伴冻结步态帕金森病患者治疗前后的脚桥核（PPN）多模态MR影像学表征研究

Freezing of gait (FOG) is a significant source of disability in Parkinson’s disease (PD). According to our previous studies, abnormal anatomical features and functional activity of Pedunculopontine nucleus (PPN) contributes to the mechanism of FOG. Unfortunately, the PPN is poorly delineated structurally in vivo. The mechanism of electrophysiologically accounting for FOG is still unclear. In the proposed study, we will: .① construct the PPN featured neural network study framework based on the results of Human Connectome Project (HCP) as well as the supervoxel-based algorithm for the whole brain, then investigate with novel data fusion techniques and motion correction methods to combine data from structural MRI, diffusion tensor imaging (DTI), myelin water imaging, and fMRI data; ② identify a robust, accurate high-resolution PPN biomarker in vivo based on the fused multi-model MR neural imaging by using discriminant analysis in pattern recognition technology, as well as illustrate the neurobiological features of the PPN biomarker; ③investigate the changes in structural and functional activities of such PPN biomarker after Ambulosono exercise intervention and illustrate its ability in assessing the severity of disease by correlate to standard motor rating scales and predicting response to treatment. If successful, the project will provide a practical imaging tool to improve patient selection for those with gait impairment, which remains largely refractory to both pharmacological and surgical therapies. This study would also help to further explore the mechanism of abnormal neurobiological activity of PPN in FOG of PD.

冻结步态（FOG）是帕金森病（PD）患者致残的重要原因之一。申请人在前期研究发现，脚桥核（PPN）形态与功能的异常可能是造成FOG的关键，并进一步提出：FOG程度可由PPN影像学表征的神经生理特征客观评估并预测对治疗的反应性和敏感性。然而，人体PPN的精细的影像学形态特征及其造成FOG的神经生理机制尚不明确。本研究将①基于人脑连接组计划研究成果，结合超体素聚类算法，搭建PPN神经网络研究框架，并创新PPN高精度影像解剖、扩散张量成像、髓鞘水成像、静息态功能成像等多模态MR数据运动校正方法及数据配准技术；②阐释多模态数据间相互映射关系，利用模式识别技术中的判别分析确立高精度的PPN影像学表征，并揭示其基线条件下的神经生理特征；③探索步态训练前后，PPN影像学表征神经生理特征的纵向演变规律，阐释其疾病评估及疗效预测能力，充分揭示PPN影像学表征神经生理活动在FOG发展与转归中的作用机制。

帕金森病中晚期患者出现冻结步态（FOG）的发生率约为63%，会引发跌倒、致残、致死等严重后果。明确FOG启动的脑运动网络变化机制对减少FOG导致的不良后果至关重要。国际及团队前期静息态fMRI研究发现，FOG以脚桥核与脑网络连接减弱为重要特征，但FOG启动的脑运动网络机制不清。基于此，本项目提出通过多模态fMRI探索“FOG运动网络”在启动中关键作用新思路。拟开展①采集PD患者多模态影像学数据提炼以PPN为种子点或连接节点的特征脑区，阐释基线期神经生理特性；②使用Ambulosono步态训练法治疗3个月后，再次获取多模态fMRI信号，分析治疗后脑区变化特征；提取治疗前后与启动密切相关的脑激活图，明确“FOG运动网络”；③应用计算机化步态毯评估治疗前、后的客观步态并分析与运动网络的相关性以评价疗效。本项目为有效抑制FOG的发生提供科学的神经影像学依据。