AICSImageIO 3.1.0 Release

Announcing the 3.1.0 release of AICSImageIO, a Python library for reading and writing image data with specific support for common formats used in computational biology (CZI, TIFF, OME-TIFF, MP4, GIF, PNG, etc.).

AICSImageIO now uses Dask arrays for all array handling. For a user this means that extremely large files are now easily handled and for users that use Dask clusters, they should see an improvement in image reading and image processing / computation time. Additionally, we added minor utility functions to help spawn or connect to Dask clusters and visualization of the image with some standard settings in napari.

Coming up we are working on adding readers for Slidebook and Zarr files, creating a Zarr file writer, and a general CZI-to-OME metadata conversion function.

For more details on how to get started with AICSImageIO check out our documentation.


Wow. Did you use our libCZI to create the python wrapper?

Sebi (from Zeiss)

Hi Seb,
In short yes. We’ll post in the near future with more info as the base library has support for mosaic files and such which isn’t propagated up to AICSImageIO. libCZI has been very helpful.
~Jamie (Allen Institute for Cell Science)

Hi Jamie,

To bad we did not really talk about this topic when in was is Seattle last fall.
When it comes to integrating Tiling Support please keep me posted and do not hesitate to contact me in case of questions. I clearly see a benefit for all of us when there would be a “reasonable” easy-to-use python wrapper for libCZI.

So far I mainly use the tools mentioned here: Read_CZI_and_OMETIFF_and_display_widgets_and_napari

1 Like

This is really exciting @JacksonMaxfield @heeler, congrats on the release! napari is working on support for file-io plugins in PR 937 and I’d love to see AICSImageIO be runnable as a napari plugin once that PR gets merged. You can see on the PR how we’re imagining that working, but it basically will involve conforming to hooks we provide via pluggy. Feel free to check out / comment on the PR now. Once that gets merged we’ll work on docs / tutorials to explain how to make plugins. I’m also looping in @talley who has been a driving force on the io-plugin architecture.