Abstract: A new data processing approach of fMRI which combines Statistical Parametric Mapping（SPM），Independent Component Analysis（ICA） and image fusion is presented.And it has good performance in extracting the brain activation patterns.Firstly，the data are obtained from the experiment of block design in various hand power forces，which associate with motor cortex.Then the preprocessing stage proceeds and the brain functional information is gotten from the SPM and ICA methods separately.Finally，the new image fusion method based upon the Principal Component Analysis（PCA） is used to process the results from the two methods mentioned above.The new approach is a way to localize the brain functional area effectively and complementary in two typical systems.

Key words: functional Magnetic Resonance Imaging（fMRI）, motor cortex, Statistical Parametric Mapping（SPM）, Independent Component Analysis（ICA）, image fusion

摘要： 提出了一种新的fMRI数据处理方法，融合了统计参数图（SPM）、独立成分分析(ICA)所提取的特征信息，实现脑功能激活区的准确提取。首先通过时段设计实验获取了反应不同握力条件下手运动相关皮层活动的fMRI数据，并且进行相应的预处理；然后采用SPM和ICA方法分别提取脑功能信息；研究了一种基于主成分分析的图像融合算法。最后，应用图像融合算法对SPM和ICA方法分别提取的脑功能信息进行融合。结果表明，该方法弥补了SPM和ICA两种方式的不足，是一种进行功能区定位更加有效的方法。

关键词: 功能磁共振成像（fMRI）, 运动皮层, 统计参数图（SPM）, 独立成分分析（ICA）, 图像融合