mask制作问题

老师们,好。

近期看到一篇文章: Resting-state functional connectivity patterns predict Chinese word reading competency. PloS one, 7(9), e44848
里面提及的mask制作,此文是这样说:
The  RSFC  analysis  was  performed  using  the  Resting-State fMRI Data Analysis Toolkit (REST) [25]. We conducted RSFC
computation and multiple comparison correction within a grey matter  (GM)  mask,  which  was  generated  using  the  following procedure. First, voxels in the SPM5 GM mask with a probability higher  than  0.4  were  included.  Second,  voxels  within  the cerebellar regions were discarded using the Automated Anatomical Labeling (AAL) template [26], given the signal distortion in the cerebellum. In total, there were 36, 272 voxels(979344mm3) in the GM mask.
========
作者拿灰质模板做mask;他设定:voxels 高于0.4的比率;并且除去了小脑。

看完视频后,我初步认为是这样:
在underlay里选择:spm8\apriori\grey.nii;在overlay里选择:REST_V1.8_130615\Template\AAL_61x73x61.nii;之后在Set a range for thresholding里选择0,90 {即把91-116后面的小脑部位的都去除了};不知这样做是否正确?

但好像需选阈值诶,因为当我点击“save clusters”,出现提示“Set cluster Connectivity criterion first!”

之后,在Threshold by cluster size里,分别设定:36272;979344;5;就可以save as clusters。但有个问题就是:保存图像非常小
再次打开该保存图像,出错了:

Error in ==> rest_sliceviewer>LoadOverlay at 3493
    if size(AllVolume,4)~=1

Error in ==> rest_sliceviewer at 449
                REST_SliceViewer_Cfg.Config(theCardinal)
                =LoadOverlay(REST_SliceViewer_Cfg.Config(theCardinal), theNewOverlay);

Error in ==> rest_sliceviewer at 558
                rest_sliceviewer('ChangeOverlay', AFigHandle);

??? Error while evaluating uicontrol Callback

所以,不知这个具体参数是怎么设置0.4参数是怎么设置的,和后面提及的 there were 36, 272 voxels in the GM mask.是如何设定的?

还有一个问题就是我一加spm8\apriori\grey.nii模板,图像就变小了,是否是需要重采样呢?

先谢谢了。

1. 在underlay里选择:spm8\apriori\grey.nii;在overlay里选择:REST_V1.8_130615\Template\AAL_61x73x61.nii;之后在Set a range for thresholding里选择0,90 {即把91-116后面的小脑部位的都去除了};不知这样做是否正确?

--不太正确。因为AAL的脑区不一定完全覆盖了你想要的所有区域。

2. 可以先卡灰质MASK。然后在AAL里面做一个小脑MASK(不完全确定这是正确的),即卡>90的脑区, save clusterS.

3. i1.*(i2==0)

4. 可以写信给作者确认他们的做法。