Dear all,

after preprocessing of my data, conn created numerous files and stored them in the folders of my original structural and functional files. Can anyone help me what files these are exactly? I didn't find anything online, and the manual does not specify it, either.

I'm mostly wondering about the files created by conn in the "structural" folder of my data:

c1ct1.nii
c2ct1.nii
c3ct1.nii
c4ct1.nii
c5ct1.nii
centering_t1.mat
ct1.nii
ct1_seg8.mat
...
and so on...


If you know what any of these files are, I'd be very happy about your help!

Specifically: Are any of the newly created files the estimated GM/WM/CSF masks?

I use the default preprocessing pipeline.


Thanks!

Cheers,
Sascha

Hi Sascha,

Conn uses SPM style for naming the files. So given your examples, I assume that your structural file was named "t1". Therefore,

ct1 ---> centered T1
c1ct1 ---> segmented centered T1 (grey matter mask)
c2ct1 ---> segmented centered T1 (white matter mask)
c3ct1 ---> segmented centered T1 (CSF)

If I remember correctly, the fourth one is 1-sum of these three, and the fifth one defines the skull (you can check by looking at the tissue probability maps of SPM). Looks like you have used SPM 12 for your analysis and hence, the segmentation module would have called the TPM.nii file located in your SPM folder. One way to understand more about these files is to simply load them and visualize them in let's say MRICron.

Regarding the other files,

centering_t1.mat contains the rotation matrix applied to center (bring the origin closer to ACPC plane)
ct1_seg8.mat would be the matrix generated while segmentation module was running. If you load it in MATLAB, you would see several variables in your workspace. They include, among other things, the original image, warping parameters, rotation matrix, the location for the TPMs, etc.


During preprocesing, the files are named as per the following convention:

uxxx.nii ---> unwarped
auxxx.nii ---> slice timing corrected, unwarped
wauxxx.nii ---> normalized, slice timing corrected, unwarped
swauxxx.nii ---> smoothed, normalized, slice timing corrected, unwarped

As you can see, conn keeps appending names to the files (SPM style). Output of one step goes in as input for the next.

Other than these, you will see the rp_xxx file which is the realignment output.
y_ and iy_ are the deformation fields  (instead of the older sn_.mat files), "which contain three image volumes encoding the x, y and z coordinates (in mm) of where each voxel maps to" (SPM 12 release notes). You would have these for structural and functional images.



For your second question, the c1ct1 - c3ct1 would be the GM/WM/CSF masks, and wc1ct1 - wc3ct1 would be the normalized GM/WM/CSF masks that you are looking for.

Hope this helps.



Best,
Pravesh
 

 
Originally posted by Sascha Froelich:

Dear Pravesh,

thanks a lot, that was exactly the kind of answer I was hoping for. One more question: What exactly is meant by "unwarp" (as in the file-prefix u)? Is it equivalent to "realign"?

Cheers,
Sascha


Originally posted by Pravesh Parekh:

Dear Pravesh,

thanks again!

Just one more question: do you know if conn also generates a brainmask.nii and if so, where and how?

Cheers,
Sascha

Hi Sascha,

I think you are referring to the brainmask.nii file which is shipped by default. It is located in the SPM folder (exact location depends on the version of SPM). In your case (SPM12), the brainmask.nii file should be in the following two locations:

Path_to_SPM\toolbox\cat12\templates_1.50mm
Path_to_SPM\toolbox\Fieldmap

Both are slightly different files (different voxel resolutions and overall dimensions). 

If you are looking for the analysis mask which is used by Conn for its analysis you would find it in the following location:

Path_to_Conn\utils\surf

the file "mask.volume.brainmask" is what is defined in the "Options" tab in the Setup window of a project (explicit mask).



Best
Pravesh

Originally posted by Sascha Froelich:

Hi Sascha,

Based on this forum post (https://www.nitrc.org/forum/forum.php?th...), I gather from Dr. Alfonso's reply that the ewc* files are image files after "erosion" which might be used to assess the quality of segmentation (check Dr. Alfonso's reply; I am not too sure about this one).

Looking at the wc0ct1.nii I think it should be the skull stripped file. If you compare wct1 file and the wc0ct1 file, you would likely only see that as the difference. Skullstripping is usually performed in SPM by calling the imcalc utility and supplying the following expression:

"i1.((i2+i3+i4)>0.5)" [https://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=spm;dd3194c2.1209]

where i1 = original anatomical scan
i2, i3, and i4 being the GM/WM/CSF masks created from the segmentation step. The 0.5 is a threshold and can be varied depending on how stringent the skull stripping needs to be.

So, its likely that the reason you do not see a c0ct1 is that the skull stripping might be happening only at the end. This part of my answer is just intuitive and you might want to check with Dr. Alfonso in case I have got it wrong here.



Best
Pravesh

Hi Sascha,

Conn uses SPM style for naming the files. So given your examples, I assume that your structural file was named "t1". Therefore,

ct1 ---> centered T1
c1ct1 ---> segmented centered T1 (grey matter mask)
c2ct1 ---> segmented centered T1 (white matter mask)
c3ct1 ---> segmented centered T1 (CSF)

If I remember correctly, the fourth one is 1-sum of these three, and the fifth one defines the skull (you can check by looking at the tissue probability maps of SPM). Looks like you have used SPM 12 for your analysis and hence, the segmentation module would have called the TPM.nii file located in your SPM folder. One way to understand more about these files is to simply load them and visualize them in let's say MRICron.

Regarding the other files,

centering_t1.mat contains the rotation matrix applied to center (bring the origin closer to ACPC plane)
ct1_seg8.mat would be the matrix generated while segmentation module was running. If you load it in MATLAB, you would see several variables in your workspace. They include, among other things, the original image, warping parameters, rotation matrix, the location for the TPMs, etc.


During preprocesing, the files are named as per the following convention:

uxxx.nii ---> unwarped
auxxx.nii ---> slice timing corrected, unwarped
wauxxx.nii ---> normalized, slice timing corrected, unwarped
swauxxx.nii ---> smoothed, normalized, slice timing corrected, unwarped

As you can see, conn keeps appending names to the files (SPM style). Output of one step goes in as input for the next.

Other than these, you will see the rp_xxx file which is the realignment output.
y_ and iy_ are the deformation fields  (instead of the older sn_.mat files), "which contain three image volumes encoding the x, y and z coordinates (in mm) of where each voxel maps to" (SPM 12 release notes). You would have these for structural and functional images.



For your second question, the c1ct1 - c3ct1 would be the GM/WM/CSF masks, and wc1ct1 - wc3ct1 would be the normalized GM/WM/CSF masks that you are looking for.

Hope this helps.



Best,
Pravesh
 

 
Originally posted by Sascha Froelich: