Extra-Neural Signals From Severed Nerves Enable Intrinsic Hand Movements In Transhumeral Amputations

Robotic prostheses controlled by myoelectric signals can restore limited but important hand function in individuals with upper limb amputation. The lack of individual finger control highlights the yet insurmountable gap to fully replace a biologic hand. Implanted electrodes around severed nerves have been used to elicit sensations perceived as arising from the missing limb, but using such extra-neural electrodes to record motor signals that allow for the decoding of phantom movements has remained elusive. Here, we showed the feasibility of using signals from non-penetrating neural electrodes to decode intrinsic hand and finger movements in individuals with above-elbow amputations. We found that information recorded with extra-neural electrodes alone was enough to decode phantom hand and individual finger movements with high accuracy, and as expected, the addition of myoelectric signals reduced classification errors both in offline and in real-time decoding.

Facilitative Exercise for Surface Myoelectric Activity Using Robot Arm Control System – Training Scheme with Gradually Increasing Difficulty Level –

Various control systems for robot arms using surface myoelectric signals have been developed. Abundant pattern-recognition techniques have been proposed to predict human motion intent based on these signals. However, it is laborious for users to train the voluntary control of myoelectric signals using those systems. In this research, we aim to develop a rehabilitation support system for hemiplegic upper limbs with a robot arm controlled by surface myoelectric signals. In this study, we construct a simple one-link robot arm that is controlled by estimating the wrist motion from the surface myoelectric signals on the forearm. We propose a training scheme with gradually increasing difficulty level for robot arm manipulation to evoke surface myoelectric signals. Subsequently, we investigate the possibility of facilitative exercise for the voluntary surface myoelectric activity of the desired muscles through trial experiments.

An Electrolarynx Control Method Using Myoelectric Signals from the Neck

Patients who have lost vocal cord function due to laryngeal cancer or laryngeal injury are incapable of speech because it is impossible to generate the laryngeal tone from which the voice originates. For such patients, various speech production substitutes have been devised and put into practical use. The electrolarynx is one of these speech production substitutes and it can be used with relative ease. However, the sound is sometimes difficult to hear and its quality is monotonous. Therefore, focusing on the control method to improve the articulation of the electrolarynx, we have proposed an electrolarynx controlled by myoelectric signals of the neck. The sternohyoid muscle, which is located in the superficial layer of the neck, was the source of the myoelectric signals. This muscle is active during speech, and its activity increases mainly at the time of speech in a low voice. We succeeded in detecting the surface myoelectric signals of the sternohyoid muscle and performing on/off control of the electrolarynx by signal processing. This report includes the derivation of a control function for converting into a control signal of the fundamental frequency of the electrolarynx from the relationship between the myoelectric signals and the fundamental frequency of the voice. This report also includes an evaluation of the controllability of the electrolarynx by comparing the obtained control signal with the user’s intention. Regarding the control of the fundamental frequency, we have proposed a method of control in three stages – high, medium, and low – and a method of control in two stages – high and low – and compared their performances. The results of the three-stage control indicated that the use of the logarithm as a control function for converting the myoelectric signals into the fundamental frequency of the electrolarynx succeeded in the control at an accuracy of 90% or more by changing the pitch of the generated sound depending on the subjects. It was also indicated that the error rate was as low as less than 20%, while maintaining a constant sound. This makes it clear that the use of the logarithm as a control function gives the highest controllability. The two-stage control exhibits a very high control success rate exceeding 90%, regardless of the type of control function; in particular, the control function using the logarithm exhibits a control success rate exceeding 95%. These results indicate that the electrolarynx control function obtained using the logarithmic function has excellent controllability.

Myoelectric Control of Robotic Leg Prostheses and Exoskeletons: A Review

Myoelectric signals from the human motor control system can improve the real-time control and neural-machine interface of robotic leg prostheses and exoskeletons for different locomotor activities (e.g., walking, sitting down, stair ascent, and non-rhythmic movements). Here we review the latest advances in myoelectric control designs and propose future directions for research and innovation. We review the different wearable sensor technologies, actuators, signal processing, and pattern recognition algorithms used for myoelectric locomotor control and intent recognition, with an emphasis on the hierarchical architectures of volitional control systems. Common mechanisms within the control architecture include 1) open-loop proportional control with fixed gains, 2) active-reactive control, 3) joint mechanical impedance control, 4) manual-tuning torque control, 5) adaptive control with varying gains, and 6) closed-loop servo actuator control. Based on our review, we recommend that future research consider using musculoskeletal modeling and machine learning algorithms to map myoelectric signals from surface electromyography (EMG) to actuator joint torques, thereby improving the automation and efficiency of next-generation EMG controllers and neural interfaces for robotic leg prostheses and exoskeletons. We also propose an example model-based adaptive impedance EMG controller including muscle and multibody system dynamics. Ongoing advances in the engineering design of myoelectric control systems have implications for both locomotor assistance and rehabilitation.

Motor unit recruitment patterns of the quadriceps differ between continuous high- and low-torque isometric knee extension to momentary failure

The size principle is a theory of motor unit (MU) recruitment that suggests MUs are recruited in an orderly manner from the smallest (lower threshold) to the largest (higher threshold) MUs. A consequence of this biophysical theory is that, for isometric contractions, recruitment is dependent on the intensity of actual effort required to meet task demands. This concept has been supported by modelling work demonstrating that, in tasks performed to momentary failure, full MU recruitment will have occurred upon reaching failure irrespective of the force requirements of the task. However, in vivo studies examining this are limited. Therefore, the aim of the current study was to examine MU recruitment of the quadriceps under both higher- and lower-torque (70% and 30% of MVC, respectively) isometric knee extension, performed to momentary failure. Specifically, we compared surface electromyography (sEMG) frequency characteristics, determined by wavelet analysis, across the two continuous isometric knee extension tasks to identify potential differences in recruitment patterns. A convenience sample of 10 recreationally active adult males (height: mean = 179.6, SD = 6.0 cm; mass: mean = 76.8, SD = 7.3 kg; age: mean = 26 SD = 7 years) with previous resistance training experience (mean = 6, SD = 3 years) were recruited. Using a within-session, repeated-measures, randomised crossover design participants performed the knee extension tasks whilst sEMG was collected from the vastus medialis (VM), rectus femoris (RF) and vastus lateralis (VL). Myoelectric signals were decomposed into intensities as a function of time and frequency using an EMG-specific wavelet transformation. Our first analysis compared the mean frequency at momentary failure; second, we investigated the effects of load on relative changes in wavelet intensities; finally, we quantified the degree of wavelet similarity over time. Wavelet-based calculation of the mean signal frequency appeared to show similar mean frequency characteristics occurring when reaching momentary failure. However, individual wavelets revealed that different changes in frequency components occurred between the two tasks, suggesting that patterns of recruitment differed. Low-torque conditions resulted in an increase in intensity of all frequency components across the trials for each muscle whereas high-torque conditions resulted in a wider range of frequency components contained within the myoelectric signals at the beginning of the trials. However, as the low-torque trial neared momentary failure there was an increased agreement between conditions across wavelets. Our results corroborate modelling studies as well as recent biopsy evidence, suggesting overall MU recruitment may largely be similar for isometric tasks performed to momentary failure with the highest threshold MUs likely recruited, despite being achieved with differences in the pattern of recruitment over time utilised.

Evaluation of an Information Flow Gain Algorithm for Microsensor Information Flow in Limber Motor Rehabilitation

This paper conducts an evaluative study on the rehabilitation of limb motor function by using a microsensor information flow gain algorithm and investigates the surface electromyography (EMG) signals of the upper limb during rehabilitation training. The surface EMG signals contain a large amount of limb movement information. By analysing and processing the surface EMG signals, we can grasp the human muscle movement state and identify the human upper limb movement intention. The EMG signals were processed by the trap and filter combination denoising method and wavelet denoising method, respectively, the signal-to-noise ratio was used to evaluate the noise reduction effect, and finally, the wavelet denoising method with a better noise reduction effect was selected to process all the EMG signals. After the noise is removed, the signal is extracted in the time domain and frequency domain, and the root mean square (RMS), absolute mean, median frequency in the time domain, and average power frequency in the frequency domain are selected and input to the classifier for pattern recognition. The support vector machine is used to classify the myoelectric signals and optimize the parameters in the support vector machine using the grid search method and particle swarm optimization algorithm and classify the test samples using the trained support vector machine. Compared with the classification results of the grid search optimized support vector machine, the optimized vector machine has a 7% higher recognition rate, reaching 85%. The action recognition classification method of myoelectric signals is combined with an upper limb rehabilitation training platform to verify the feasibility of using myoelectric signals for rehabilitation training. After the classifier recognizes the upper limb movements, the upper computer sends movement commands to the controller to make the rehabilitation platform move according to the recognition results, and finally, the movement execution accuracy of the rehabilitation platform reaches 80% on average.

Sensing and decoding the neural drive to paralyzed muscles during attempted movements of a person with tetraplegia using a sleeve array

Motor neurons in the brain and spinal cord convey information about motor intent that can be extracted and interpreted to control assistive devices, such as computers, wheelchairs, and robotic manipulators. However, most methods for measuring the firing activity of single neurons rely on implanted microelectrodes. Although intracortical brain-computer interfaces (BCIs) have been shown to be safe and effective, the requirement for surgery poses a barrier to widespread use. Here, we demonstrate that a wearable sensor array can detect residual motor unit activity in paralyzed muscles after severe cervical spinal cord injury (SCI). Despite generating no observable hand movement, volitional recruitment of motor neurons below the level of injury was observed across attempted movements of individual fingers and overt wrist and elbow movements. Subgroups of motor units were coactive during flexion or extension phases of the task. Single digit movement intentions were classified offline from the EMG power (RMS) or motor unit firing rates with median classification accuracies >75% in both cases. Simulated online control of a virtual hand was performed with a binary classifier to test feasibility of real time extraction and decoding of motor units. The online decomposition algorithm extracted motor units in 1.2 ms, and the firing rates predicted the correct digit motion 88 ± 24% of the time. This study provides the first demonstration of a wearable interface for recording and decoding firing rates of motor neurons below the level of injury in a person with tetraplegia after motor complete SCI.Significance StatementA wearable electrode array and machine learning methods were used to record and decode myoelectric signals and motor unit firing in paralyzed muscles of a person with motor complete tetraplegia. Motor unit action potentials were extracted from myoelectric signals during attempted movements of the fingers and voluntary movements of the wrist and elbow. The patterns of EMG and motor unit firing rates were highly task-specific, even in the absence of visible motion in the limb, enabling accurate classification of attempted movements of single digits. These results demonstrate the potential to create a wearable sensor for determining movement intentions from spared motor neurons, which may enable people with severe tetraplegia to control assistive devices such as computers, wheelchairs, and robotic manipulators.