Unknown input segmentation type

[Michael-shannon]

Hi there

I would love to try your tool, but I can't understand what kind of segmented image to supply.

My first try was to supply a segmented mask as a tif from cellpose, but i don't think this satisfies the requirements.

Could you provide an example image, and tell me which tool you used? If at all possible, that would be very helpful.

Even better, would be to tell me how to adapt a cellpose mask to this workflow - then i can stick with the upstream tools that i know work well.

Thanks for any help you can offer - your tool looks amazing,

Michael

[borgesaugusto]

Hi! Thank you for your comment and for wanting to try the tool! Currently, the supported format is a skeletonized tiff image. We usually generate them using the Epyseg software (https://github.com/baigouy/EPySeg). Examples of how this looks like are in the "tests" and "examples" directory, as files in TIF format. For example, the in vivo examples were generated from these tif files: https://github.com/borgesaugusto/forsys/tree/main/examples/data/in_vivo

We are thinking of extending the functionality to also work with cellpose outputs, so I'll mark this issue as an enhancement. It would be very helpful if you could provide a small example of how the tif image from Cellpose would look, so that we can help you further.

[Michael-shannon]

test_data.zip

Hi again! Thanks a lot - it works fine with the example data you pointed to.

I used epyseg on this example file, both the raw data and the segmentation are in the zip attached.

I played around a little with the settings but could only achieve a very homogenous result where I would expect some different forces/pressures to be at play. Any help you could offer would be much appreciated!