1

I have some nii.gz files which, when I open using imajeJ , or any python library such as SimpleITK or Monai I get a stack of MRI images, as if taken from top to bottom. That being said, when I open the same file with slicer, I also get a stack of images, taken from the side and from the front, the G and Y channels. My question is, do the images from the different angles exist inside the file, or is 3D slicer inferring how these channels would look like based on the R channel. I am asking this because the other methods don’t seem to be able to find the other views. Thank you.

3D slicer enter image description here

ImageJ

enter image description here

Monai. SimpleITK in google colab enter image description here

Iliasp
  • 143
  • 8

2 Answers2

2

SimpleITK and Slicer both represent your Nifti file as a 3d image, i.e., a 3d array of pixels. Monai/SimpleITK is simply choosing to display that 3d image as a 2d stack of slices, and it is choosing to use the Z direction to cut the 3d volume into slices.

Slicer, in its 4-up display, is also displaying 2d cuts into the original 3d image. However it is choosing to cut the 3d image along all 3 axes, (X, Y, and Z).

Dave Chen
  • 1,905
  • 1
  • 12
  • 18
2

Viewers like Slicer, ITK-Snap, etc. also display the view from the different anatomical planes. Depending on the orientation of the patient inside the volume, the contained volumetric information can be displayed along the three planes and their corresponding axis. Each picture in that view is just a 2D representation of a slice along one of the axes.

So yes, this information is just inferred and is usually not contained in a DICOM (or NIFTI) file and is purely inferred for viewing purposes.

code-lukas
  • 1,586
  • 9
  • 19