脑卒中后运动环路重建的“宏观-介观”多模态高维特征分析及神经编码机制研究

Stroke has a high incidence and disability rate. Most of stroke patients have different degrees of motor dysfunction which affect activities of daily living. Brain plasticity is the main intrinsic mechanism of functional recovery after central nervous system injury. And the neuroplasticity characteristics caused by different types of stroke with different lesion location are different. Brain plasticity changes depend on brain injury location, which influence recovery of motor function. Routine rehabilitation training and neuromodulation have poor clinical effect. Because there is no concern about different strokes of the specific neural circuit damage result in different neuroplasticity characteristics and different functional outcome, precise intervention cannot be performed. It is important to explore the key adaptive brain plasticity changes of different types of stroke. We attempt to establish a rat model of focal cerebral ischemia with different lesion location (cortical and/or subcortical), divided into natural recovery and functional training groups, using behavioral assessments to evaluate motor function recovery. And we use the neuroimaging and artificial intelligence to find the key brain region from a lot of dynamic brain plasticity features. Furthermore, we investigate the discharge properties of neurons in target brain region to decode neural information. We explore the macroscopic and mesoscopic mechanism of motor circuit reconstruction after stroke based on multi-modal and high-dimensional feature analysis and decoding neural information, which can provide a solid theoretical basis and objective research evidence for the optimization of rehabilitation strategies and the development of new intervention methods for motor rehabilitation of stroke.

脑卒中发病率高，致残率高，运动障碍极多见，严重影响患者日常生活能力。脑重塑是脑卒中后功能恢复的主要内在机制，不同类型脑卒中所引发的重塑特征也不同。常规康复训练包括神经调控疗效欠佳，是因为没有关注不同脑卒中其特定神经环路受损和重塑特征不同，则功能恢复路径和结局不同，而不能进行精准干预。本课题拟建立不同脑区（皮质、皮质下、皮质合并皮质下）缺血大鼠模型，纵向行为学观察运动功能恢复特征，并结合多模态脑功能影像及人工智能算法，首先在宏观层面从高维脑重塑特征中定位出影响功能预后的关键脑区，进一步在介观层面对该关键区域的神经元在体放电活动进行分析，从而在宏观和介观水平进行运动环路重建的脑重塑规律及对应的神经编码机制研究，有利于建立基于此机制的、靶点明确的精准神经调控，为脑卒中运动障碍康复策略的优化及新干预方法的开发提供坚实的理论和客观证据支持。

脑卒中发病率高，致残率高，运动障碍极多见，严重影响患者日常生活能力。脑重塑是脑卒中后功能恢复的主要内在机制，不同类型脑卒中所引发的重塑特征也不同。常规康复训练包括神经调控疗效欠佳，是因为没有关注不同脑卒中其特定神经环路受损和重塑特征不同，则功能恢复路径和结局不同，而不能进行精准干预。本课题建立不同脑区（皮质和/或皮质下）缺血大鼠模型，纵向行为学观察运动功能恢复特征，并结合多模态脑功能影像及人工智能算法，首先在宏观层面从高维脑重塑特征中定位出影响功能预后的关键脑区，进一步在介观层面对该关键区域的神经元在体放电活动进行分析，从而在宏观和介观水平进行运动环路重建的脑重塑规律及对应的神经编码机制研究。结果发现脑损伤后运动执行-反馈调节环路连接受损。将M1-纹状体环路的功能连接值、结构连接值作为特征进行功能恢复好坏的分类，支持向量机（Support Vector Machine，SVM）分类算法结果表明M1-纹状体环路的连接属性可以在个体水平较准确区分功能恢复好与差，提示脑缺血后运动功能的恢复可能有赖于M1-纹状体环路的高连接强度。在体多通道记录中M1和背外侧纹状体（Dorsolateral Striatum，DLS）的神经振荡活动的同步性分析结果显示与功能恢复不好的大鼠比较，功能恢复好的大鼠M1和DLS的LFP相关性整体水平明显增加，在 Alpha显著增加。本研究结果提示M1-纹状体环路是脑卒中运动康复的关键靶向环路。关键神经环路的重建是脑卒中后肢体运动功能恢复的核心机制。针对性地进行脑卒中运动执行-反馈调节环路的个体化映射和精准调控，可能突破目前脑卒中运动康复疗效的瓶颈。