基于脑电网络分析的脑卒中患者言语认知康复评估方法研究

项目名称： 基于脑电网络分析的脑卒中患者言语认知康复评估方法研究

项目编号： No.61501518

项目类型： 青年科学基金项目

立项/批准年度： 2016

项目学科： 无线电电子学、电信技术

项目作者： 应俊

作者单位： 中国人民解放军总医院

项目金额： 21万元

中文摘要： 言语认知功能评估是脑卒中康复的关键诊断方法，现行的行为学认知量表法存在人员专业性限制、主观因素影响、语言依赖性以及早期敏感度低等问题，制约了对症治疗与康复效果。本项目旨在提出一种脑卒中言语认知的客观量化评估方法，研究从患者的听性稳态反应的脑功能网络入手，分析脑功能网络拓扑属性的异常变化，提取聚类系数、最短路径、全局效率、信息流增益等特性参数对认知功能缺陷进行量化评估，并在此基础上，完成诱发脑电检测与计算的软硬件平台，实现在线分析评估。本项目提出以大脑活动协同性与因果性来分析认知缺损的研究思路，探讨基于网络拓扑分析来解决认知量化难题的研究模式，探索脑电网络实时计算与分析的技术创新，最终在系统水平上为揭示大脑疾病的病理生理机制与认知信息处理机制提供新的启示。

中文关键词： 信号分析；诱发电位；特征提取；信号检测

英文摘要： Functional assessment of speech perception is a key diagnostic method for stroke rehabilitation. There are limites for current behavioral cognitive scaling method such as professional restrictions, negative impact of subjective factors, language-dependent and low diagnosis sensitivity for early stage, which restrict the targeted treatment and effectiveness of rehabilitation. This project proposes an objective quantitative assessment methods of speech perception for stroke rehabilitation. We intend to research on brain functional network of Auditory Steady State Response, investigate the abnormal changes in topology attributes of brain functional network,and extract the characteristic parameters to quantitative assessment of cognitive dysfunction, including clustering coefficient, shortest path length, Eglob and flow gain etc.. Based on this study, a hardware and software platform to online acquisition and analysis of evoked potentials is developed. We research conitive deficits on analysing interoperability and causality of brain activity, solve the problem of quantified cognitive assessment by topology analysis, achieve a technology innovation on real-time analysis of EEG networks. Eventually, gain new inspiration with macro analysis on the pathophysiology and cognitive information processing mechanisms of the brain disease.

英文关键词： Signal analysis;Evoked potentials;Feature extraction;Signal detection