目的:内源信号光学成像分析方法通常使用空白本底或交混本底作为基准图像,然后将刺激条件下获得的皮层图像与基准图像相除或者相减,得到刺激条件下的皮层功能图。但是空白本底和交混本底本身是含有皮层活动信息的,二者也存在结构上的差异,说明本底方法本身可能会带来计算误差。本文的目的是探索一种不使用本底基准图像的内源信号分析方法。方法:将主成分分析与空间独立成分分析相结合,不使用本底相减方法,以自行记录的麻醉猫初级视皮层对正弦光栅刺激反应所得到的成像图像为数据源,直接提取方位功能图。结果:我们提出的基于空间独立成分分析的内源信号分析方法同样可以提取出效果较好的方位功能图。将我们得到的方位功能图与传统本底方法得到的方位功能图的轮廓进行比较,两种分析方法得到的功能构筑信息基本相同。结论:我们提出的基于空间独立成分分析的内源信号提取方法可以有效直接从原始图像提取出方位功能图。
Objective: For the optical imaging of ina-insic signals, genuine blank or cocktail blank was usually used as baseline image, which would be subtracted from the cortical image under stimuli conditions to get functional image. However, the cortical activities in the baseline images and the different organization between these two blank images would introduce certain errors. The objective of this paper is to explore a new method rather than traditional subtracting ones. Methods: Spatial independent component analysis combined with principal component analysis was applied to extract cortical orientation functional images directly t~om recording images collected from anesthetized cat primary visual cortex. Results: Orientation functional images extracted by the method we proposed had almost the same cortical functional architecture compared with that by traditional blank method. Conclusions: Spatial independent component analysis we proposed could extracte the intrinsic optical imaging signals directly and effectively.