Isotropic scaling

[zimengwu]

Hi all,

Thank you for this awesome package - very exciting!
I just had a question about the scaling of the image after import. I'm trying to use this on confocal data that's not isotropic, but I noticed my image was scaled isotopically after importing and running the viewer so it looked a bit squashed. I can overcome this by rescaling z manually before feeding it into the package pipeline. However, I was just wondering if there was anyway to set the x,y,z resolution from the metadata before generating the mesh as I assume it would change any tension/pressure inferences not having the correct shape?

Thank you for any help,

Zimeng

[hturlier]

Hi Zimeng,
Sorry for late answer. We assume that the masks you provide are isotropic in x-y-z resolution indeed, so you need to rescale them beforehand with your preferred method.
Otherwise, as you guessed, the inferences may not be correct.
We are working on solutions to remedy - generically - about the common problem of confocal 3D image anisotropy, but that's a different project.
Best wishes,
Hervé

[parsazarei1]

Hi,
Thanks for the instructions! However, that did not solve the problem for me. My image is scaled correctly in Fiji and has a round cross-section, but when the embryo creates the color-coded inference maps, the cross-section is flattened. Can you help me understand why? I would like to use your method in my paper.