Submitted by Chen0075 on Wed, 02/29/2012 - 21:51
老师，
您好！
最近有一篇稿件被审稿人提了几个问题，都是有关REST软件的，总结如下，能否帮忙回答一下，十分十分感谢！！！
1. 我用的是REST slice view 里的FDR检验（one-tail）,一个统计学者提问说这里的FDR检验采用的类型是什么？：if FDR was used, what FDR method was it (Benjamini & Hochberg, Benjamini & Yekutieli, etc)。
2. 做完two-sample t-test, ANOVA检验，或者correlation analysis后，REST的报告里只会显示peak t-value, peak F-value, 和 peak r-value; 怎样把t值，F值，和r值 转换为相应的Z值？转换后得到的Z值是什么含义？
再次感谢！！！
Submitted by YAN Chao-Gan on Fri, 03/02/2012 - 05:20 Permalink
Re
1. http://en.wikipedia.org/wiki/False_discovery_rate#Independent_tests
Controlling procedures: Independent tests.
2. 参照：http://restfmri.net/forum/node/852#comment-1349
Submitted by Chen0075 on Fri, 03/02/2012 - 16:04 Permalink
谢谢！
谢谢！老师的回答！我也下了y_TFRtoZ程序，但是应该怎么在Metlab里面调用它呢，应该输入什么命令？我已经把y_TFRtoZ.m放在当前目录下了，在command里面输入y_TFRtoZ,回车，但是会出现:Input argument "ImgFile" is undefined.
而changbvalue那个程序倒是能运行得很正常！
能否解答一下，谢谢！
Submitted by YAN Chao-Gan on Fri, 03/02/2012 - 23:36 Permalink
Re
help y_GRF_Threshold
function [Data_Corrected]=y_GRF_Threshold(StatsImgFile,VoxelPThreshold,IsTwoTailed,ClusterPThreshold,OutputName,MaskFile,Flag,Df1,Df2)
Function to perform Gaussian Random Field theory multiple comparison correction like easythresh in FSL.
References:
Flitney, D.E., & Jenkinson, M. 2000. Cluster Analysis Revisited. Tech. rept. Oxford Centre for Functional Magnetic Resonance Imaging of the Brain, Department of Clinical Neurology, Oxford University, Oxford, UK. TR00DF1.
K. Friston, K. Worsley, R. Frackowiak, J. Mazziotta, and A. Evans. Assessing the significance of focal activations using their spatial extent. Human Brain Mapping, 1:214?220, 1994.
Input:
StatsImgFile - The statistical image file name. Could be either Z, T, F or R statistical image. T/F/R images will transformed into Z image (according to their degree of freedom Df1, Df2) to perform further smoothness estimation and cluster thresholding.
VoxelPThreshold - P threshold for each voxel. Will be transformed into Z threshold in two-tailed way or one-tailed way. Note: FSL's Z>2.3 corresponds to one-tailed VoxelPThreshold = 0.0107. |Z|>2.3 corresponds to two-tailed VoxelPThreshold = 0.0214.
IsTwoTailed - 0: in one-tailed way. 1: converting voxel P threshold to z threshold in two-tailed way. Correct positive values to Cluster P at ClusterPThreshold/2, and correct negative values to Cluster P at ClusterPThreshold/2. Together the Cluster P < ClusterPThreshold.
ClusterPThreshold - The final cluster-level P threshold
OutputName - The output file name. Will be suffixed by'Z_ClusterThresholded_'.
MaskFile - The mask file name. If empty (i.e., ''), then all voxels are included.
Flag - 'Z', 'T', 'F' or 'R'. Indicate the type of the input statistical image
Df1 - The degree of freedom of the statistical image. For F statistical image, there is also Df2
Df2 - The second degree of freedom of F statistical image
Output:
The image file (Z statistical image) after correction.
Data_Corrected - The Data matrix after correction
ClusterSize - The cluster size for cluster-level p threshold and can be used in REST Slice Viewer. (Note: corner connection is used in FSL)
这个没有GUI，按照说明给出输入参数调用即可。
y_GRF_Threshold(StatsImgFile,VoxelPThreshold,IsTwoTailed,ClusterPThreshold,OutputName,MaskFile,Flag,Df1,Df2)