ADAPTATION TO NOVEL FORCE FIELDS IN HEMIPARETIC STROKE

机器人操纵器被视为潜在的重要康复工具，因为它们可以在受试者移动时施加各种力（力场）。它们也可以重复使用，同时监测和记录受试者的表现，允许比目前人类治疗师更密集和更长时间的治疗。仍然存在的问题是，哪种类型的力场最适合帮助偏瘫患者恢复运动功能，以及中风幸存者应该如何在这种力场中接受训练。当正常受试者暴露在一个新的力场中，以一种系统的方式干扰他们的运动时，他们会逐渐适应，因此当干扰力场出乎意料地消失时，他们会表现出特有的后遗症。初始适应和后效都是在被试没有意识到适应过程的情况下发生的。由于后遗症可以通过神经-肌肉-骨骼系统的动态模型来预测，我计划使用动态建模技术作为设计力的工具，从而在表现出刻板印象的运动错误的偏瘫中风患者中产生理想的后遗症。目前有证据表明，当偏瘫中风患者经历辅助力场时，他们的康复得到加强。其他证据表明，在伸展运动过程中针对阻力场进行训练，也可能有助于患者在阻力场移除后完成理想的运动。该项目的主要目标是确定哪种类型的场-辅助或电阻-最适合获得理想的康复效果。为此，我将使用一个可编程的双连杆机器人，它在到达水平平面运动时施加力。我打算首先确定后遗症能持续到什么程度。然后，我将开发和测试一个计算框架，设计辅助和阻力力场。最后将决定是否辅助或阻力力场是最适合调解所需的代偿作用的运动偏瘫幸存者。这些实验结果将有利于机器人在更复杂的康复环境中的开发和应用，并为机器人辅助康复领域更长期的临床试验提供明确的方向。

Robotic manipulators are seen as a potentially important instrument for rehabilitation because they can be programmed to exert a variety of forces (force fields) while the subject moves. They can also be used repetitively, while monitoring and recording subject's performance, allowing more intensive and prolonged treatment than is currently feasible with a human therapist. The question that remains is what type of force field is best suited for assisting in the recovery of motor function in hemiparetic human subjects, and how should the stroke survivor be trained in such a force field. When normal subjects are exposed to a novel force field that disturbs their motion in a systematic fashion, they adapt progressively so that they exhibit characteristic after-effects when the disturbing force field is unexpectedly removed. Both the initial adaptation and the after-effects occur without subjects being aware of the adaptive process. Since after-effects can be predicted by dynamic models of the neuro-musculoskeletal system, I plan to use dynamic modeling techniques as a tool for designing forces that will yield desirable after- effects in hemiparetic stroke patients that show stereotypical errors in movement. Currently there is evidence that hemiparetic stroke patients recovery is enhanced when they experience assistive force fields. Other evidence suggests that training against resistive force fields during reaching movements may also help the patient perform a desirable motion once the resistive field is removed. The main goal of this project is to determine which type of field -- assistive or resistive -- is best suited to attaining desirable rehabilitation outcomes. To this end I will use a programmable two-link robot that exerts forces during reaching movements in the horizontal plane. I plan to determine first to what extent after-affects can be made to persist. I will then develop and test a computational framework that designs assistive and resistive force fields. Finally will determine whether assistive or resistive force fields are best suited for mediating the desired compensatory effects on movements in hemiparetic stroke survivors. The results of these experiments should favor the development and application of more sophisticated rehabilitation environments using robots and provide a clear direction for more long-term clinical trials in the area of robot assisted rehabilitation.