search

MRtrix3 / mrtrix3

MRtrix3 provides a set of tools to perform various advanced diffusion MRI analyses, including constrained spherical deconvolution (CSD), probabilistic tractography, track-density imaging, and apparent fibre density
http://www.mrtrix.org
Mozilla Public License 2.0
281 stars 176 forks source link

Investigate 5ttgen fsl -nocrop option #1469

Open Lestropie opened 5 years ago

Lestropie commented 5 years ago

From forum thread.

What might be happening is that an image downstream of the standard_space_roi step is being used as the output image template, but that step is itself cropping the image FoV independently of the later mrcrop step.

benjamin-kalloch commented 5 years ago

I experience the same issue as mentioned in the forum thread and would like to share my observations hoping that it may help in resolving the bug:

[PROBLEM] The segmentation image is re-oriented (as compared to the input T1 image) after "5ttgen fsl" despite the use of the "-nocrop" flag.

[EXPECTATION] The segmentation image should have the same dimension and orientation as the original T1 image. It should be possible to load both images in, for example, FSL view (segmentation image & T1 image) and overlay them.

I would support the suspicion that the reason for this behavior is the use of the fsl tool "standard_space_roi". In my case, I have noticed that the re-orientation only occurs, when the input MR image is oriented in a non-standard way. For example: (A) no re-orientation happens, i.e. the segmentation image overlays perfectly fine the T1 image in FSLView when the T1 image is oriented the following way: Orientation: Axial X axis orientation: left to right Y axis orientation: posterior to anterior Z axis orientation: inferior to superior

(B) re-orientation happens, i.e. the segmentation image DOES NOT overlay the T1 image (FSLView prints: "Unable to load incompatible overlay!") when the T1 image is oriented like Orientation: Sagittal X axis orientation: anterior to posterior Y axis orientation: inferior to superior Z axis orientation: left to righ

Lestropie commented 5 years ago

OK, that does actually give me a useful place to start when trying to reproduce. Thanks for the input @BennyKay :+1: