I've submitted a paper to #nldl2024! This is my first real #conference submission.

Title: Towards AI for approximating hydrodynamic simulations as a 2D segmentation task

Alternate title: Abusing image segmentation to approximate physics based simulations /lh

https://www.nldl.org/

#AI #imagesegmentation #hydrodynamic #deeplabv3plus #nldl

Dominik Kutra from @ilastik_team help trainees with #ImageSegmentation in ilastik, explore and discover the tools in the @bioimageio 🦒 and run the model zoo models in ilastik.

Learn to segment your images with the Segment Anything Model (SAM) by Meta AI in just a few lines of #Python code! Check out my YouTube tutorial: https://youtu.be/fVeW9a6wItM
#imageSegmentation #MetaAI #segmentanything #microscopy

Quick #ai #deeplearning tip about sensitivity vs specificity:

I can never remember which is which, but think of it like a camera.

* Sensitivity: If the camera's sensitivity (e.g. ISO) isn't high enough, it won't see anything (positive).
* Specificity: If the image is out of focus, it will be blurry and the bit you're interested in will overflow into the negative region.

#imagesegmentation

I have this silly question about #imagesegmentation. So is there any advantage to using different software for image segmentation? I see #ImageJ, #CellProfiler, #Napari and #Python all can do this, but if the underlying algorithm is the same, then why not just stick with ImageJ?