We are pleased to announce the release of SlideIO version 2.8.1, an update to our open-source library for pathology image analysis.
This version introduces a new structured metadata API that exposes slide and scene metadata as a JSON tree, adds channel attributes (color, name, emission wavelength) for fluorescence formats, and brings significant performance improvements to the converter through multithreaded encoding. It also includes an extensive stability pass with numerous bug fixes across the library.
Highlights
Structured Metadata API
- New
Slide::getMetadata()andScene::getMetadata()methods expose file and scene metadata as a navigable JSON tree. - Format-specific structured metadata implemented for SVS (Aperio header), NDPI, DCM, GDAL, and ZVI drivers.
- ZVI: complete file metadata is now exposed as a JSON tree organized by Channels, Z-Slices, and Time-Frames.
Channel Attributes
- New
Scene::getChannelAttributes()method exposes per-channel attributes such as channel name, color, and emission wavelength. - Channel attributes are available for CZI, VSI, SCN, PKE, OME-TIFF, and ZVI formats.
- The converter preserves channel color information when converting to OME-TIFF.
Converter Improvements
- Multithreaded encoding with multiple parallel readers significantly speeds up conversions.
- Thread count and batch size are configurable via
SVSConverterParameters/OMETIFFConverterParametersand the converter command line. - JPEG compression support for conversion of CZI images.
- Fixed multithreaded TIFF conversion with JPEG compression.
Stability and Bug Fixes
- Extensive code-review pass fixing more than 25 issues, including integer overflows, missing bounds checks (CZI, VSI/ETS), resource leaks, uninitialized reads, and endianness conversion errors.
- Improved thread safety: atomic reference counters and proper exception propagation in multithreaded code paths.
- Big-endian platform support fixes.
- Fixed automatic driver detection for QPTIFF files.
SlideandSceneobjects now report the driver used to open them.- SVS: zoom levels are now reported correctly for small images.
We welcome your feedback and issue reports on GitHub and appreciate your continued support of the SlideIO project.