TIAToolbox 2.0.1

TIAToolbox v2.0.1 (2026-03-16)

Bug Fixes and Other Changes

• Fixes filtering of tiles for post-processing within masked region (#1037).
• Fixes Calculation of max_inst_value for Instances where overlap is empty (#1038)

Full Changelog: v2.0.0...v2.0.1

TIAToolbox 2.0.0

TIAToolbox v2.0.0 (2026-03-11)

✨ Major Updates and Feature Improvements

⚙️ Engine Redesign (PR #578)

TIAToolbox 2.0.0 introduces a completely re-engineered inference engine designed for significant performance, scalability, and memory-efficiency improvements.

Key Enhancements

• A modern processing stack built on Dask (parallel/distributed execution) and Zarr (chunked, out-of-core storage)
• Standardised output formats across all engines:
  • Python dict
  • Zarr
  • AnnotationStore (SQLite-backed)
  • QuPath JSON
• Cleaner runtime behavior with reduced warning noise and a unified progress bar
• More predictable memory usage through chunked streaming
• Broader test coverage across engine components

🗺️ Improved QuPath Support

Enhancements include:

• Better handling of GeoJSON
• Support for multipoint geometries (#841)
• Improved semantic output helpers:
  • dict_to_store_semantic_segmentor (#926)
  • OME-TIFF probability overlays (#929)

🔬 New Nucleus Detection Engine

A dedicated nucleus detection pipeline has been added, built on the redesigned engine for improved accuracy and efficient large-scale processing.

🧠 KongNet Model Family

TIAToolbox 2.0.0 introduces KongNet, a high-performance architecture that achieved top results across multiple international challenges:

• 🥇 1st place: MONKEY Challenge (overall detection)
• 🥇 1st place: MIDOG (mitosis detection)
• ⭐ Top-tier performance on PUMA

Multiple pretrained variants are available (CoNIC, PanNuke, MONKEY, PUMA, MIDOG), each with standardised IO configurations.

🧬 Expanded Foundation Model Support

Additional foundation models are now supported (#906), broadening the range of high-capacity architectures available for feature extraction and downstream tasks.

🖼️ SAM Segmentation in TIAViz

TIAViz now integrates Meta’s Segment Anything Model (SAM), enabling:

• Interactive segmentation
• Rapid region extraction
• Exploratory annotation workflows

Simplified SAM usage (#968) streamlines its integration into analysis pipelines.

🖼️ Multichannel Image (mIF) Support in WSIReader

• Multichannel image support (PR #825) for immunofluorescence and non-RGB modalities

🖼️ WSI Registration Visualization in TIAViz

TIAViz now supports interactive WSI registration visualisation, allowing users to compare aligned slides in two modes:

• Side‑by‑side view — view fixed and moving WSIs next to each other for direct comparison
• Overlay mode — blend registered WSIs with adjustable transparency for visual inspection of alignment quality

This feature enables intuitive, high‑resolution exploration of slide registration results directly within TIAViz.

🧩 Enhanced WSIReader & Metadata Handling

Major improvements include:

• More robust cross-vendor metadata extraction (#1001)
• Simplified Windows installation using openslide-bin (no manual DLL steps)
• macOS Tileserver fix (#976)
• Improved DICOM reading (#934)

☁️ New Cloud-Native Reader: FsspecJSONWSIReader (PR #897)

A new reader supporting fsspec-compatible filesystems, enabling seamless access to WSIs stored on:

• S3
• GCS
• Azure
• HPC clusters
• Any fsspec-supported backend

This enables cloud-native and distributed data workflows.
Contributed by @aacic

🤗 Pretrained Models Migrated to Hugging Face

All pretrained models and sample assets have been migrated (#945, #983), improving:

• Download reliability
• Versioning and reproducibility
• Caching and CI integration
• Licensing clarity per model family

🛡️ Security, Compatibility & Tooling

🔐 Security & Dependency Updates

• Dependency upgrades
• Internal security improvements
• Explicit workflow permissions added (#1021, #1023)

🐍 Python Version Support

• Dropped: Python 3.9
• Added: Python 3.13
• Supported: Python 3.10–3.13
• Updated CUDA wheel source to cu126

🛠️ Developer Tooling & CI/CD

• Expanded mypy type-checking coverage (#912, #931, #935, #951)
• Updated pre-commit hooks and general formatting
• CI uses CPU-only PyTorch for faster, more reliable builds (#974, #979)
• Updated pip install workflow (#1013)
• Added new Python 3.13 Docker images (#1014, #1019)

🧹 Bug Fixes & Stability Improvements

• Fixed multi-GPU behaviour with torch.compile (#923)
• Fixed DICOM reading issue (#934)
• Fixed annotation contour handling with holes (#956)
• Fixed consecutive annotation load bug (#927)
• Fixed SCCNN model issues (#970)
• Fixed MapDe dist_filter shape issue (#914)
• Improved notebook reliability on Colab (#1026–#1030)
• macOS TileServer issues resolved (#976)

🧭 Migration Guide for Users

🔄 Updating from 1.x to 2.0.0

Update calls: replace .predict() with .run()

# Old
results = segmentor.predict(imgs=[...], ioconfig=config)

# New
results = segmentor.run(images=[...], ioconfig=config)

Use patch_mode: replace mode="patch" with patch_mode=True and mode="tile" or "wsi" with patch_mode=False

# Old
results = segmentor.predict(imgs=[...], mode="patch", ioconfig=config)

# New
results = segmentor.run(images=[...], patch_mode=True, ioconfig=config)

# Old
results = segmentor.predict(imgs=[...], mode="wsi", ioconfig=config)

# New
results = segmentor.run(images=[...], patch_mode=False, ioconfig=config)

Use the new I/O configs

from tiatoolbox.models.engine.io_config import IOSegmentorConfig

config = IOSegmentorConfig(
    patch_input_shape=(256, 256),
    stride_shape=(240, 240),
    input_resolutions=[{"resolution": 0.25, "units": "mpp"}],
    save_resolution={"units": "baseline", "resolution": 1.0},
)

Specify the output format

results = segmentor.run(
    images=[...],
    ioconfig=ioconfig,
    output_type="zarr",  # or "dict", "annotationstore", "qupath"
    save_dir="outputs/",
)

Update imports

• tiatoolbox.typing → tiatoolbox.type_hints

Install requirements

• Python 3.10+ required
• On Windows: install OpenSlide via pip install openslide-bin

Full Changelog: v1.6.0...v2.0.0

TIAToolbox 1.6.0

TIAToolbox v1.6.0 (2024-12-12)

Major Updates and Feature Improvements

• Foundation Models Support via timm API (#856, contributed by @GeorgeBatch)
  • Introduced TimmBackbone for running additional PyTorch Image Models.
  • Tested models include UNI, Prov-GigaPath, and H-optimus-0.
  • Added an example notebook demonstrating feature extraction with foundation models.
  • timm added as a dependency.
• Performance Enhancements with torch.compile (#716)
  • Improved performance on newer GPUs using torch.compile.
• Multichannel Input Support in WSIReader (#742)
• AnnotationStore Filtering for Patch Extraction (#822)
• Python 3.12 Support
• Deprecation of Python 3.8 Support
• CLI Response Time Improvements (#795)

API Changes

• Device Specification Update (#882)
  • Replaced has_gpu with device for specifying GPU or CPU usage, aligning with PyTorch's Model.to() functionality.
• Windows Compatibility Enhancement (#769)
  • Replaced POWER with explicit multiplication.

Bug Fixes and Other Changes

• TIFFWSIReader Bound Reading Adjustment (#777)
  • Fixed read_bound to use adjusted bounds.
  • Reduced code complexity in WSIReader (#814).
• Annotation Rendering Fixes (#813)
  • Corrected rendering of annotations with holes.
• Non-Tiled TIFF Support in WSIReader (#807, contributed by @GeorgeBatch)
• HoVer-Net Documentation Update (#751)
  • Corrected class output information.
• Citation File Fix for cffconvert (#869, contributed by @Alon-Alexander)
• Bokeh Compatibility Updates
  • Updated bokeh_app for compatibility with bokeh>=3.5.0.
  • Switched from size to radius for bokeh>3.4.0 compatibility (#796).
• JSON Extraction Fixes (#772)
  • Restructured SQL expression construction for JSON properties with dots in keys.
• VahadaneExtractor Warning (#871)
  • Added warning due to changes in scikit-learn>0.23.0 dictionary learning (#382).
• PatchExtractor Error Message Refinement (#883)
• Immutable Output Fix in WSIReader (#850)

Development-Related Changes

• Mypy Checks Added
  • Applied to utils, tools, data, annotation, and cli/common.
• ReadTheDocs PDF Build Deprecation
• Formatter Update
  • Replaced black with ruff-format.
• Dependency Removal
  • Removed jinja2.
• Test Environment Update
  • Updated to Ubuntu 24.04.
• Conda Environment Workflow Update
  • Implemented micromamba setup.
• Codecov Reporting Fix (#811)
  Full Changelog: v1.5.1...v1.6.0

TIAToolbox 1.5.1

1.5.1 (2023-12-16)

Development related changes

• Specifies compatible Python versions
  • Fixes tiatoolbox-feedstock build for conda-forge release #763

Full Changelog: v1.5.0...v1.5.1

TIAToolbox 1.5.0

1.5.0 (2023-12-15)

Major Updates and Feature Improvements

• Adds the bokeh visualization tool. #684
  • The tool allows a user to launch a server on their machine to visualise whole slide images, overlay the results of deep learning algorithms or to select a patch from whole slide image and run TIAToolbox deep learning engines.
  • This tool powers the TIA demos server. For details please see https://tiademos.dcs.warwick.ac.uk/.
• Extends Annotation to Support Init from WKB #639
• Adds IOConfig for NuClick in pretrained_model.yaml #709
• Adds functions to save the TIAToolbox Engine outputs to Zarr and AnnotationStore files. #724
• Adds Support for QuPath Annotation Imports #721

Changes to API

• Adds model.to(device) and model.load_model_from_file() functionality to make it compatible with PyTorch API. #733
• Replaces pretrained with weights to make the engines compatible with the new PyTorch API. #621
• Adds support for high-level imports for various utility functions and classes such as WSIReader, PatchPredictor and imread #606, #607,
• Adds tiatoolbox.typing for type hints. #619
• Fixes incorrect file size saved by save_tiles, issue with certain WSIs raised by @TomastpPereira
• TissueMasker transform now returns mask instead of a list. #748
  • Fixes #732

Bug Fixes and Other Changes

• Fixes pixman incompability error on Colab #601
• Removes shapely.speedups. The module no longer has any affect in Shapely >=2.0. #622
• Fixes errors in the slidegraph example notebook #608
• Fixes bugs in WSI Registration #645, #670, #693
• Fixes the situation where PatchExtractor.get_coords() can return patch coords which lie fully outside the bounds of a slide. #712
  • Fixes #710
• Fixes #738 raised by @xiachenrui

Development related changes

• Replaces flake8 and isort with ruff #625, #666
• Adds mypy checks to root and utils package. This will be rolled out in phases to other modules. #723
• Adds a module to detect file types using magic number/signatures #616
• Uses poetry for version updates instead of bump2version. #638
• Removes setup.cfg and uses pyproject.toml for project configurations.
• Reduces runtime for some unit tests e.g., #627, #630, #631, #629
• Reuses models and datasets in tests on GitHub actions by utilising cache #641, #644
• Set up parallel tests locally #671

Full Changelog: v1.4.0...v1.5.0

TIAToolbox 1.4.1

Bug Fixes and Other Changes

• Fix dictionary changed size Error #626 (#605)

Full Changelog: v1.4.0...v1.4.1

TIAToolbox 1.4.0

1.4.0 (2023-04-24)

Major Updates and Feature Improvements

• Adds Python 3.11 support [experimental] #500
  • Python 3.11 is not fully supported by pytorch pytorch/pytorch#86566 and openslide openslide/openslide-python#188
• Removes Python 3.7 support
  • This allows upgrading all the dependencies which were dependent on an older version of Python.
• Adds Neighbourhood Querying Support To AnnotationStore #540
  • This enables easy and efficient querying of annotations within a neighbourhood of other annotations.
• Adds MultiTaskSegmentor engine #424
• Fixes an issue with stain augmentation to apply augmentation to only tissue regions.
  • #546 contributed by @navidstuv
• Filters logger output to stdout instead of stderr.
  • Fixes #255
• Allows import of some modules at higher level for improved usability
  • WSIReader can now be imported as from tiatoolbox.wsicore import WSIReader
  • WSIMeta can now be imported as from tiatoolbox.wsicore import WSIMeta
  • HoVerNet, HoVerNetPlus, IDaRS, MapDe, MicroNet, NuClick, SCCNN can now be imported as from tiatoolbox.models import HoVerNet, HoVerNetPlus, IDaRS, MapDe, MicroNet, NuClick, SCCNN
• Improves PatchExtractor performance. Updates WSIPatchDataset to be consistent. #571
• Updates documentation for License for clarity on source code and model weights license.

Changes to API

• Updates SCCNN architecture to make it consistent with other models. #544

Bug Fixes and Other Changes

• Fixes Parsing Missing Omero Version NGFF Metadata #568
  • Fixes #535 raised by @benkamphaus
• Fixes reading of DICOM WSIs at the correct level #564
  • Fixes #529
• Fixes scipy, matplotlib, scikit-image deprecated code
• Fixes breaking changes in DICOMWSIReader to make it compatible with latest wsidicom version. #539, #580
• Updates shapely dependency to version >=2.0.0 and fixes any breaking changes.
• Fixes bug with DictionaryStore.bquery and geometry=None, i.e. only a where predicate given.
  • Partly Fixes #532 raised by @blaginin
• Fixes local tests for Windows/Linux
• Fixes flake8, deepsource errors.
• Uses logger instead of warnings and print statements to properly log runs.

Development related changes

• Upgrades dependencies which are dependent on Python 3.7
• Moves requirements*.txt files to requirements folder
• Removes tox
• Uses pyproject.toml for bdist_wheel, pytest and isort
• Adds joblib and numba as dependencies.

Full Changelog: v1.3.3...v1.4.0

TIAToolbox 1.3.3

Major Updates and Feature Improvements

• Restricts dependency versions for long term stability of the current version

Changes to API

None

Bug Fixes and Other Changes

• Fix bug related to reading scikit-image

Development related changes

• Restricts dependency versions for compatibility

TIAToolbox 1.3.2

Major Updates and Feature Improvements

• None

Changes to API

• None

Bug Fixes and Other Changes

• Fix bug related to reading DICOM files

Development related changes

• Restricts wsidicom version to <0.7.0 for compatibility

TIAToolbox 1.3.1

Major Updates and Feature Improvements

• Adds NuClick architecture #449
• Adds Annotation Store Reader #476
• Adds DFBR method for registering pair of images #510

Changes to API

• Adds a sample SVS loading function tiatoolbox.data.small_svs() to the data module #517

Bug Fixes and Other Changes

• Simplifies example notebook for image reading for better readability
• Restricts Shapely version to <2.0.0 for compatibility

Development related changes

• Adds GitHub workflow for automatic generation of docker image
• Updates dependencies
• Updates bump2version config
• Enables flake8 E800 checks for commented codes.
• Fixes several errors generated by DeepSource.
• Prevent test dumping file to root
• Removes duplicate functions to generate parameterized test scenarios

Note: Please note that Python 3.7 support will be removed after this release. We plan to add support for Python 3.11 in the next release.