Brain Neurophysiological Biomarkers of Functional Recovery in Stroke

描述（由申请人提供）： 背景/依据：中风仍然是我们退伍军人长期残疾的主要原因，绝大多数中风幸存者的上肢都有明显的运动障碍。中风幸存者在恢复的早期阶段恢复了大部分功能，这一过程可能是由潜在的神经生理学变化介导的。这些变化，但是，没有得到很好的理解，在自愿达到运动，但仍然是成功的康复至关重要。 脑电图（EEG）是了解中风恢复机制的几种方法之一，但基于EEG的研究主要依赖于与运动障碍和/或功能无关的整体脑电活动（即在静息状态下）。因此，我们的目标是更好地了解在关键的恢复期，在计划和执行意志达到的神经机制。我们打算使用EEG来识别脑源性生物标志物的变化，这些生物标志物反映了意图达到的时间和方向，并确定这些生物标志物在自然恢复过程中的变化，即，随着中风幸存者从亚急性期进展到慢性期。目的：我们的初步数据表明，它是可行的，以收集高质量的脑卒中幸存者，因为他们自愿执行运动任务。此外，我们检测到脑源性生物标志物，这些生物标志物可靠地编码了范围的各个方面，即，运动的预期方向和运动开始的时间。具体而言，第一项研究表明，当使用所提出的任务时，可以提取大脑衍生的生物标志物，其编码预期的到达方向。第二项研究显示，使用相同的任务，慢性中风患者表现出良好的特征与意志运动相关的皮质动力学。此外，使用这些生物标志物作为贝叶斯分类器的输入，可以在逐个试验的基础上稳健地预测移动意图的时间。因此，只有当更多的脑源性 生物标志物在时间和方向方面可用于该分类系统。 在这些初步研究中，我们的目标是表征最能预测中风幸存者意志性伸展运动的发生和方向的神经生理信号，并确定这些识别出的信号如何随着中风幸存者从亚急性期恢复到慢性期而变化。研究方法：将在亚急性恢复期（卒中后2-6周）和慢性期（> 6个月）对40例卒中患者的便利样本进行研究。在每次访视时，将要求参与者在InMotion平面机器人的两个目标中选择一个并到达其中一个，同时收集EEG数据。将提取与运动开始之前的时期相对应的EEG时间序列。这些数据将被处理以便提取大量功能性EEG度量（例如，频谱功率、相干性和运动相关皮层电位）。这些指标将被用作贝叶斯分类器的输入，以确定哪些生物标志物可靠地编码预期运动的意志定时和方向。最后，将确定这些生物标志物的变化作为恢复的函数。发现/结果：我们预测，我们将从众多的EEG指标中识别出一组生物标志物，这些指标将一致地指示意志运动的预期方向和时间。此外，我们预测这些生物标志物将随着时间和位置的恢复而变化，因此将提供恢复的机制见解以及与脑机接口相关的数据。现状：该项目目前处于规划和开发阶段。影响：这项研究将对指导未来旨在最大限度地恢复功能的工作产生重大影响。识别脑源性生物标志物编码中风幸存者的特定成分，将揭示自然恢复的神经生理学基础，并指导未来的康复策略。

DESCRIPTION (provided by applicant): Background/Rationale: Stroke remains the leading cause of long-term disability in our Veterans and the vast majority of stroke survivors experience significant motor impairments in the upper extremity. Survivors of stroke regain the most function during the early phases of recovery, a process that is likely mediated by changes in underlying neurophysiology. These changes, however, are not well understood in relation to voluntary reaching movements, but are nonetheless critical to successful rehabilitation. Electroencephalography (EEG) is one of several methods to understand the mechanisms underlying recovery in stroke, but EEG-based investigations have mainly relied on global brain electrical activity (i.e. during resting state) t correlate with motor impairment and/or function. Our goal, therefore, is to better understand the neural mechanisms during planning and execution of volitional reach across critical periods of recovery. We intend to use EEG to identify changes in brain-derived biomarkers that reflect the timing and direction of the intent to reach, and determine how these change over the course of natural recovery, i.e., as stroke survivors progress from the subacute to chronic phases. Objectives: Our preliminary data indicate that it is feasible to collect quality EEG in survivors o stroke as they volitionally perform motor tasks. Moreover, we detected brain-derived biomarkers that reliably encoded for aspects of reach, i.e., the intended direction to move and timing of movement onset. Specifically the first study demonstrated that when using the proposed task a brain-derived biomarker could be extracted which encoded for the intended direction of reach. The second study revealed that using this same task, a chronic stroke patient exhibited well-characterized cortical dynamics associated with volitional movement. Further, using these biomarkers as input to a Bayesian classifier robustly predicted the timing of the intent to move on a trial- by-trial basis. Thus, the classification will only improve when many more brain-derived biomarkers are available to this classification system with regard to timing and direction. In lin with these preliminary studies, our objective is to characterize the neurophysiological signals that best predict the onset and direction of volitional reaching movements in survivors of stroke, and determine how these identified signals change as stroke survivors recover from subacute to chronic stages. Methods: A convenience sample of forty stroke patients will be studied in the subacute phase of recovery (2-6 weeks post-stroke) and followed to the chronic stage (> 6 months.) At each visit, participants will be asked to choose between, and reach to, one of two targets in an InMotion planar robot while EEG data are collected concurrently. The EEG time series corresponding to the period prior to movement onset will be extracted. These data will be processed in order to extract a multitude of functional EEG metrics (e.g., spectral power, coherence, and movement related cortical potentials.) These metrics will be used as inputs into a Bayesian classifier to determine which biomarkers reliably encode for volitional timing and direction of intended movement. Finally, the changes in these biomarkers as a function of recovery will be determined. Findings/Results: We predict that we will identify a subset of biomarkers from the multitude of EEG metrics that will consistently indicate the intended direction and timing of volitional movement. Further, we predict that these biomarkers will change as a function of recovery with regard to timing and location, and will therefore provide both mechanistic insights to recovery as well as data relevant to brain-computer interfaces. Status: This project is currently in the planning and development stage. Impact: This research will have a significant impact on steering future work aimed at maximizing functional recovery. Identifying the brain-derived biomarkers encoding for specific components of reach in stroke survivors will reveal the neurophysiology underlying natural recovery and guide future rehabilitation strategies.