napari-wsi

A plugin to read whole-slide images within napari.

---

Installation via pip

You can install napari-wsi via pip:

pip install "napari-wsi[all]>=1.0"

This automatically installs all optional backends, as a shortcut for:

pip install "napari-wsi[openslide,rasterio,wsidicom]>=1.0"

In addition, to be able to read images using the openslide backend, it is required to install the OpenSlide library itself, for example by installing the openslide-bin python package (also via pip).

Installation via conda

You can also install napari-wsi via conda:

conda install -c conda-forge "napari-wsi>=1.0"

This already installs all optional dependencies, including OpenSlide.

Description

This napari plugin provides a widget for reading various whole-slide image formats using a common zarr store inteface, based on the libraries openslide, rasterio, and wsidicom.

Quickstart

After installation, open the Plugins menu in the viewer and select WSI Reader to open the widget. Then select a Backend to use, select a Path to open, and click Load.

If sRGB is selected in the Color Space menu and an ICC profile is attached to the given image, a transformation to this color space will be applied when the image data is read. Otherwise, the raw RGB image data will be displayed.

This plugin can also be used to open image files via drag and drop into the viewer window. The file suffixes '.bif', '.ndpi', '.scn', '.svs' are registered with the openslide backend, while the suffixes '.tif' and '.tiff' are registered with the rasterio backend. These files can also be opened directly from the command line or from a python script, in various ways:

napari CMU-1.svs

from napari import Viewer

viewer = Viewer()
viewer.open("CMU-1.svs", plugin="napari-wsi")

from napari import Viewer
from napari_wsi.backends.openslide import OpenSlideStore

viewer = Viewer()
store = OpenSlideStore("CMU-1.svs")
store.to_viewer(viewer)

Known Issues & Other Notes

• This plugin is prototype research software and there may be breaking changes with each release of the plugin, which is also the case for current releases of the napari viewer itself.
• The wsidicom backend supports loading annotations together with the image data. However, this may take several minutes, depending on the number and complexity of the annotations. When loading more than a few thousand polygon annotations, make sure that the experimental "triangles speedup" setting is enabled.