HistoKit

HistoKit is a Python package for processing Whole Slide Images (WSI).
It provides a toolkit for common operations in digital pathology workflows.

Features

• Tissue detection — identifies tissue regions within whole-slide images.
• Artifact detection — uses the GrandQC deep learning model.
• Patch extraction — tiles slides into patches at customizable magnification and stride.
• Augmentation & Stain Normalisation — applies common augmentation methods and stain normalisation techniques (using staintools library).

In Development

• Feature extraction — extracts representations using foundation models.
• Visualization — explore high-dimensional features using UMAP.
• Classification with Multiple Instance Learning — classification using various MIL models.

Installation

From remote repository

The first way to use HistoKit is to download the source code from a remote repository. It is recommended to create a virtual environment with Anaconda or Venv. Then it is necessary to install the required dependencies from the requirements.txt file. Which can be done with the following command:

pip install -r requirements.txt

Then OpenSlide library needs to be installed with:

pip install openslide-bin

If everything has been installed correctly, now it is possible to use functions from the HistoKit package after extracting files from a .zip archive. As a local pip package

It is also possible to install the package from .zip archive with pip. To do that repeat the steps described in the previous section, but now you don’t need to extract the files. You just have to install histo_kit from your .zip file with the following command:

pip install {path-to-histokit-archive}.zip

Documentation and Examples

HistoKit Documentation

References

Article describing the first version of HistoKit (1.0.0)

@inproceedings{Merta2026HistoKit,
  author    = {Merta, Julia and Marczyk, Micha{\l}},
  title     = {HistoKit: Fast and Accurate Tissue and Artifact Detection and Data Processing for Whole Slide Histopathological Imaging},
  booktitle = {Proceedings of the 19th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2026), Volume 2: BIODEVICES, BIOIMAGING, BIOINFORMATICS},
  year      = {2026},
  month     = mar,
  address   = {Marbella, Spain},
  publisher = {SCITEPRESS -- Science and Technology Publications, Lda},
  isbn      = {978-989-758-802-0},
  issn      = {2184-4305},
  doi       = {10.5220/0000222000004070},
  url       = {https://doi.org/10.5220/0000222000004070}
}

Post-processing for tissue segmentation

@inproceedings{Marczyk2025PostProcessing,
  author    = {Marczyk, M. and Wrobel, A. and Merta, J. and Polanska, J.},
  title     = {Post-Processing of Thresholding or Deep Learning Methods for Enhanced Tissue Segmentation of Whole-Slide Histopathological Images},
  booktitle = {Proceedings of the 18th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOIMAGING 2025), Volume 1},
  pages     = {229--238},
  publisher = {SciTePress},
  year      = {2025},
  isbn      = {978-989-758-731-3},
  doi       = {10.5220/0013174700003911},
  url       = {https://doi.org/10.5220/0013174700003911},
  note      = {GitHub: https://github.com/ZAEDPolSl/WSI_TissueSeg}
}

GrandQC artifact detection model

@article{Weng2024GrandQC,
  author  = {Weng, Z. and Seper, A. and Pryalukhin, A. and others},
  title   = {GrandQC: A comprehensive solution to quality control problem in digital pathology},
  journal = {Nature Communications},
  volume  = {15},
  pages   = {10685},
  year    = {2024},
  doi     = {10.1038/s41467-024-54769-y},
  url     = {https://doi.org/10.1038/s41467-024-54769-y},
  note    = {GitHub: https://github.com/cpath-ukk/grandqc}}