Explanation of Result outputs

Submitted by Vivek Trivedi on Thu, 09/27/2018 - 02:18

Dear DPABI experts:

I am a bit confused regarding the outputs of the files in the 'Results' folder. There seem to be individual subject image files, m* files and z* files. My questions are: 

1. What do these different files represent?

2. Which files should I use for standardization and statistical analysis?

3. Would it make sense to use an ANOVA test to compare the ReHo results z* files of 3 groups (Control, Early Disease, Advanced Disease)? Will this test be able to show me significant differences in Regional Homogeneity between the 3 groups?

I appreaciate your patience and look forward to a response.

Thank you,

Vivek

Forums:

puyunfashi

Thu, 09/27/2018 - 03:26

You may refer to http://mrirc.psych.ac.cn/RfMRIMapsDataSharingStructure for more info about the data structure.

we recommend you to use sm* or sz* because these metrics have been smoothed and centralized (subtracted mean) or standardized (subtracted mean and divided by standard deviation)

It's a good idea to conduct such analysis, hope you get what you expected.

Best!

Xiao

YAN Chao-Gan

Thu, 09/27/2018 - 04:30

m* doesn't stand for "subtracted mean", but stand for "divide mean".

Vivek Trivedi

Fri, 09/28/2018 - 02:12

Thank you for your replies. 

I was able to use the statistical analysis tool to test for differences between the three groups using ANOVA. I have some questions about these results.

1. What do the numbers in the results viewer represent? Are these p-values? For my results the entire brain shows up as mostly red (close to zero).

2. Is it standard to add a covariate such as grey matter when comparing for differences between groups?

3. Is it necessary to run an FDR correction to this data? I have 14 advanced disease, 18 early disease, and 10 control subjects.

 

Thank you again for taking the time.

 

puyunfashi

Fri, 09/28/2018 - 06:41

1. It's a F-map because you conducted an anova;

2. It's decent to do this, although it may depend on the popular practice in your research area;

3. We recommend permutation test with TFCE or GRF correction, FDR is sometimes too stringent.

Best,

Xiao

Vivek Trivedi

Fri, 09/28/2018 - 20:41

Thank you for your reply. I followed your advise and tried using TFCE correction, but encountered the following error:

=======================================================================
                 Permutation Analysis of Linear Models
=======================================================================
Error using palm_takeargs (line 1133)
Unknown option: "folder\Temp\DependentVolume.nii"
 
Error in palm_core (line 33)
[opts,plm] = palm_takeargs(varargin{:});
 
Error in palm (line 81)
palm_core(varargin{:});
 
Error in y_GroupAnalysis_PermutationTest_Image (line 102)
palm([TempDir,filesep,'PALMConfig.txt']);
 
Error in y_ANCOVA1_Image (line 95)
    y_GroupAnalysis_PermutationTest_Image(DependentVolume,Regressors,OutputName,MaskFile,CovariateVolume,Contrast,'F',0,Header,PALMSettings);
    
Error in DPABI_STAT_TOOL>ComputeButton_Callback (line 526)
            y_ANCOVA1_Image(S, OutputName, MaskFile, ImageCell, TextCell, PALMSettings);
 
Error in gui_mainfcn (line 95)
        feval(varargin{:});
 
Error in DPABI_STAT_TOOL (line 42)
    gui_mainfcn(gui_State, varargin{:});
 
Error in
matlab.graphics.internal.figfile.FigFile/read>@(hObject,eventdata)DPABI_STAT_TOOL('ComputeButton_Callback',hObject,eventdata,guidata(hObject)) 
Error while evaluating UIControl Callback.
 
 
 
 
 
My environment is as follows:
MATLAB R2018a
DPABI v4.4 Basic
Windows 10
 
Thank you for your help with this matter.

Vivek Trivedi

Fri, 09/28/2018 - 20:41

Thank you for your reply. I followed your advise and tried using TFCE correction, but encountered the following error:

=======================================================================
                 Permutation Analysis of Linear Models
=======================================================================
Error using palm_takeargs (line 1133)
Unknown option: "folder\Temp\DependentVolume.nii"
 
Error in palm_core (line 33)
[opts,plm] = palm_takeargs(varargin{:});
 
Error in palm (line 81)
palm_core(varargin{:});
 
Error in y_GroupAnalysis_PermutationTest_Image (line 102)
palm([TempDir,filesep,'PALMConfig.txt']);
 
Error in y_ANCOVA1_Image (line 95)
    y_GroupAnalysis_PermutationTest_Image(DependentVolume,Regressors,OutputName,MaskFile,CovariateVolume,Contrast,'F',0,Header,PALMSettings);
    
Error in DPABI_STAT_TOOL>ComputeButton_Callback (line 526)
            y_ANCOVA1_Image(S, OutputName, MaskFile, ImageCell, TextCell, PALMSettings);
 
Error in gui_mainfcn (line 95)
        feval(varargin{:});
 
Error in DPABI_STAT_TOOL (line 42)
    gui_mainfcn(gui_State, varargin{:});
 
Error in
matlab.graphics.internal.figfile.FigFile/read>@(hObject,eventdata)DPABI_STAT_TOOL('ComputeButton_Callback',hObject,eventdata,guidata(hObject)) 
Error while evaluating UIControl Callback.
 
 
 
 
 
My environment is as follows:
MATLAB R2018a
DPABI v4.4 Basic
Windows 10
 
Thank you for your help with this matter.

YAN Chao-Gan

Tue, 10/23/2018 - 05:33

Hi, you might have a space in your path. "XXXX folder"