Hello everyone,
I used CAT12 to clean my T1 scans and it outputted a neat T1 as well as WM and GM segmentation. For the step of creating the gmwm seed for the ACT tractography I tried two things
- Inputting 5ttgen the mri/p0T1.nii image. Here is how it looks
When I tried running 5ttgen -premasked (whiven it has already the skull extracted) with this image it gave me 5 volumes of which all were empy except the 4th one like this:
5ttgen with the raw T1 image gave me a 5tt segmentation, but of course it was very bad.
- Seeing that CAT12 gave me nice GM segmentation
I thought I could create the gmwm seed with that. But 5ttgmwmi takes as input the 5tt output.
Beyond solving problem in 1. My question is the following:
How can I generate the gmwmi mask for tck -act without 5ttgmwi using my CAT12 segmented GM image?
Thank you in advance! Hope i’m clear with my problem
Hi @Ivan_Mindlin,
5ttgmwi allows you to use any segmentation you want, as long as you put it in the correct format. That is a 5 volume image, where the first volume is the cortical GM, second volume the subcortical GM, third volume WM, forth volume CSF and last volume pathological tissue (if you have healthy data, this is just a blank volume). Each tissue should be between 0 and 1 and the sum of each voxel has to be one (however I think these 2 requirements have been relaxed). If you already have your segmentation from CAT12, you can combine the data using MRtrix tools to achieve this. I hope this helps.
Best regards,
Manuel
Hello!
Thank you for the suggestion. It’s true that the input to 5ttgmwi can be constructed manually but my concern was regarding the input to the tckgen command.
I insisted on using 5ttgen output because I wanted to benefit from the ACT framework, and I understand that requires having CSF and subcortical-GM which CAT12 wasn’t giving me.
The problem was that I was inputting an extracted brain and the -premasked option was not working.
I found a script that uses CAT12 and outputs something similar to the raw T1 but also with the face and skull of the subkect. This was what 5ttgen wanted as input
Sorry for the late reply, hope this is useful for someone!