These deep learning methods can give extremely good results out of the box. Stardist requires a bit of manual annotation, but I hear ~40 nuclei is enough, which should be doable. There is an image.sc post that shows how to visualise stardist results in napari.
These deep learning methods can give extremely good results out of the box. Stardist requires a bit of manual annotation, but I hear ~40 nuclei is enough, which should be doable. There is an image.sc post that shows how to visualise stardist results in napari.