NRI:BMI Control of a Therapeutic Exoskeleton

描述（由申请人提供）：本研究旨在通过利用患者意图和损伤实时评估的益处，加速卒中后机器人康复的开发、有效性和使用。验证将使用位于德克萨斯州休斯顿的康复研究所（TIRR）的MAHI EXO-II外骨骼机器人进行。机器人康复是脑卒中患者感觉运动训练的有效平台。机器人设备能够准确定位受损肢体，同时提供辅助和阻力，并收集可用于表征患者运动质量的运动数据。MAHI EXO-II是一种物理人机接口，将通过非侵入性脑机接口（BMI）进行增强，以将患者纳入控制回路，从而使治疗“活跃”，并使患者参与康复任务中的各种损伤严重程度。这种方法利用了在机器人康复的其他临床研究中观察到的患者运动启动意图的已知益处以及BMI使用对皮质可塑性的有益影响。机器人措施 根据MAHI EXO-II上的传感器和BMI获得的实时数据，运动障碍将驱动适应个人能力的患者特定治疗课程，机器人提供帮助或适当地挑战参与者，以最大限度地提高康复效果。机器人康复中的按需辅助范例已被证明是有效的;然而，这种范例是被动的，并且由尚未得到充分验证和验证的性能指标驱动。此外，在治疗过程中，高强度的练习和持续的“挑战”可以改善康复效果。主要贡献：1）使最先进的EEG-BMI方法适应中风患者，并开发用于控制MAHI EXO-II的BMI，其将a）增加上肢功能，B）促进对大脑可塑性的理解，以及c）创新康复; 2）确定运动损伤的适当机器人和电生理测量以及用于患者特定治疗的相关控制算法;和3）初步研究中的临床验证，以评估所提出的方法。

DESCRIPTION (provided by applicant): This research aims to accelerate the development, efficacy and use of robotic rehabilitation after stroke by capitalizing on the benefits of patient intent and real-time assessment of impairment. Validation will occur using the MAHI EXO-II exoskeleton robot at The Institute for Rehabilitation and Research (TIRR) in Houston, Texas. Robotic rehabilitation is an effective platform for sensorimotor training in stroke patients. A robotic device enables accurate positioning of the impaired limb while simultaneously providing assistance & resistance forces and collection of motion data that can be used to characterize the quality of the patient's movements. The MAHI EXO-II, a physical human-robot interface, will be augmented with a non-invasive brain-machine interface (BMI) to include the patient in the control loop, thereby making the therapy 'active' and engaging patients across a broad spectrum of impairment severity in the rehabilitation tasks. This approach capitalizes on the known benefits of patient intent in movement initiation observed in other clinical studies of robotic rehabilitation and on the beneficial effects of BMI use on cortical plasticity. Robotic measures of motor impairment, derived from real-time data acquired from sensors on the MAHI EXO-II and from the BMI, will drive patient-specific therapy sessions adapted to the capabilities ofthe individual, with the robot providing assistance or challenging the participant as appropriate, in order to maximize rehabilitation outcomes. Assist-as-needed paradigms in robotic rehabilitation have been shown to be efficacious; however, such paradigms are passive and driven by performance metrics that have not been sufficiently validated and verified. Additionally, intense practice and continual 'challenge' during therapy is known to improve rehabilitation outcomes. Key contributions: 1) Adapting most advanced EEG- BMI methods to stroke patients and developing a BMI for the control of the MAHI EXO-II that will a) increase upper limb function, b) advance understanding of brain plasticity, and c) innovate rehabilitation; 2) Determining appropriate robotic and electrophysiological measures of motor impairment and associated control algorithms for patient-specific therapy; and 3) Clinical validation in pilot studies to evaluate the proposed approach.