One sample t-test with Rest 1.8

Submitted by Wei on

Dear experts,

I got some problems with Rest 1.8 while doing statistics for fALff. mfAllf-1 files are generated by Rest strictly following the course video. 

1. About mask: If I leave the mask option empty, does it mean that Rest will run the default mask or whole brain? Is there any default mask in Rest programme that I can use?

2. I use DMN mask created by Pickatlas. This mask was tested in SPM before for statistics analysis and it worked well. But when I put it into Rest, It comes back with errors.

Read 3D EPI functional images: "N:\test1\mfALFF-1"...

 

                             Image Files in the Group:

                             mfALFF-100001.nii

                             mfALFF-100002.nii

                             mfALFF-100003.nii

                             mfALFF-100004.nii

                             mfALFF-100005.nii

                             mfALFF-100006.nii

                             mfALFF-100007.nii

                             mfALFF-100008.nii

                             mfALFF-100009.nii

                             mfALFF-100010.nii

                             mfALFF-100011.nii

                             mfALFF-100012.nii

 

                             One Sample T Test Calculating...

Array dimensions must match for binary array op.

 

Error in rest_ttest1_Image (line 49)

TTest1_T=TTest1_T.*MaskData;

 

Error in rest_ttest1_gui>btnCompute_Callback (line 130)

rest_ttest1_Image(DependentDirs,OutputName,MaskFile,Base);

 

Error in gui_mainfcn (line 95)

        feval(varargin{:});

 

Error in rest_ttest1_gui (line 27)

gui_mainfcn(gui_State, varargin{:});

 

Error while evaluating uicontrol Callback

 

Could you help me with these two problems? Thank you very much in advance.

 

Best regards,

Wei

 

YAN Chao-Gan

Mon, 02/15/2016 - 02:22

Hi Wei,

1. If you left it empty, is means whole brain. You can use BrainMask_05_61x73x61.img.

2. You have to resample the DMN mask to the resolution of your functional images. You can use image reslicer.

Best,

Chao-Gan