wsireg 0.2.1

python package for registering multimodal whole slide microscopy images

wsireg

Multi-modal or mono-modal whole slide image registration in a graph structure for complex registration tasks using elastix.

• Documentation: https://wsireg.readthedocs.io.

Features

• Graph based approach to defining modalities and arbitrary transformation paths between associated images

• Use of elastix (through ITKElastix) to perform registration

• Support for linear and non-linear transformation models

• Transform associated data (masks, shape data) along the same path as the images.

• Supports images converted to OME-TIFF using bioformats2raw -> raw2ometiff pipeline

• All registered images exported as pyramidal OME-TIFF or OME-zarr that can be viewed in software such as Vitessce , vizarr, QuPath, OMERO or any platform that supports these formats.

• All transforms for complex registration paths are internally composited and only 1 interpolation step is performed, avoiding accumulation of interpolation error from many registrations

• Shape data (polygons, point sets, etc.) in GeoJSON format (future portable format for QuPath detection/annotation data) can be imported and transformations applied producing a modified GeoJSON

• Some support for reading native WSI formats: currently reads .czi and .scn but could be expanded to other formats supported by python package tifffile

History

0.0.2 (2021)

• First release on PyPI.

0.2.1 (2021-04-14)

• add RegImage sub-classes for different file types

  • TiffFileRegImage (.scn, .ndpi,.tiff,.tif) : uses dask + zarr to do memory-efficient computation of necessary data for registration / transformation

  • CziRegImage (.czi) : Carl Zeiss image format, can perform read-time pre-processing like RGB -> greyscale or selection of individual channels to limit memory footprint

  • OmeTiffRegImage (.ome.tiff,ome.tif): uses TiffFile to read images and parses OME metadata to get interleaved RGB information

  • MergeRegImage (meta): used to transform multiple images’ channels to a single OME-TIFF after registration if they output to the same size and data type (i.e. for cyclic IF)

  • NpRegImage (np.ndarray): Supports adding a registration image from a numpy array

  • SitkRegImage (everything else): uses SimpleITK to read images as a last resort. Will read entire image into memory!

• support masks for registration

  • Masks can be used in elastix to define pixels used in metric calculation

  • add ability to automatically crop images based on associated masks’s bounding box (can be useful if image dimensions differ greatly)

• use RegTransform class to manage transformations