ITKIT: Feasible Medical Image Operation based on SimpleITK API

PyPI:

Docker Image:

Readthedocs:

ITKIT is a comprehensive toolkit for medical image preprocessing and analysis, providing command-line tools, a GUI application, and deep learning framework integrations for CT and MRI image processing.

✨ Core Features

• 🔧 Preprocessing Tools: Check, resample, orient, patch, augment, and convert medical images
• 🖥️ GUI Application: User-friendly PyQt6 interface for all operations
• 🧠 Neural Networks: 16+ state-of-the-art segmentation models (SegFormer, MedNeXt, VMamba, etc.)
• 🔌 Framework Support: Integration with OpenMMLab, MONAI, TorchIO, and PyTorch Lightning
• 🏥 3D Slicer Extension: Native extension for running inference directly in 3D Slicer
• 📊 Dataset Conversion: Scripts for 12+ popular medical imaging datasets
• ⚡ High Performance: Multiprocessing support for faster batch processing

🚀 Quick Start

Installation

pip install itkit
# Optional: Install GUI support
pip install "itkit[gui]"

We also provide a docker image:

docker pull mgam1009/itkit:latest

[!NOTE] ITKIT experiences BC in v4. The previous version is v3.6.0rc0.

Basic Usage

# Check dataset integrity
itk_check check /path/to/dataset --min-spacing 0.5 0.5 0.5

# Resample images to uniform spacing
itk_resample dataset /path/to/source /path/to/output --spacing 1.0 1.0 1.0 --mp

# Launch GUI application
itkit-app

📚 Documentation

Full documentation is available at docs

You can also find the docs on readthedocs.

Quick Links

• Installation Guide - Detailed installation instructions
• Quick Start Tutorial - Get started in 5 minutes
• Dataset Structure - Required dataset format
• Preprocessing Tools - Complete command reference
• Framework Integration - OpenMMLab, MONAI, TorchIO
• 3D Slicer Integration - Run inference in 3D Slicer
• Neural Network Models - Available segmentation models
• Supported Datasets - Dataset conversion scripts
• FAQ & Troubleshooting - Common issues and solutions
• Contributing Guide - How to contribute

🛠️ Command-Line Tools

ITKIT provides several preprocessing commands:

Command	Description
itk_check	Validate dataset integrity (spacing, size, pairing)
itk_resample	Resample images to target spacing or size
itk_orient	Orient images to standard directions (LPI, RAS, etc.)
itk_patch	Extract patches for training
itk_aug	Data augmentation with random rotations
itk_extract	Extract specific classes from segmentation maps
itk_convert	Convert between formats (MHA, NIfTI, NRRD) and frameworks (MONAI, TorchIO)
itkit-app	Launch graphical user interface
itk_slicer	Start ITKIT inference server for 3D Slicer integration
mmrun	Run OpenMMLab experiments

Use --help with any command for detailed usage information.

🖼️ GUI Application

Install GUI support and launch:

pip install "itkit[gui]"
itkit-app

# Adjust DPI if needed
QT_SCALE_FACTOR=2 itkit-app

📦 Optional Features

ITKIT provides optional dependency groups:

pip install "itkit[gui]"        # GUI application (PyQt6)
pip install "itkit[advanced]"   # Deep learning frameworks (OpenMMLab)
pip install "itkit[dev]"        # Development tools (pytest, black, mypy)
pip install "itkit[pathology]"  # Pathology image processing
pip install "itkit[onnx]"       # Model deployment (ONNX, TensorRT)

📖 Citation

If you use ITKIT in your research, please cite:

@misc{ITKIT,
    author = {Yiqin Zhang},
    title = {ITKIT: Feasible Medical Image Operation based on SimpleITK API},
    year = {2025},
    url = {https://github.com/MGAMZ/ITKIT}
}

🤝 Contributing

We welcome contributions! Please see our Contributing Guide for:

• Development setup
• Code style guidelines
• Testing requirements
• Pull request process

📄 License

ITKIT is released under the MIT License.

📧 Contact

For questions or suggestions, reach out at: 312065559@qq.com

🌟 Acknowledgments

ITKIT builds upon:

• SimpleITK - Medical image processing
• OpenMMLab - Deep learning framework
• MONAI - Medical imaging AI
• TorchIO - Medical image preprocessing

---

⭐ Star us on GitHub if you find ITKIT useful!