search

volBrain / platform

0 stars 1 forks source link

superimposition of native results and native T1 image #2

Open IFaillenot opened 9 months ago

IFaillenot commented 9 months ago

Bonjour,

I used Vol2brain to detect white matter lesion/abnormalities in a patient. This was well detected. When the mni_tissue label 8 ("lesion") is overlayed on the mni_t1 image (left part of the figure joined) , it is perfect. When the native_tissue label 8 is overlayed on the T1 image given as input to vol2brain (right part of the figure), we can see that this T1 is not the "native" image...
the readme file says "{native, mni}t1.nii.gz Filtered and normalized T1 image" but there isn't nativet1* image in the directory.

Capture d’écran 2023-12-08 à 13 53 52

pierrickvolbrain commented 9 months ago

I guess that the issue is coming from your viewer which do not correctly read the matrix header. With ITKSnap (the viewer recommended in the tutorial), your case seems perfect.

Screenshot 2023-12-08 at 14 54 07

IFaillenot commented 9 months ago

usually, MRIcro_GL correctly reads the matrix header.... I am quite self confident about the format (Siemens data acquired this year, converted in bids with heudiconv) where did you find the "original_t1_job image or the "native_t1_job ? it will solve the issue if I have it

pierrickvolbrain commented 9 months ago

This is the submitted image. I will send it to you by PM.

"the readme file says "{native, mni}t1.nii.gz Filtered and normalized T1 image" : this was true, but finally we decided to remove it in order to reduce the size of the archive. We will update the readme.

IFaillenot commented 9 months ago

ok the original is exactly the same that mine. So I still have a problem.