独立分量分析在功能磁共振成像信号功能活动区检测中的应用

主要讨论独立分量分析(ICA)在功能磁共振成像(fMRI)信号功能区检测中的应用。fMRI利用血氧水平依赖(BOLD)效应成像，根据大脑神经元兴奋后局部血氧饱和度增高的原理间接显示神经元活动。假设fMRI信号中包含反映血氧饱和度事件相关的信号、生理噪声和仪器产生的随机噪声等独立分量，首先对fMRI信号进行去噪、配准等预处理，然后利用fastlCA算法对独立分量进行分离，有效抑制噪声对功能区检测的影响，利用相关原理检测出fMRI信号的功能活动区。

fMRI Signal Analysis Using ICA Algorithm

To discuss the application of Independent Component Analysis (ICA) to the detection of Functional Magnetic Resonance Imaging (fMRI) signal. When a brain area is being used, arterial oxygenated blood will be redistributed and increased in this area. fMRI is a new imaging technique based on the blood oxygenation level dependent (BOLD) effect, it can indicate the activity of neuron indirectly. It is well known that fMRI signals comprise some independent components, such as the event-related signal which reflect the saturation of blood oxygenation, the physiological noise, the random noise caused by the apparatus and so on. Firstly, some pre-processing(such as noise smoothing and image registration), to the fMRI signals is done. Then, the fastICA algorithm to decompose the independent components is used, which is a very useful way to restrain the impact caused by noise. Finally, the true functional activity by computing the linear correlation coefficients between decomposed time-series of fMRI signals and stimulating reference function can be detected.