相干脑电源成像方法研究

【作者】 张军鹏；

【导师】 尧德中；

【作者基本信息】 电子科技大学 ， 生物医学工程， 2005， 硕士

【摘要】 脑电源的定位在临床、康复和认知研究中均占有重要地位。脑电源大致可以分为相干源和非相干源两类。脑电源的相干性被认为是皮层功能耦合的表现,定位相干源具有重要意义。受文献[9]的启发,我们发展了一种定位相干源的方法。这种方法的原理是,通过把各个通道之间的互相关系数矩阵进行基于最小二乘法的分解,得到各个通道和所有通道之间加权和的相关系数,然后用该相关系数的空间分布来成像。计算机模拟实验证明,相比传统的功率成像方法,该方法对相干但能量水平不同的源组合有较好的敏感性以及抗噪能力。真实数据测试表明,这种方法成像敏感度高,能对多个相干源准确定位,且对噪声是稳健的,因而有明显的理论意义和应用价值。更多还原

【Abstract】 The localization of brain sources plays an important role on clinical diagnosis, cognitive research and rehabilitation. Brain sources can be classified into coherent sources and incoherent ones. Coherence between signals of sensors covering different scalp areas is commonly taken as a measure of functional coupling. Motivated by literature [9], we develop a method of mapping coherent brain sources. The principle of this method is decomposition of correlation coefficient between different channels of EEG. This techniques decomposes, in the terms of the least squares , the mutual correlation coefficient(CC ) matrix into individual CC between any channel and the weighted sum of all the channels, and then utilizes the spatial distribution of the individual CC as the imaging map. Computer simulation test show that compared with power mapping, this algorithm is sensitive to the combination of coherent sources with different energy levels. Real VEP data tests show that it has high sensitivity in localizing 2D coherent sources, it can accurately localize multiple coherent sources, and it is also robust to noise and to referrence. Thus, it is of theoretical significance and of practical value.更多还原