Estimation of Muscle Force Based on Neural Drive in a Hemispheric Stroke Survivor

Decoding hand and wrist movement intention from chronic stroke survivors with hemiparesis using a user-friendly, wearable EMG-based neural interface

OBJECTIVE

Seventy-five percent of stroke survivors, caregivers, and health care professionals (HCP) believe current therapy practices are insufficient, specifically calling out the upper extremity as an area where innovation is needed to develop highly usable prosthetics/orthotics for the stroke population. A promising method for controlling upper extremity technologies is to infer movement intention non-invasively from surface electromyography (EMG). However, existing technologies are often limited to research settings and struggle to meet user needs.

APPROACH

To address these limitations, we have developed the NeuroLife® EMG System, an investigational device which consists of a wearable forearm sleeve with 150 embedded electrodes and associated hardware and software to record and decode surface EMG. Here, we demonstrate accurate decoding of 12 functional hand, wrist, and forearm movements in chronic stroke survivors, including multiple types of grasps from participants with varying levels of impairment. We also collected usability data to assess how the system meets user needs to inform future design considerations.

MAIN RESULTS

Our decoding algorithm trained on historical- and within-session data produced an overall accuracy of 77.1 ± 5.6% across 12 movements and rest in stroke participants. For individuals with severe hand impairment, we demonstrate the ability to decode a subset of two fundamental movements and rest at 85.4 ± 6.4% accuracy. In online scenarios, two stroke survivors achieved 91.34 ± 1.53% across three movements and rest, highlighting the potential as a control mechanism for assistive technologies. Feedback from stroke survivors who tested the system indicates that the sleeve's design meets various user needs, including being comfortable, portable, and lightweight. The sleeve is in a form factor such that it can be used at home without an expert technician and can be worn for multiple hours without discomfort.

SIGNIFICANCE

The NeuroLife EMG System represents a platform technology to record and decode high-resolution EMG for the real-time control of assistive devices in a form factor designed to meet user needs. The NeuroLife EMG System is currently limited by U.S. federal law to investigational use.

Toward a generalizable deep CNN for neural drive estimation across muscles and participants

Objective.High-density electromyography (HD-EMG) decomposition algorithms are used to identify individual motor unit (MU) spike trains, which collectively constitute the neural code of movements, to predict motor intent. This approach has advanced from offline to online decomposition, from isometric to dynamic contractions, leading to a wide range of neural-machine interface applications. However, current online methods need offline retraining when applied to the same muscle on a different day or to a different person, which limits their applications in a real-time neural-machine interface. We proposed a deep convolutional neural network (CNN) framework for neural drive estimation, which takes in frames of HD-EMG signals as input, extracts general spatiotemporal properties of MU action potentials, and outputs the number of spikes in each frame. The deep CNN can generalize its application without retraining to HD-EMG data recorded in separate sessions, muscles, or participants.Approach.We recorded HD-EMG signals from the vastus medialis and vastus lateralis muscles from five participants while they performed isometric contractions during two sessions separated by ∼20 months. We identified MU spike trains from HD-EMG signals using a convolutive blind source separation (BSS) method, and then used the cumulative spike train (CST) of these MUs and the HD-EMG signals to train and validate the deep CNN.Main results.On average, the correlation coefficients between CST from the BSS and that from deep CNN were0.983±0.006for leave-one-out across-sessions-and-muscles validation and0.989±0.002for leave-one-out across-participants validation. When trained with more than four datasets, the performance of deep CNN saturated at0.984±0.001for cross validations across muscles, sessions, and participants.Significance.We can conclude that the deep CNN is generalizable across the aforementioned conditions without retraining. We could potentially generate a robust deep CNN to estimate neural drive to muscles for neural-machine interfaces.

A generic neural network model to estimate populational neural activity for robust neural decoding

BACKGROUND

Robust and continuous neural decoding is crucial for reliable and intuitive neural-machine interactions. This study developed a novel generic neural network model that can continuously predict finger forces based on decoded populational motoneuron firing activities.

METHOD

We implemented convolutional neural networks (CNNs) to learn the mapping from high-density electromyogram (HD-EMG) signals of forearm muscles to populational motoneuron firing frequency. We first extracted the spatiotemporal features of EMG energy and frequency maps to improve learning efficiency, given that EMG signals are intrinsically stochastic. We then established a generic neural network model by training on the populational neuron firing activities of multiple participants. Using a regression model, we continuously predicted individual finger forces in real-time. We compared the force prediction performance with two state-of-the-art approaches: a neuron-decomposition method and a classic EMG-amplitude method.

RESULTS

Our results showed that the generic CNN model outperformed the subject-specific neuron-decomposition method and the EMG-amplitude method, as demonstrated by a higher correlation coefficient between the measured and predicted forces, and a lower force prediction error. In addition, the CNN model revealed more stable force prediction performance over time.

CONCLUSIONS

Overall, our approach provides a generic and efficient continuous neural decoding approach for real-time and robust human-robot interactions.

Finger Force Estimation using Motor Unit Discharges Across Forearm Postures

BACKGROUND

Myoelectric-based decoding has gained popularity in upper-limb neural-machine interfaces. Motor unit (MU) firings decomposed from surface electromyographic (EMG) signals can represent motor intent, but EMG properties at different arm configurations can change due to electrode shift and differing neuromuscular states. This study investigated whether isometric fingertip force estimation using MU firings is robust to forearm rotations from a neutral to either a fully pronated or supinated posture.

METHODS

We extracted MU information from high-density EMG of the extensor digitorum communis in two ways: (1) Decomposed EMG in all three postures (MU-AllPost); and (2) Decomposed EMG in neutral posture (MU-Neu), and extracted MUs (separation matrix) were applied to other postures. Populational MU firing frequency estimated forces scaled to subjects' maximum voluntary contraction (MVC) using a regression analysis. The results were compared with the conventional EMG-amplitude method.

RESULTS

We found largely similar root-mean-square errors (RMSE) for the two MU-methods, indicating that MU decomposition was robust to postural differences. MU-methods demonstrated lower RMSE in the ring (EMG = 6.23, MU-AllPost = 5.72, MU-Neu = 5.64 %MVC) and pinky (EMG = 6.12, MU-AllPost = 4.95, MU-Neu = 5.36 %MVC) fingers, with mixed results in the middle finger (EMG = 5.47, MU-AllPost = 5.52, MU-Neu = 6.19% MVC).

CONCLUSION

Our results suggest that MU firings can be extracted reliably with little influence from forearm posture, highlighting its potential as an alternative decoding scheme for robust and continuous control of assistive devices.

Decoding Muscle Force From Motor Unit Firings Using Encoder-Decoder Networks

Appropriate interpretation of motor unit (MU) activities after surface EMG (sEMG) decomposition is a key factor to decode motor intentions in a noninvasive and physiologically meaningful way. However, there are great challenges due to the difficulty in cross-trial MU tracking and unavoidable loss of partial MU information resulting from incomplete decomposition. In light of these issues, this study presents a novel framework for interpreting MU activities and applies it to decode muscle force. The resulting MUs were clustered and classified into different categories by characterizing their spatially distributed firing waveforms. The process served as a general MU tracking method. On this basis, after transferring the MU firing trains to twitch force trains by a twitch force model, a deep network was designed to predict the normalized force. In addition, MU category distribution was examined to calibrate the actual force level, while functions of some unavailable MUs were compensated. To investigate the effectiveness of this framework, high-density sEMG signals were recorded using an 8×8 electrode array from the abductor pollicis brevis muscles of eight subjects, while thumb abduction force was measured. The proposed method outperformed three common methods ( ) yielding the lowest root mean square deviation of 6.68% ± 1.29% and the highest fitness ( R² ) of 0.94 ± 0.04 between the predicted force and the actual force. This study offers a valuable, computational solution for interpreting individual MU activities, and its effectiveness was confirmed in muscle force estimation.

Adaptive Real-time Decomposition of Electromyogram During Sustained Muscle Activation: A Simulation Study

OBJECTIVE

Real-time decomposition of electromyogram (EMG) into constituent motor unit (MU) activity has shown promising applications in neurophysiology and human-machine interactions. Existing decomposition methods could not accommodate stochastic variations in EMG signals such as drifts of action potential amplitudes and MU recruitment-derecruitment (rotation) patterns during long-term recordings. The objective of this study was to develop an adaptive real-time decomposition approach suitable for prolonged muscle activation.

METHODS

We developed a parallel-double-thread computation algorithm. The backend thread initiated and periodically refined and updated the MU information (separation matrix) using independent component analysis and convolution kernel compensation. The frontend thread performed the real-time decomposition. We evaluated our algorithm on synthesized high-density EMG signals, in which MUs were recruited-derecruited sporadically and MU action potentials amplitude drifted over time. Different signal-to-noise levels were also simulated.

RESULTS

Compared with the decomposition without the adaptive processes, periodically fine-tuned and updated separation matrix increased identifiable MU number by 3-4 fold over 30-minute of recordings. The increased MU number was more prominent at higher signal-to-noise ratios. The decomposition accuracy also increased by up to 10% with greater improvement observed at higher muscle contraction levels.

CONCLUSION

The adaptive algorithm can maintain the decomposition performance over time, allows us to continuously track the same MUs during sustained activation, and, at the same time, can add newly recruited MU information to existing separation matrix.

SIGNIFICANCE

Our approach showed robust performance over time, which has the potential to longitudinally evaluate MU firing and recruitment properties and improve neural decoding performance for neural-machine interactions.

Concurrent Estimation of Finger Flexion and Extension Forces Using Motoneuron Discharge Information

OBJECTIVE

A reliable neural-machine interface offers the possibility of controlling advanced robotic hands with high dexterity. The objective of this study was to develop a decoding method to estimate flexion and extension forces of individual fingers concurrently.

METHODS

First, motor unit (MU) firing information was identified through surface electromyogram (EMG) decomposition, and the MUs were further categorized into different pools for the flexion and extension of individual fingers via a refinement procedure. MU firing rate at the populational level was calculated, and the individual finger forces were then estimated via a bivariate linear regression model (neural-drive method). Conventional EMG amplitude-based method was used as a comparison.

RESULTS

Our results showed that the neural-drive method had a significantly better performance (lower estimation error and higher correlation) compared with the conventional method.

CONCLUSION

Our approach provides a reliable neural decoding method for dexterous finger movements.

SIGNIFICANCE

Further exploration of our method can potentially provide a robust neural-machine interface for intuitive control of robotic hands.

Muscle Force Estimation Based on Neural Drive Information From Individual Motor Units

Estimation of muscle contraction force based on the macroscopic feature of surface electromyography (SEMG) has been widely reported, but the use of microscopic neural drive information has not been thoroughly investigated. In this study, a novel method is proposed to process individual motor unit (MU) activities (firing sequences and action potential waveforms) derived from the decomposition of high density SEMG (HD-SEMG), and it is applied to muscle force estimation. In the proposed method, a supervised machine learning approach was conducted to determine the twitch force of each MU according to its action potential waveforms, which enables separate calculation of every MU's contribution to force. Thus, the muscle force was predicted through a physiologically meaningful muscle force model. In the experiment, HD-SEMG data were recorded from the abductor pollicis brevis muscles of eight healthy subjects during their performance of thumb abduction with the force increasing gradually from zero to four force levels (10%, 20%, 30%, 40% of the maximal voluntary contraction), while the true muscle force was measured simultaneously. When the proposed method was used, the root mean square difference (RMSD) of the error of the estimated force with respect to the measured force was reported to be 8.3% ± 2.8%. The proposed method also significantly outperformed the other four common methods for force estimation (RMSD: from 11.7% to 20%, ), demonstrating its effectiveness. This study offers a useful tool for exploiting the neural drive information towards muscle force estimation with improved precision. The proposed method has wide applications in precise motor control, sport and rehabilitation medicine.

Electrically compensated, tattoo-like electrodes for epidermal electrophysiology at scale

Epidermal electrophysiology is widely carried out for disease diagnosis, performance monitoring, human-machine interaction, etc. Compared with thick, stiff, and irritating gel electrodes, emerging tattoo-like epidermal electrodes offer much better wearability and versatility. However, state-of-the-art tattoo-like electrodes are limited in size (e.g., centimeters) to perform electrophysiology at scale due to challenges including large-area fabrication, skin lamination, and electrical interference from long interconnects. Therefore, we report large-area, soft, breathable, substrate- and encapsulation-free electrodes designed into transformable filamentary serpentines that can be rapidly fabricated by cut-and-paste method. We propose a Cartan curve-inspired transfer process to minimize strain in the electrodes when laminated on nondevelopable skin surfaces. Unwanted signals picked up by the unencapsulated interconnects can be eliminated through a previously unexplored electrical compensation strategy. These tattoo-like electrodes can comfortably cover the whole chest, forearm, or neck for applications such as multichannel electrocardiography, sign language recognition, prosthetic control or mapping of neck activities.

Hybrid Encoder-Decoder Deep Networks for Decoding Neural Drive Information towards Precise Muscle Force Estimation

How to utilize and interpret microscopic motor unit (MU) activities after surface electromyogram (sEMG) decomposition towards accurate decoding of the neural control remains a great challenge. In this study, a novel framework of hybrid encoder-decoder deep networks is proposed to process the microscopic neural drive information and it is applied to precise muscle force estimation. After a high-density sEMG (HD-sEMG) decomposition was performed using the progressive FastICA peel-off algorithm, a muscle twitch force model was then applied to basically convert each channel's electric waveform (i.e., action potential) of each MU into a twitch force. Next, hybrid encoder-decoder deep networks were performed on every 50 ms of segment of the summation of twitch force trains from all decomposed MUs. The encoder network was designed to characterize spatial information of MU's force contribution over all channels, and the decoder network finally decoded the muscle force. This framework was validated on HD-sEMG recordings from the abductor pollicis brevis muscles of five subjects by a thumb abduction task using an 8 × 8 grid. The proposed framework yielded a mean root mean square error of 6.62% ± 1.26% and a mean coefficient of determination value of 0.95 ± 0.03 from a linear regression analysis between the estimated force and actual force over all data trials, and it outperformed three common methods with statistical significance (p < 0.001). This study offers a valuable solution for interpreting microscopic neural drive information and demonstrates its success in predicting muscle force.

Dexterous Force Estimation during Finger Flexion and Extension Using Motor Unit Discharge Information

With the development of advanced robotic hands, a reliable neural-machine interface is essential to take full advantage of the functional dexterity of the robots. In this preliminary study, we developed a novel method to estimate isometric forces of individual fingers continuously and concurrently during dexterous finger flexion and extension. Specifically, motor unit (MU) discharge activity was extracted from the surface high-density electromyogram (EMG) signals recorded from the finger extensors and flexors, respectively. The MU information was separated into different groups to be associated with the flexion or extension of individual fingers and was then used to predict individual finger forces during multi-finger flexion and extension tasks. Compared with the conventional EMG amplitude-based method, our method can obtain a better force estimation performance (a higher correlation and a smaller estimation error between the predicted and the measured force) when a linear regression model was used. Further exploration of our method can potentially provide a robust neural-machine interface for intuitive control of robotic hands.