COLLABORATIVE RESEARCH: ARWED - AUGMENTED PERCEPTION FOR UPPER-LIMB REHABILITATION

项目负责人：Banerjee， Amarnath/Perez Gracia, Alba/Robson， nina提案编号：1403502/1403688/1404011标题：合作研究：ARWED -上肢康复增强感知更广泛的意义和重要性训练和再训练因中风而丧失运动能力的个体的运动是一项具有挑战性的任务。在许多情况下，经过长时间的训练，只能取得部分成功。鉴于中风后患者恢复的局限性，开发更好的再训练工具和方法势在必行。尽管存在一些训练假设，例如需要精确地重复训练模式，但解决方案仍有待发现。克服其中一些障碍是本提案的目标。最近的研究表明，人类的行动感知和行动执行之间存在直接联系。有一些证据表明，对人类行为的视觉观察对观察者的运动有影响。当运动由非人类设备（如机器人）执行时，这种启动效应会减弱或不存在。该提案旨在开发和测试一种用于人类手臂训练的新型可穿戴系统。该系统将允许用户将设备感知为上肢的一部分（因此关闭感知/行动循环）。基于观察学习结果的训练方案将与ARWED一起实施，以使其成为中风后患者运动恢复的康复训练设备。技术描述智力优势集中在开发动作观察和动作执行之间的联系，以便制定训练方案以促进中风后的康复。目前，利用虚拟现实重新学习生物运动的系统有限。增强可穿戴系统（ARWED）的发展需要解决计算机视觉/建模和机器人运动学映射方面的几个挑战。测试设备对启动感知-行动效应的有效性将需要人体运动学、信号分析、虚拟现实、机器人故障恢复理论和康复等领域的综合专业知识。因此，这个多学科合作研究团队的理论贡献将不仅在医学领域内，而且在上述研究领域内推进知识和理解。更广泛的影响是基于待开发设备的特点，在运动障碍患者的培训和再培训方面具有扩展的能力。有效的设计和操作将使所提出的ARWED系统成为一种可靠的解决方案，将广泛应用于康复、运动治疗和康复等医疗专业人员。拟议的研究将为任何程度的肢体限制患者的训练和身体康复提供有效的工具，范围从患有部分运动能力丧失的个体到由于中风，出生缺陷或事故导致的严重行动限制。运动学习理论表明，在练习过程中减少反馈有利于运动技能训练的长期保留。与基于机器人的中风训练相比，观察学习可能在向adl转移方面提供更大的好处。因此，对这些患者来说，一个成功的指标是将训练从未受影响的肢体转移到受影响的肢体，反之亦然。ARWED系统有望通过在认知、远程操作和虚拟现实的实验和教育工作中实现该设备，显著推进工程和科学知识的基础。医学界也将从ARWED的发展中受益，因为它进一步了解了使用技术训练如何提高不同人群运动控制的恢复，同时提供了一种可能比目前可用的更有效的新干预措施。研究团队的合作性质将导致工程和运动机能学领域的本科生和研究生的教育。从长远来看，该提案的结果将促进想要进入物理和职业治疗专业的学生与那些想要设计和开发可以帮助人类神经损伤后恢复的机械设备的学生之间的交流。

PI: Banerjee, Amarnath/Perez Gracia, Alba/Robson, NinaProposal Number: 1403502/1403688/1404011Title: Collaborative Research: ARWED - Augmented Perception for Upper-Limb RehabilitationBroader Significance & Importance Training and retraining the movement of individuals suffering loss of motor ability due to stroke is a challenging task. In many cases, only partial success is accomplished after long training sessions. Given the limitations of recovery for post-stroke patients, it is imperative that better tools and methods for retraining be developed. Even though several training hypotheses exist, such as need for an exact repetition of the training pattern, the solutions are still to be found. Overcoming some of these obstacles is the goal of this proposal. Recent studies show a direct link between human action perception and action execution. There is some evidence that the visual observation of human actions has an effect on the movement of the observer. This priming effect is reduced, or not present, when the motion is executed by a non-human device, such as a robot. This proposal aims to develop and test a novel wearable system for the training of the human arm. The system will allow the user to perceive the device as part of their upper limb (hence closing the perception/action loop). Training protocols based on observational learning findings will be implemented with the ARWED in order to develop it as rehabilitation training device for motor recovery in post-stroke patients.Technical description The intellectual merit centers on exploiting the link between action-observation and action execution in order to develop training protocols to facilitate rehabilitation following stroke. Currently, there are limited systems that utilize virtual reality in the relearning of biological movements. The proposed development of augmented wearable system (ARWED) requires solving several challenges in computer vision/modeling, and robot kinematic mapping. Testing the effectiveness of the device on priming the perception-action effect will require combined expertise from the areas of human kinematics, signal analysis, virtual reality, robotic fault recovery theory and rehabilitation. Therefore, the theoretical contributions emerging from this multidisciplinary collaborative research team will advance knowledge and understanding not only within the medical field, but also across the above-mentioned research areas. The broader impact is based on the characteristics of the device to be developed, with an expanded ability in the training and re-training of patients with motor disabilities. Efficient design and manipulation will make the proposed ARWED system a reliable solution, which will be broadly utilized by medical professionals working in rehabilitation, sports therapy and convalescence. The proposed research will provide effective tools for the training and physical rehabilitation of patients with limb limitations at any scale, ranging from individuals suffering partial loss of motor ability to those with severe limitations in mobility due to strokes, birth defects or accidents. Motor learning theory shows that reducing feedback during practice benefits long term retention of motor skill training. Observational learning may offer greater benefits regarding transfer to ADLs, in comparison to robotic-based stroke training. Thus, a success indicator for these patients would be the beneficial transfer of training from the un-affected to affected limb or vice-versa. The ARWED system is expected to advance significantly the fundamentals of engineering and scientific knowledge, by implementing the device in experimental and educational work on cognition, telemanipulation and virtual reality. The medical community would also benefit from the development of the ARWED by furthering the understanding of how training using technology may enhance the recovery of motor control in diverse populations while providing a novel intervention that may prove more effective than what is currently available. The collaborative nature of the research team will lead to the education of undergraduate and graduate students in the areas of Engineering and Kinesiology. In the long run, the outcomes of the proposal will facilitate the communication between students that want to enter Physical and Occupational Therapy professions and those that want to design and develop mechanical devices that can aid in human recovery following neurological injury.