• 1) Introduction to MONAI and the MONAI Community - Stephen Aylward

  2) MONAI medical deep learning capabilities and roadmap - Jorge Cardoso 

  3) MONAI and clinical workflows: Clara and federated learning - Brad Genereaux

Abstract: The MONAI framework is the open-source foundation for medical deep learning that began as a collaboration between King's College London and NVidia and that is now being led by Project MONAI. The MONAI framework is a freely available, community-supported, PyTorch-based framework for deep learning that is specifically designed for healthcare imaging AI. It is an extensible framework that is meant to include the state-of-the-art neural networks, data augmentation methods, and training systems for medical imaging.  The framework also includes streamlined access to medical imaging grand challenges and is designed to promote reproducibility and open science. Project MONAI is developing the MONAI framework and associated tutorials and course material as part of its commitment to creating an inclusive community of AI researchers for the development and exchange of best practices for AI in healthcare imaging, for academia and industry. Project MONAI and the MONAI framework would not have been possible without existing toolkits such as Nvidia Clara Train, NiftyNet, DLTK, and DeepNeuro and the communities that have grown around them.