If you can't explain how this is going to get survivors recovered; THEN TOTALLY FUCKING USELESS!
Proportional myoelectric control of a virtual bionic arm in participants with hemiparesis, muscle spasticity, and impaired range of motion
Journal of NeuroEngineering and Rehabilitation volume 21, Article number: 222 (2024)
Abstract
Background
This research aims to improve the control of assistive devices for individuals with hemiparesis after stroke by providing intuitive and proportional motor control. Stroke is the leading cause of disability in the United States, with 80% of stroke-related disability coming in the form of hemiparesis, presented as weakness or paresis on half of the body. Current assistive exoskeletonscontrolled via electromyography do not allow for fine force regulation. Current control strategies provide only binary, all-or-nothing control based on a linear threshold of muscle activity.
Methods
In this study, we demonstrate the ability of participants with hemiparesis to finely regulate their muscle activity to proportionally control the position of a virtual bionic arm. Ten stroke survivors and ten healthy, aged-matched controls completed a target-touching task with the virtual bionic arm. We compared the signal-to-noise ratio (SNR) of the recorded electromyography (EMG) signals used to train the control algorithms and the task performance using root mean square error, percent time in target, and maximum hold time within the target window. Additionally, we looked at the correlation between EMG SNR, task performance, and clinical spasticity scores.
Results
All stroke survivors were able to achieve proportional EMG control despite limited or no physical movement (i.e., modified Ashworth scale of 3). EMG SNR was significantly lower for the paretic arm than the contralateral nonparetic arm and healthy control arms, but proportional EMG control was similar across conditions for hand grasp. In contrast, proportional EMG control for hand extension was significantly worse for paretic arms than healthy control arms. The participants’ age, time since their stroke, clinical spasticity rate, and history of botulinum toxin injections had no impact on proportional EMG control.
Conclusions
It is possible to provide proportional EMG control of assistive devices from a stroke survivor’s paretic arm. Importantly, information regulating fine force output is still present in muscle activity, even in extreme cases of spasticity where there is no visible movement. Future work should incorporate proportional EMG control into upper-limb exoskeletons to enhance the dexterity of stroke survivors.
Introduction
Stroke is the leading cause of disability in the United States, with more than 795,000 people suffering a stroke each year [1]. 80% of stroke-related motor deficits are in the form of upper-limb hemiparesis [1, 2]. Hemiparesis presents as a one-sided weakness or paralysis and is caused by damage to the central nervous system from the stroke. This damage to the central nervous system interrupts descending motor control, dissociates motor responses and sensory inputs, and can lead to hyperexcitability of the muscles, causing spasticity [3, 4].
After a stroke, residual muscle activity in the hemiparetic arm can be recorded using surface electromyography (EMG), even in patients with no detectable muscle activity as measured by traditional clinical assessments [5]. Even in the chronic phase, muscle activity persists and can be improved over time [6, 7]. However, the ability to modulate muscle activity is diminished in chronic stroke patients [8], which can lead to abnormal muscle activations and task difficulty [9]. Motor deficits have also been found in the nonparetic arm after a stroke [10].
Ultimately, hemiparesis makes it difficult to complete activities of daily living, reducing the quality of life and autonomy [11]. Assistive devices, like powered orthoses [12, 13] and functional electrical stimulation (FES) [14], have been used to restore hand function to stroke patients with hemiparesis, thereby restoring independence and increasing quality of life [13, 15, 16]. Because muscle activity still persists in hemiparetic stroke patients [5], EMG can serve as an intuitive control signal for FES [17, 18] and powered orthoses [13, 19]. However, one challenge when using paretic EMG for control is the presence of involuntary EMG increases when the individual moves another part of their arm [20]; this has been shown to decrease the accuracy of EMG-based control algorithms [21].
Due to the complexities of EMG and the abnormalities of paretic EMG [8, 20, 22,23,24,25], current EMG control algorithms most often employ a binary, “all-or-nothing” approach that simply detects if the muscle is active or inactive. When this binary control is used to control the position of a hand exoskeleton, individuals are limited to maximally closing or maximally opening their hand. Because there is a fixed force output from the exoskeleton, binary control makes it difficult, if not impossible, to perform fine motor actions. Variable force output is critical in tasks like manipulating fragile objects [26], preventing slips [27], and grasping under uncertain conditions [28].
Pattern recognition has been used to extract more precise control from EMG activity for research applications with exoskeletons [29,30,31] and commercial applications with prostheses [32,33,34]. However, pattern recognition systems still provide only discrete class predictions and do not inherently provide proportional position control or fine force regulation. Proportional control of upper-limb exoskeletons has been shown with proportional force [35,36,37], torque [38, 39], velocity [40, 41], and position control [41, 42]. The proportional control demonstrated in these studies mainly focuses on the arm from the wrist up to the shoulder; those that do look at control of the hand use force [37], torque [39], and velocity control [40], and not position control. The joint from the wrist to the shoulder collectively supports gross motor function (i.e., positioning of the hand in space) and leverages large, anatomically distinct muscles for control (e.g., the biceps and triceps). In contrast, fine motor control of the hand involves the coordination of multiple small muscles densely packed in the forearm. In the adjacent field of upper-limb prosthetic control, proportional position control is common [43,44,45,46,47] and has been shown to increase performance relative to velocity control for a prosthetic hand [48]. Proportional position control is also more closely aligned with the natural encoding for hand control, which is in terms of joint position [49,50,51].
A key challenge in realizing proportional position control for upper-limb exoskeletons is that the primary patient population, stroke patients, often has severe muscle spasticity [52, 53]. Muscle spasticity often manifests as lower EMG SNR [54], excessive co-contractions [55, 56], and delayed muscle activation/relaxation [22,23,24,25, 56]. Due to these signal challenges, proportional position control of exoskeletons has often been performed using EMG from the nonparetic, contralateral limb [31, 57, 58] rather than the affected paretic limb, which limits the ability to perform bilateral tasks. Others have only tested exoskeleton control with healthy participants rather than the target patient population of stroke survivors [30, 59].
Using high-density EMG in conjunction with machine learning can be a solution to obtain more robust and dexterous control from paretic EMG. High-density EMG gathers data from most, if not all, the muscles, and avoids the need to meticulously identify isolated EMG from desired muscles. Machine learning is then used to identify and exploit even the smallest differences among the ensemble of muscle activity when attempting different movements. Indeed, high-density EMG has already been used to classify hand gestures with high accuracy with paretic EMG from stroke survivors [29, 60]. Building on these works, here we propose high-density EMG in conjunction with machine learning to provide proportional control of the hand. To do this, we leverage a modified Kalman filter, which has been demonstrated to provide robust proportional position control with healthy individuals and with upper-limb amputees [46].
In this study, we specifically investigated the ability to extract proportional position control from the extrinsic hand muscles of stroke survivors with hemiparesis. We show that all participants were able to achieve proportional EMG control, regardless of their age, time since their stroke, clinical spasticity rate, and history of botulinum toxin injections. We also show that EMG signal-to-noise ratio and proportional control are better for hand grasp than hand extension, consistent with the neurophysiology of post-stroke spasticity [61]. These results can help guide the implementation and patient inclusion criteria for future assistive hand exoskeletons with proportional EMG control.
More at link.
No comments:
Post a Comment