Cognitive-based Rehabilitation Platform of Hand Grasp after Spinal Cord Injury using Virtual Reality and Instrumented Wearables

项目摘要 颈椎脊髓损伤（SCI）会损害手的功能和参与环境的能力。作为 结果，执行日常生活活动的能力可能严重受损，直接导致降低 生活质量物理治疗仍然是SCI后恢复功能性能力的主要途径。物理 康复过程要求参与者承诺在功能方面取得有意义的进展。 认知参与的康复方法可以促进对实践的更大承诺， 改善运动学习。我们建议开发利用虚拟现实（VR）的创新平台， 仪器可穿戴设备，增强SCI后手部抓握和伸展运动学习期间的认知因素。 这些因素包括更强的代理感，或控制感，以及多感官反馈。感 的代理涉及更大的运动控制，和各种感觉反馈方式（视觉，听觉， 和触觉）在运动训练中被证明是有效的。然而，这些因素并没有得到很好的考虑， 传统的物理治疗方法。我们已经开发了两个新颖的基于认知的平台， 恢复抓握和伸展功能。我们建议在患有慢性SCI的退伍军人中测试每个平台， 宫颈水平。在目标1中，我们将研究我们的第一个平台，"认知"手套，如何提高功能， 抓紧这款手套包括力和弯曲传感器，它们为机器学习算法提供输入， 预测何时实现安全抓取。手套通过板载传感器提醒用户安全抓握 提供视觉（LED）、音频（蜂鸣器）和触觉（振动器）反馈的模块。在培训过程中， 以逐渐缩短的时间间隔提供，以基于神经科学原理逐步诱导代理 “故意捆绑”。这一原则表明，随着更大的机构，一个人认为他们的行动（即，安全 抓握）在时间上与感觉结果（即，手套反馈）。我们的手套可供用户随时使用，并且 现在与定制的VR应用程序兼容， 以及定制的视觉和声音警报。我们假设VR中增强的反馈甚至会产生 更大的改进抓持性能比机载反馈单独。在目标2中，我们研究如何 SCI退伍军人可以在虚拟达到过程中学习更好的手臂肌肉控制。我们开发了一种"感官" 支具，为一只手臂提供等长阻力，以引出可以驱动运动的肌电图（EMG）模式 人接收来自VR的视觉反馈和来自支架的肌肉肌腱触觉反馈 在训练中。肌腱刺激可以引起调节肌肉激活模式的运动感觉。 VR反馈将提供有意识的运动训练提示，而振动反馈将潜意识地 基于聚类分析得出更明显的EMG模式。我们假设，促进不同的肌电图 通过最大化集群间距离实现的模式将提高到达触摸任务的性能。 每个目标中的抓握和伸展任务的性能都根据描述 时间、效率和运动的准确性。对于未来的MERIT提案，我们打算将这些 培训电动辅助设备更好操作的平台，包括混合动力（电机功率和功能） 肌肉的电刺激）手的外骨骼。重要的是，加强认知能力的概念 使用可穿戴技术，多感官反馈和虚拟现实进行运动的代理和学习， 体能训练将适用于所有形式的神经肌肉损伤，包括中风和脑外伤 除SCI外，我们的团队处于理想的位置，可以采取下一步的开发，翻译， 应用这种技术方法，它与康复研究和发展服务相匹配， 目标.

PROJECT SUMMARY Spinal cord injury (SCI) at the cervical level impairs hand function and the ability to engage the environment. As a result, the ability to perform activities of daily living can be severely compromised, directly leading to reduced quality of life. Physical therapy is still the primary pathway to restore functional abilities after SCI. The physical rehabilitation process requires commitment by the participant to achieve meaningful gains in function. Rehabilitation approaches that are cognitively engaging can facilitate greater commitment to practice and improved movement learning. We propose to develop innovative platforms that utilize virtual reality (VR) and instrumented wearables that enhance cognitive factors during motor learning of hand grasp and reach after SCI. These factors include greater sense of agency, or perception of control, and multi-sensory feedback. Sense of agency is implicated with greater movement control, and various sensory feedback modalities (visual, audio, and haptic) are proven effective in movement training. However, these factors are not well considered in traditional physical therapy approaches. We have developed two novel cognitive-based platforms for rehabilitating grasp and reach function. We propose to test each platform in Veterans with chronic SCI at the cervical level. In Aim 1, we will investigate how our first platform, the “cognition” glove, may improve functional grasp. This glove includes force and flex sensors that provide inputs to a machine learning algorithm trained to predict when secure grasp is achieved. The glove alerts the user of secure grasp through onboard sensory modules providing visual (LED), audio (beeper), and tactile (vibrator) feedback. During training, feedback is provided at gradually shorter time-intervals to progressively induce agency based on the neuroscience principle of ‘intentional binding’. This principle suggests that with greater agency, one perceives their action (i.e., secure grasp) is more coupled in time to a sensory consequence (i.e., glove feedback). Our glove is user-ready, and now has compatibility with customized VR applications to provide enhanced sensory feedback through engaging and customized visual and sound alerts. We hypothesize that enhanced feedback in VR will produce even greater improvements in grasp performance than onboard feedback alone. In Aim 2, we investigate how Veterans with SCI may learn greater arm muscle control during virtual reaching. We have developed a “sensory” brace that provides isometric resistance to one arm to elicit electromyography (EMG) patterns that can drive a virtual arm. The person receives visual feedback from VR and muscle tendon haptic feedback from the brace during training. Tendon stimulation can elicit movement sensations that modulate muscle activation patterns. The VR feedback will provide conscious movement training cues while vibration feedback will subconsciously elicit more distinct EMG patterns based on cluster analysis. We hypothesize that the promotion of distinct EMG patterns, achieved by maximizing inter-cluster distances, will improve performance of a reach-to-touch task. Performance of both the grasp and reach task in each Aim is assessed according to metrics that describe the timing, efficiency, and accuracy of movement. For a future MERIT proposal, we intend to integrate these platforms to train better operation of powered assistive devices, including a hybrid (motor power and functional electrical stimulation of muscle) exoskeleton of the hand. Importantly, the concept of strengthening cognitive agency and learning of movement using wearable technology, multi-sensory feedback, and virtual reality during physical training will be applicable to all forms of neuromuscular impairment, including stroke and traumatic brain injury in addition to SCI. Our team is ideally positioned to take the next steps in developing, translating, and applying this technological approach, which matches the Rehabilitation Research and Development Service’s goals.