroprosthetics.

Integration of neural architecture within a finite element framework for improved neuromusculoskeletal modeling

Neuromusculoskeletal (NMS) models can aid in studying the impacts of the nervous and musculoskeletal systems on one another. These computational models facilitate studies investigating mechanisms and treatment of musculoskeletal and neurodegenerative conditions. In this study, we present a predictive NMS model that uses an embedded neural architecture within a finite element (FE) framework to simulate muscle activation. A previously developed neuromuscular model of a motor neuron was embedded into a simple FE musculoskeletal model. Input stimulation profiles from literature were simulated in the FE NMS model to verify effective integration of the software platforms. Motor unit recruitment and rate coding capabilities of the model were evaluated. The integrated model reproduced previously published output muscle forces with an average error of 0.0435 N. The integrated model effectively demonstrated motor unit recruitment and rate coding in the physiological range based upon motor unit discharge rates and muscle force output. The combined capability of a predictive NMS model within a FE framework can aid in improving our understanding of how the nervous and musculoskeletal systems work together. While this study focused on a simple FE application, the framework presented here easily accommodates increased complexity in the neuromuscular model, the FE simulation, or both.

Design and Validation of a Sensor Fault-Tolerant Module for Real-Time High-Density EMG Pattern Recognition

With the advancements in electronics technology, high-density (HD) EMG sensing systems have become available and have been investigated for their feasibility and performance in neural-machine interface (NMI) applications. Comparing to the traditional single channel-based targeted muscle sensing method, HD EMG sensing performs a sampling of the electrical activity over a larger surface area and has the promise of 1) providing richer neural information from one temporal and two spatial dimensions and 2) ease of wear in real life without the need of anatomically targeted electrode placement. To use HD EMG in real-time NMI applications, challenges including high computational burden and unreliability of EMG recordings over time need to be addressed. This paper presented an HD EMG PR based NMI which seamlessly integrates HD EMG PR with a Sensor Fault-Tolerant Module (SFTM) which aimed to provide robust PR in real time. Experimental results showed that the SFTM was able to recover the PR accuracies by 6%-22% from disturbances including contact artifacts and loose contacts. A Python-based implementation of the proposed HD EMG SFTM was developed and was demonstrated to be computationally efficient for real-time performance. These results have demonstrated the feasibility of a robust real-time HD EMG PR-based NMI.

Identification of Motor Unit Twitch Properties in the Intact Human In Vivo

Restoring natural motor function in neurologically injured individuals is challenging, largely due to the lack of personalization in current neurorehabilitation technologies. Signal-driven neuro-musculoskeletal models may offer a novel paradigm for devising novel closed-loop rehabilitation strategies according to an individual's physiology. However, current modelling techniques are constrained to bipolar electromyography (EMG), thereby lacking the resolution necessary to extract the activity of individual motor units (MUs) in vivo. In this work, we decoded MU spike trains from high-density (HD)-EMG to obtain relevant neural properties across multiple isometric plantar-dorsiflexion tasks. Then, we sampled MU statistical distributions and used them to reproduce MU specific activation profiles. Results showed bimodal distributions which may correspond to slow and fast MU populations. The estimated activation profiles showed a high degree of similarity to the reference torque (R²>0.8) across the recorded muscles. This suggests that the estimation of MU twitch properties is a crucial step for the translation of neural information into muscle force.Clinical Relevance- This work has multiple implications for understanding the underlying mechanism of motor impairment and for developing closed-loop strategies for modulating alpha motor circuitries in neurologically injured individuals.

Design and Assist-as-Needed Control of Flexible Elbow Exoskeleton Actuated by Nonlinear Series Elastic Cable Driven Mechanism

Exoskeletons can assist the daily life activities of the elderly with weakened muscle strength, but traditional rigid exoskeletons bring parasitic torque to the human joints and easily disturbs the natural movement of the wearer’s upper limbs. Flexible exoskeletons have more natural human-machine interaction, lower weight and cost, and have great application potential. Applying assist force according to the patient’s needs can give full play to the wearer’s remaining muscle strength, which is more conducive to muscle strength training and motor function recovery. In this paper, a design scheme of an elbow exoskeleton driven by flexible antagonistic cable actuators is proposed. The cable actuator is driven by a nonlinear series elastic mechanism, in which the elastic elements simulate the passive elastic properties of human skeletal muscle. Based on an improved elbow musculoskeletal model, the assist torque of exoskeleton is predicted. An assist-as-needed (AAN) control algorithm is proposed for the exoskeleton and experiments are carried out. The experimental results on the experimental platform show that the root mean square error between the predicted assist torque and the actual assist torque is 0.00226 Nm. The wearing experimental results also show that the AAN control method designed in this paper can reduce the activation of biceps brachii effectively when the exoskeleton assist level increases.

Joint Torque Prediction via Hybrid Neuromusculoskeletal Modelling during Gait Using Statistical Ground Reaction Estimates: An Exploratory Study

Joint torques of lower extremity are important clinical indicators of gait capability. This parameter can be quantified via hybrid neuromusculoskeletal modelling that combines electromyography-driven modelling and static optimisation. The simulations rely on kinematics and external force measurements, for example, ground reaction forces (GRF) and the corresponding centres of pressure (COP), which are conventionally acquired using force plates. This bulky equipment, however, hinders gait analysis in real-world environments. While this portability issue could potentially be solved by estimating the parameters through machine learning, the effect of the estimation errors on joint torque prediction with biomechanical models remains to be investigated. This study first estimated GRF and COP through feedforward artificial neural networks, and then leveraged them to predict lower-limb sagittal joint torques via (i) inverse dynamics and (ii) hybrid modelling. The approach was evaluated on five healthy subjects, individually. The predicted torques were validated with the measured torques, showing that hip was the most sensitive whereas ankle was the most resistive to the GRF/COP estimates for both models, with average metrics values being 0.70 < R2 < 0.97 and 0.069 < RMSE < 0.15 (Nm/kg). This study demonstrated the feasibility of torque prediction based on personalised (neuro)musculoskeletal modelling using statistical ground reaction estimates, thus providing insights into potential real-world mobile joint torque quantification.

A Muscle Synergy-Inspired Method of Detecting Human Movement Intentions Based on Wearable Sensor Fusion

Detecting human movement intentions is fundamental to neural control of robotic exoskeletons, as it is essential for achieving seamless transitions between different locomotion modes. In this study, we enhanced a muscle synergy-inspired method of locomotion mode identification by fusing the electromyography data with two types of data from wearable sensors (inertial measurement units), namely linear acceleration and angular velocity. From the finite state machine perspective, the enhanced method was used to systematically identify 2 static modes, 7 dynamic modes, and 27 transitions among them. In addition to the five broadly studied modes (level ground walking, ramps ascent/descent, stairs ascent/descent), we identified the transition between different walking speeds and modes of ramp walking at different inclination angles. Seven combinations of sensor fusion were conducted, on experimental data from 8 able-bodied adult subjects, and their classification accuracy and prediction time were compared. Prediction based on a fusion of electromyography and gyroscope (angular velocity) data predicted transitions earlier and with higher accuracy. All transitions and modes were identified with a total average classification accuracy of 94.5% with fused sensor data. For nearly all transitions, we were able to predict the next locomotion mode 300-500ms prior to the step into that mode.

Electromyography-informed modeling for estimating muscle activation and force alterations in Parkinson's disease

Electromyography (EMG)-driven neuromusculoskeletal modeling (NMSM) enables simulating the mechanical function of multiple muscle-tendon units as controlled by nervous system in the generation of complex movements. In the context of clinical assessment this may enable understanding biomechanical factor contributing to gait disorders such as one induced by Parkinson's disease (PD). In spite of the challenges in the development of patient-specific models, this preliminary study aimed at establishing a feasible and noninvasive experimental and modeling pipeline to be adopted in clinics to detect PD-induced gait alterations. Four different NMSM have been implemented for three healthy controls using CEINMS, an OpenSim-compatible toolbox. Models differed in the EMG-normalization methods used for calibration purposes (i.e. walking trial normalization and maximum voluntary contraction normalization) and in the set of experimental EMGs used for the musculotendon-unit mapping (i.e. 4 channels vs. 15 channels). Model accuracy assessment showed no statistically significant differences between the more complete model (non-clinically viable) and the proposed reduced one (clinically viable). The clinically viable reduced model was systematically applied on a dataset including ten PD's and thirteen healthy controls. Results showed significant differences in the neuromuscular control strategy of the PD group in term of muscle forces and joint torques. Indeed, PD patients displayed a significantly lower magnitude on force production and revealed a higher amount of force variability with the respect of the healthy controls. The estimated variables could become a measurable biomechanical outcome to assess and track both disease progression and its impact on gait in PD subjects.

Gesture Recognition Using Surface Electromyography and Deep Learning for Prostheses Hand: State-of-the-Art, Challenges, and Future

Amputation of the upper limb brings heavy burden to amputees, reduces their quality of life, and limits their performance in activities of daily life. The realization of natural control for prosthetic hands is crucial to improving the quality of life of amputees. Surface electromyography (sEMG) signal is one of the most widely used biological signals for the prediction of upper limb motor intention, which is an essential element of the control systems of prosthetic hands. The conversion of sEMG signals into effective control signals often requires a lot of computational power and complex process. Existing commercial prosthetic hands can only provide natural control for very few active degrees of freedom. Deep learning (DL) has performed surprisingly well in the development of intelligent systems in recent years. The significant improvement of hardware equipment and the continuous emergence of large data sets of sEMG have also boosted the DL research in sEMG signal processing. DL can effectively improve the accuracy of sEMG pattern recognition and reduce the influence of interference factors. This paper analyzes the applicability and efficiency of DL in sEMG-based gesture recognition and reviews the key techniques of DL-based sEMG pattern recognition for the prosthetic hand, including signal acquisition, signal preprocessing, feature extraction, classification of patterns, post-processing, and performance evaluation. Finally, the current challenges and future prospects in clinical application of these techniques are outlined and discussed.

Using Reinforcement Learning to Estimate Human Joint Moments From Electromyography or Joint Kinematics: An Alternative Solution to Musculoskeletal-Based Biomechanics

Reinforcement learning (RL) has potential to provide innovative solutions to existing challenges in estimating joint moments in motion analysis, such as kinematic or electromyography (EMG) noise and unknown model parameters. Here, we explore feasibility of RL to assist joint moment estimation for biomechanical applications. Forearm and hand kinematics and forearm EMGs from four muscles during free finger and wrist movement were collected from six healthy subjects. Using the proximal policy optimization approach, we trained two types of RL agents that estimated joint moment based on measured kinematics or measured EMGs, respectively. To quantify the performance of trained RL agents, the estimated joint moment was used to drive a forward dynamic model for estimating kinematics, which was then compared with measured kinematics using Pearson correlation coefficient. The results demonstrated that both trained RL agents are feasible to estimate joint moment for wrist and metacarpophalangeal (MCP) joint motion prediction. The correlation coefficients between predicted and measured kinematics, derived from the kinematics-driven agent and subject-specific EMG-driven agents, were 98% ± 1% and 94% ± 3% for the wrist, respectively, and were 95% ± 2% and 84% ± 6% for the metacarpophalangeal joint, respectively. In addition, a biomechanically reasonable joint moment-angle-EMG relationship (i.e., dependence of joint moment on joint angle and EMG) was predicted using only 15 s of collected data. In conclusion, this study illustrates that an RL approach can be an alternative technique to conventional inverse dynamic analysis in human biomechanics study and EMG-driven human-machine interfacing applications.

A convolutional neural network to identify motor units from high-density surface electromyography signals in real time

OBJECTIVES

This paper aims to investigate the feasibility and the validity of applying deep convolutional neural networks (CNN) to identify motor unit (MU) spike trains and estimate the neural drive to muscles from high-density electromyography (HD-EMG) signals in real time. Two distinct deep CNNs are compared with the convolution kernel compensation (CKC) algorithm using simulated and experimentally recorded signals. The effects of window size and step size of the input HD-EMG signals are also investigated.

APPROACH

The MU spike trains were first identified with the CKC algorithm. The HD-EMG signals and spike trains were used to train the deep CNN. Then, the deep CNN decomposed the HD-EMG signals into MU discharge times in real time. Two CNN approaches are compared with the CKC: 1) multiple single-output deep CNN (SO-DCNN) with one MU decomposed per network, and 2) one multiple-output deep CNN (MO-DCNN) to decompose all MUs (up to 23) with one network.

MAIN RESULTS

The MO-DCNN outperformed the SO-DCNN in terms of training time (3.2 to 21.4 s/epoch vs. 6.5 to 47.8 s/epoch, respectively) and prediction time (0.04 vs. 0.27 s/sample, respectively). The optimal window size and step size for MO-DCNN were 120 and 20 data points, respectively. It results in sensitivity of 98% and 85% with simulated and experimentally recorded HD-EMG signals, respectively. There is a high cross-correlation coefficient between the neural drive estimated with CKC and that estimated with MO-DCNN (range of r-value across conditions: 0.88-0.95).

SIGNIFICANCE

We demonstrate the feasibility and the validity of using deep CNN to accurately identify MU activity from HD-EMG with a latency lower than 80 ms, which falls within the lower bound of the human electromechanical delay. This method opens many opportunities for using the neural drive to interface humans with assistive devices.

Anterior Cruciate Ligament Loading Increases With Pivot-Shift Mechanism During Asymmetrical Drop Vertical Jump in Female Athletes

Background:

Frontal plane trunk lean with a side-to-side difference in lower extremity kinematics during landing increases unilateral knee abduction moment and consequently anterior cruciate ligament (ACL) injury risk. However, the biomechanical features of landing with higher ACL loading are still unknown. Validated musculoskeletal modeling offers the potential to quantify ACL strain and force during a landing task.

Purpose:

To investigate ACL loading during a landing and assess the association between ACL loading and biomechanical factors of individual landing strategies.

Study Design:

Descriptive laboratory study.

Methods:

Thirteen young female athletes performed drop vertical jump trials, and their movements were recorded with 3-dimensional motion capture. Electromyography-informed optimization was performed to estimate lower limb muscle forces with an OpenSim musculoskeletal model. A whole-body musculoskeletal finite element model was developed. The joint motion and muscle forces obtained from the OpenSim simulations were applied to the musculoskeletal finite element model to estimate ACL loading during participants’ simulated landings with physiologic knee mechanics. Kinematic, muscle force, and ground-reaction force waveforms associated with high ACL strain trials were reconstructed via principal component analysis and logistic regression analysis, which were used to predict trials with high ACL strain.

Results:

The median (interquartile range) values of peak ACL strain and force during the drop vertical jump were 3.3% (–1.9% to 5.1%) and 195.1 N (53.9 to 336.9 N), respectively. Four principal components significantly predicted high ACL strain trials, with 100% sensitivity, 78% specificity, and an area of 0.91 under the receiver operating characteristic curve (P < .001). High ACL strain trials were associated with (1) knee motions that included larger knee abduction, internal tibial rotation, and anterior tibial translation and (2) motion that included greater vertical and lateral ground-reaction forces, lower gluteus medius force, larger lateral pelvic tilt, and increased hip adduction.

Conclusion:

ACL loads were higher with a pivot-shift mechanism during a simulated landing with asymmetry in the frontal plane. Specifically, knee abduction can create compression on the posterior slope of the lateral tibial plateau, which induces anterior tibial translation and internal tibial rotation.

Clinical Relevance:

Athletes are encouraged to perform interventional and preventive training to improve symmetry during landing.

A Conceptual Blueprint for Making Neuromusculoskeletal Models Clinically Useful

The ultimate goal of most neuromusculoskeletal modeling research is to improve the treatment of movement impairments. However, even though neuromusculoskeletal models have become more realistic anatomically, physiologically, and neurologically over the past 25 years, they have yet to make a positive impact on the design of clinical treatments for movement impairments. Such impairments are caused by common conditions such as stroke, osteoarthritis, Parkinson’s disease, spinal cord injury, cerebral palsy, limb amputation, and even cancer. The lack of clinical impact is somewhat surprising given that comparable computational technology has transformed the design of airplanes, automobiles, and other commercial products over the same time period. This paper provides the author’s personal perspective for how neuromusculoskeletal models can become clinically useful. First, the paper motivates the potential value of neuromusculoskeletal models for clinical treatment design. Next, it highlights five challenges to achieving clinical utility and provides suggestions for how to overcome them. After that, it describes clinical, technical, collaboration, and practical needs that must be addressed for neuromusculoskeletal models to fulfill their clinical potential, along with recommendations for meeting them. Finally, it discusses how more complex modeling and experimental methods could enhance neuromusculoskeletal model fidelity, personalization, and utilization. The author hopes that these ideas will provide a conceptual blueprint that will help the neuromusculoskeletal modeling research community work toward clinical utility.

The Sensor-Based Biomechanical Risk Assessment at the Base of the Need for Revising of Standards for Human Ergonomics

Due to the epochal changes introduced by "Industry 4.0", it is getting harder to apply the varying approaches for biomechanical risk assessment of manual handling tasks used to prevent work-related musculoskeletal disorders (WMDs) considered within the International Standards for ergonomics. In fact, the innovative human-robot collaboration (HRC) systems are widening the number of work motor tasks that cannot be assessed. On the other hand, new sensor-based tools for biomechanical risk assessment could be used for both quantitative "direct instrumental evaluations" and "rating of standard methods", allowing certain improvements over traditional methods. In this light, this Letter aims at detecting the need for revising the standards for human ergonomics and biomechanical risk assessment by analyzing the WMDs prevalence and incidence; additionally, the strengths and weaknesses of traditional methods listed within the International Standards for manual handling activities and the next challenges needed for their revision are considered. As a representative example, the discussion is referred to the lifting of heavy loads where the revision should include the use of sensor-based tools for biomechanical risk assessment during lifting performed with the use of exoskeletons, by more than one person (team lifting) and when the traditional methods cannot be applied. The wearability of sensing and feedback sensors in addition to human augmentation technologies allows for increasing workers' awareness about possible risks and enhance the effectiveness and safety during the execution of in many manual handling activities.