NiftyNet 0.2.0.post1

NiftyNet is a TensorFlow-based [1] open-source convolutional neural networks (CNN) platform for research in medical image analysis and image-guided therapy. NiftyNet’s modular structure is designed for sharing networks and pre-trained models. Using this modular structure you can:

• Get started with established pre-trained networks using built-in tools
• Adapt existing networks to your imaging data
• Quickly build new solutions to your own image analysis problems

NiftyNet is a consortium of research groups (WEISS – Wellcome EPSRC Centre for Interventional and Surgical Sciences, CMIC – Centre for Medical Image Computing, HIG – High-dimensional Imaging Group), where WEISS acts as the consortium lead.

[1]	Please install the appropriate TensorFlow PyPI package (tensorflow or tensorflow-gpu) before executing pip install niftynet – see the instructions on the NiftyNet source code repository for details.

Features

NiftyNet currently supports medical image segmentation and generative adversarial networks. NiftyNet is not intended for clinical use. Other features of NiftyNet include:

• Easy-to-customise interfaces of network components
• Sharing networks and pretrained models
• Support for 2-D, 2.5-D, 3-D, 4-D inputs [2]
• Efficient discriminative training with multiple-GPU support
• Implementation of recent networks (HighRes3DNet, 3D U-net, V-net, DeepMedic)
• Comprehensive evaluation metrics for medical image segmentation

[2]	2.5-D: volumetric images processed as a stack of 2D slices; 4-D: co-registered multi-modal 3D volumes

Getting started

Installation

Please follow the installation instructions.

Examples

Please see the NiftyNet demos.

Network (re-)implementations

Please see the list of network (re-)implementations in NiftyNet.

API documentation

The API reference is available on Read the Docs.

Contributing

Please see the contribution guidelines on the NiftyNet source code repository.

Useful links

NiftyNet website

NiftyNet source code on CmicLab

NiftyNet source code mirror on GitHub

NiftyNet mailing list: [e-mail]

Citing NiftyNet

If you use NiftyNet in your work, please cite Li et. al. 2017:

Li W., Wang G., Fidon L., Ourselin S., Cardoso M.J., Vercauteren T. (2017) On the Compactness, Efficiency, and Representation of 3D Convolutional Networks: Brain Parcellation as a Pretext Task. In: Niethammer M. et al. (eds) Information Processing in Medical Imaging. IPMI 2017. Lecture Notes in Computer Science, vol 10265. Springer, Cham. DOI: 10.1007/978-3-319-59050-9_28

BibTeX entry:

@InProceedings{niftynet17,
  author = {Li, Wenqi and Wang, Guotai and Fidon, Lucas and Ourselin, Sebastien and Cardoso, M. Jorge and Vercauteren, Tom},
  title = {On the Compactness, Efficiency, and Representation of 3D Convolutional Networks: Brain Parcellation as a Pretext Task},
  booktitle = {International Conference on Information Processing in Medical Imaging (IPMI)},
  year = {2017}
}

Licensing and copyright

Copyright 2017 University College London and the NiftyNet Contributors. NiftyNet is released under the Apache License, Version 2.0. Please see the LICENSE file in the NiftyNet source code repository for details.

Acknowledgements

This project is grateful for the support from the Wellcome Trust, the Engineering and Physical Sciences Research Council (EPSRC), the National Institute for Health Research (NIHR), the Department of Health (DoH), University College London (UCL), the Science and Engineering South Consortium (SES), the STFC Rutherford-Appleton Laboratory, and NVIDIA.