关于rest工具箱里的reho.m

老师、学长好~ 请教个问题
在使用 REST_V1.5_101101 这个工具包后,自己根据定义写了一个计算reho的代码。即对每个体素的所有时间点直接升序sort,然后计算KCC值。没有加入任何优化。但计算出来的结果,每个体素的reho值普遍比工具箱里算的值相对偏小。模糊可以看到,自己算的图像黑白的区域与工具箱的结果是很相似的。
回头看工具箱里的代码时,发现reho.m计算rank时取了平均,注释写的是"Compute mean of tied ranks"。
我的问题是:
1、tied ranks是什么意思啊?
2、为什么不直接用[result, Index] = sort(matrix) 里的Index,而自己重新写了一个rank呢?
3、reho.m里的处理方法比原定义的方法有什么优点吗?
谢谢老师、学长回复哈~~

Functional Connectivity with multiple subjects?

Hello, I'm a new user of REST and I'm trying to use the functional connectivity tool with multiple subject's data. I followed the instructions provided in the documentation but I have a question about inputting multiple subject's folders. What I did was selected one subject's folder containing their nifti pair data set. Then added another directory to the input list containing another subject's data folder.

ER设计的数据用DPARSF做功能连接分析

大家好,

我现在想要对一组ER设计的数据用DPARSF做功能连接分析。在阅读相关文献时,我注意到很多使用任务激活数据做连接的研究中,都将实验设计的条件回归掉
如:The task-related regressor
(tactile or visual moving stimuli) of each run was also included as
an additional regressor of no interest in the correlation analysis in
order to minimize the effect of the specific tasks (Whalley et al.,
2005) ----- Sani et al, 2010

DPARSF 出错

Win XP, Matlab 7.01(R14), SPM5, REST 1.6,DPARSF Basic Edition

??? Error using ==> print
Error using ==> graphics\private\prepare
Error using ==> capturescreen
The rectangle passed to getframe must be at least partially on screen.

Error in ==> DPARSF_run at 635
eval(['print(''-dtiff'',''-r100'',''',AutoDataProcessParameter.SubjectID{i},'.tif'',h);']);