Welcome to the Imaging Server Kit’s documentation!

The Imaging Server Kit lets you turn Python-based image processing workflows into algorithms that gain extra functionalities.

@sk.algorithm  # <- Turn your function into an algorithm
def my_algo(image, parameter):
  (...)

Server Kit algorithms are versatile objects that allow you run computations in a variety of ways.

For example, you can

• Turn your algorithm into a web server, connect to it and run computations from QuPath, Napari, or Python via HTTP requests.

• Generate a dock widget to run your algorithm interactively in Napari.

• Run your algorithm tile-by-tile on the input image.

• Stream results to inspect them in real-time.

On top of that, you can validate data passed as algorithm parameters, provide samples, and generate documentation for your algorithm that you can share with users.

This documentation will give you a conceptual overview of the package, and walk you through the steps to learn how to create an algorithm, and give you some suggestions of use cases.

Contents

• Getting started
• Creating algorithms
• Providing samples
• Algorithm metadata
• Combining algorithms
• Live updates
• Tiled inference
• Serving algorithms
• Usage in Python
• Using Docker
• Data layers
• Examples

Installation

Install the imaging-server-kit package with pip:

pip install imaging-server-kit

or clone the project and install the development version:

git clone https://github.com/Imaging-Server-Kit/imaging-server-kit.git
cd imaging-server-kit
pip install -e .

Napari plugin

Install the napari-serverkit plugin with pip to enable all Napari-related functionalities.

pip install napari-serverkit

QuPath extention

To use the Imaging Server Kit in QuPath, follow the installation instructions at qupath-extension-serverkit.

License

This software is distributed under the terms of the BSD-3 license.

Citing

If you use imaging-server-kit in the context of scientific publication, you can cite it as below.

BibTeX:

@software{mallory_wittwer_2025_15673152,
  author       = {Mallory Wittwer and Edward Andò and Maud Barthélemy and Florian Aymanns},
  title        = {Imaging-Server-Kit/imaging-server-kit: v0.0.14},
  url          = {https://doi.org/10.5281/zenodo.15673152},
  doi          = {10.5281/zenodo.15673152},
  version      = {v0.0.14},
  year         = 2025,
}

Acknowledgements

We thank the Personalized Health and Related Technologies for funding this project.