ITKIT Documentation

Welcome to the ITKIT documentation! ITKIT is a user-friendly toolkit built on SimpleITK and Python, designed for common data preprocessing operations in data-driven CT medical image analysis.

📖 Table of Contents

Getting Started

• Installation Guide - Install ITKIT and its dependencies
• Dataset Structure - Understand the required dataset format

Processing Tools

• itk_check - Image checking and validation
• itk_resample - Resampling to target spacing/size
• itk_orient - Image re-orientation
• itk_patch - Patch extraction
• itk_aug - Data augmentation
• itk_extract - Label extraction
• itk_combine - Label merging and intersection
• itk_convert - Format conversion
• itk_infer - Batch inference with MMEngine/ONNX backends
• itk_evaluate - Segmentation evaluation with comprehensive metrics

Advanced Topics

• Framework Integration - Integration with deep learning frameworks
• OpenMMLab extensions
• MONAI integration
• TorchIO integration

• PyTorch Lightning support

• 3D Slicer Integration - 3D Slicer extension for inference

• Install and configure the Slicer extension
• Run inference directly in 3D Slicer
• MMEngine and ONNX backend support

• Sliding window inference for large volumes

• Web Interface - Browser-based GUI (itkit-web)

• Full-featured alternative to the PyQt desktop GUI
• Embedded file browser and per-tool parameter panels
• Real-time log streaming and progress display

• REST API for scripted access

• Neural Network Models - State-of-the-art segmentation models

• Transformer-based models (SegFormer, UNETR, DA-TransUNet)
• State space models (VMamba, SwinUMamba, SegMamba)

• CNN-based models (MedNeXt, UNet3+, DconnNet)

• Supported Datasets - Dataset conversion scripts

• AbdomenCT-1K, BraTS 2024, KiTS23
• FLARE 2022/2023, TotalSegmentator
• LiTS, LUNA16, CTSpine1K
• And more...

Community

• Contributing Guide
• Development setup
• Code style guidelines
• Submission process
• Release policy

🚀 Key Features

• 🔧 Feasible Operations: Simple command-line interface for complex ITK operations
• 🖥️ GUI Support: PyQt6-based graphical interface for easier interaction
• 🔌 Framework Integration: Seamlessly works with MONAI, TorchIO, and OpenMMLab
• 🧠 Comprehensive Models: State-of-the-art medical segmentation networks
• 📊 Multiple Datasets: Conversion scripts for 12+ popular medical imaging datasets
• ⚡ High Performance: Multiprocessing support for faster preprocessing
• 🎨 Flexible: Works with multiple file formats (MHA, NIfTI, NRRD, DICOM)

📝 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}
}

📧 Contact

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

📄 License

ITKIT is released under the MIT License.