Skip to content

Installation Guide

Requirements

ITKIT is tested on:

  • Python >= 3.10
  • numpy >= 2.2.6
  • SimpleITK >= 2.5.0

Lower versions may work but are not guaranteed.

Installation Methods

The simplest way to install ITKIT is via pip:

pip install itkit

From Source

First, clone the repository:

git clone https://github.com/MGAMZ/ITKIT.git
cd ITKIT

Then, install the package:

pip install .

Optional Dependencies

ITKIT provides several optional dependency groups for different use cases:

Development Tools

For development and testing:

pip install "itkit[dev]"

Includes: pytest, pylint, black, isort, mypy, and other development tools.

Advanced Features

For advanced image processing and deep learning features:

pip install "itkit[advanced]"

Includes: torchio, onedl-mmcv, onedl-mmengine, onedl-mmsegmentation.

Pathology Support

For pathology image processing features:

pip install "itkit[pathology]"

Includes: opensdpc, openslide-python, openslide-bin.

GUI Support

For the PyQt desktop graphical user interface:

pip install "itkit[gui]"

Includes: PyQt6.

Web Interface

For the browser-based web interface (itkit-web):

pip install "itkit[web]"

Includes: flask, flask-cors.

ONNX Deployment

For model deployment with ONNX:

pip install "itkit[onnx]"

Includes: onnx, onnxruntime, tensorrt, and related tools.

Combined Installation

You can install multiple optional dependencies at once:

pip install "itkit[dev,advanced,gui]"

Deep Learning Framework Compatibility

If you plan to run deep learning tasks, we recommend installing monai to avoid potential dependency issues:

pip install --no-deps monai

Note: The itk_convert monai and itk_convert torchio commands do not require the monai or torchio Python packages to perform the conversion. Install these packages only if you plan to run MONAI or TorchIO-based deep learning workflows.

OpenMMLab Integration

Important: The upstream OpenMMLab project has gradually fallen out of maintenance. ITKIT now recommends users to use the OneDL redistribution of OpenMMLab instead:

  • OneDL-mmengine
  • OneDL-mmcv
  • OneDL-mmsegmentation

These are included in the advanced optional dependencies.

Verifying Installation

After installation, verify that ITKIT is properly installed:

# Check version
python -c "import itkit; print(itkit.__version__)"

# List available commands
itk_check --help
itk_resample --help