摘要
人在想象但未实施肢体或其他身体部位动作时,与该动作相关的大脑运动皮层区域会发生与该动作实施时相似的电生理响应,称为想象动作电位。想象动作电位的提取与分类是脑-机接口(BCI)技术的关键和难点。本文分别介绍了想象动作电位的时频分析、复杂度分析、相位耦合测量、多通道线性描述符、多维统计分析等特征提取方法和线性判别分析、人工神经网络、支持向量机等分类算法,以供BCI系统设计与研究时参考。
Specific electrophysiological response will occur in the related region of brain motor cortex when human pursues limbs or other body parts in imagination, which is similar to the phenomenon caused by real motor of the corresponding body action. Extracting and classifying the motor imaginary potential are the key and difficult points in brain-computer interface. In this article, some feature extraction methods such as time-frequency analysis, complexity analysis, phase coupling measurement, multi-dimensional statistical analysis and etc. , and some classification means such as linear discrimination analysis, artificial neural network, support vector machine and etc, are introduced as the references in design and study of brain-computer interface system.
出处
《仪器仪表学报》
EI
CAS
CSCD
北大核心
2008年第8期1772-1778,共7页
Chinese Journal of Scientific Instrument
基金
"十一五"国家高技术研究发展计划(863计划
2007AA04Z236)
国家自然科学基金(60501005)
天津市科技支撑计划重点项目(07ZCKFSF01300)资助
关键词
脑-机接VI
想象动作电位
特征提取
分类算法
brain-computer interface （BCI）
motor imaginary potential
feature extraction
classification
