conda create -n Napari python=3.9
mamba install -c conda-forge tensorflow-gpu=2.10
mamba install napari pyqt
mamba install -c conda-forge openslide
mamba install -c conda-forge openslide-python
pip install stardist
Preload svs data into memory with openslide. It took 90s to load the 100MB test image:
TCGA-AK-3440-01Z-00-DX1.ea0763b1-4262-4c75-9f76-080af1ffeab7.svs
To install, run python -m pip install -e . in ./preloadsvs/.
An On-The-Fly reader of svs data with the help of Dask [https://docs.dask.org/en/stable/].
It took 3s to load the 100MB test image:
TCGA-AK-3440-01Z-00-DX1.ea0763b1-4262-4c75-9f76-080af1ffeab7.svs
It took 7s to load the 1000MB test image:
TCGA-CZ-4863-01Z-00-DX1.7f3da3a1-5d70-4d30-a9f3-80ab883cf9f6.svs
To install, run python -m pip install -e . in ./svsotf/.
Further apply rgb2hed preprocessing in scikit-image to identify nuclei when we are zoomed in enough. in the On-The-Fly reader.
To install, run python -m pip install -e . in ./svsotfhed/.
Further apply StarDist [https://github.com/stardist/stardist/] to identify nuclei when we are zoomed in enough. in the On-The-Fly reader.
To install, run python -m pip install -e . in ./svsmodel/.