GaNDLF

The Generally Nuanced Deep Learning Framework (GaNDLF) for reproducible segmentation and classification.

Why use GaNDLF?

GaNDLF was developed to lower the barrier to AI, enabling reproducibility, translation, and deployment. As an out-of-the-box solution, GaNDLF alleviates the need to build from scratch. Users may kickstart their project by modifying only a configuration (config) file that provides guidelines for the envisioned pipeline and CSV inputs that describe the training data.

Range of GaNDLF functionalities:

• Supports multiple
  • Deep Learning model architectures
  • Channels/modalities
  • Prediction classes
• Robust data augmentation, courtesy of TorchIO and Albumentations
• Built-in cross validation, with support for parallel HPC-based computing
• Multi-GPU (on the same machine) training
• Leverages robust open source software
• Zero-code needed to train robust models
• Low-code requirement for customization
• Automatic mixed precision support

Table of Contents

• Getting Started
• Application Setup
• Usage
  • Customize the training and inference
• Extending GaNDLF
• ITCR Connectivity
• FAQ
• Acknowledgements

Citation

Please cite the following article for GaNDLF (full paper):

@article{pati2023gandlf,
    author={Pati, Sarthak and Thakur, Siddhesh P. and Hamamc{\i}, {\.{I}}brahim Ethem and Baid, Ujjwal and Baheti, Bhakti and Bhalerao, Megh and G{\"u}ley, Orhun and Mouchtaris, Sofia and Lang, David and Thermos, Spyridon and Gotkowski, Karol and Gonz{\'a}lez, Camila and Grenko, Caleb and Getka, Alexander and Edwards, Brandon and Sheller, Micah and Wu, Junwen and Karkada, Deepthi and Panchumarthy, Ravi and Ahluwalia, Vinayak and Zou, Chunrui and Bashyam, Vishnu and Li, Yuemeng and Haghighi, Babak and Chitalia, Rhea and Abousamra, Shahira and Kurc, Tahsin M. and Gastounioti, Aimilia and Er, Sezgin and Bergman, Mark and Saltz, Joel H. and Fan, Yong and Shah, Prashant and Mukhopadhyay, Anirban and Tsaftaris, Sotirios A. and Menze, Bjoern and Davatzikos, Christos and Kontos, Despina and Karargyris, Alexandros and Umeton, Renato and Mattson, Peter and Bakas, Spyridon},
    title={GaNDLF: the generally nuanced deep learning framework for scalable end-to-end clinical workflows},
    journal={Communications Engineering},
    year={2023},
    month={May},
    day={16},
    volume={2},
    number={1},
    pages={23},
    issn={2731-3395},
    doi={10.1038/s44172-023-00066-3},
    url={https://doi.org/10.1038/s44172-023-00066-3}
}

Contact

GaNDLF developers can be reached via the following ways:

• GitHub Discussions
• Email