Nvidia's MONAI Framework for DL in Health Care Imaging Now in Alpha
- By John K. Waters
Nvidia has announced the alpha release of the MONAI open source framework for deep learning (DL) in health care imaging.
Based on the Ignite and PyTorch DL frameworks, MONAI was developed primarily to help health care researchers reproduce their experiments so they can build on each other's work, the company says.
MONAI (Medical Open Network for AI) is a user-friendly framework designed to deliver reproducible results, and it's domain-optimized for the demands of health care data, including the unique formats, resolutions and specialized meta-information of medical images, wrote Kimberly Powell, vice president of Nvidia's Healthcare group, in a blog post. "Our first public release provides domain-specific data transforms, neural network architectures, and evaluation methods to measure the quality of medical imaging models," she wrote.
The modular, open source framework gives researchers the flexibility to customize their DL development without needing to replace their existing workflows with an end-to-end system, Powell explained. "An advanced researcher could, for instance, adopt MONAI code for data preprocessing and transformations, and then switch over to an existing AI pipeline for training," she said.
The framework was developed under the auspices of the MONAI Project, which Nvidia started with King's College London to establish "an inclusive community of AI researchers for the development and exchange of best practices for AI in healthcare imaging across academia and enterprise researchers."
NVIDIA and King's College London are leading the initiative in collaboration with an advisory board made up of members from Stanford University, the German Cancer Research Center, Kitware, MGH & BWH Center for Clinical Data Science, the Chinese Academy of Sciences, and the Technical University of Munich.
"Project MONAI has outstanding potential to accelerate the pace of medical imaging AI research," said Stephen Aylward, chair of the MONAI advisory board and a senior director at open source software company Kitware, in a statement. "It provides a high-quality, open-source foundation that is specialized for medical imaging, that welcomes everyone to build upon, and that anyone can use to communicate and compare their ideas."
Santa Clara, Calif.-based Nvidia has grown in a few short years into a multibillion-dollar company by marketing its graphical processing units (GPUs) to datacenter operators as the right silicon for processing the flood of data demanded by a new generation of AI-oriented applications. The company made more moves into the AI and robotics space last year with the release of its Isaac SDK. The company described the open source offering as "a developer toolbox for accelerating the development and deployment of AI-powered robots."
Available now on GitHub, the framework builds on the best practices from existing tools, such as Nvidia Clara, NiftyNet, DLTK, and DeepNeuro. Future releases of Nvidia Clara, the company's a health care application framework for AI-powered imaging and genomics, will also leverage the MONAI framework.
John K. Waters is the editor in chief of a number of Converge360.com sites, with a focus on high-end development, AI and future tech. He's been writing about cutting-edge technologies and culture of Silicon Valley for more than two decades, and he's written more than a dozen books. He also co-scripted the documentary film Silicon Valley: A 100 Year Renaissance, which aired on PBS. He can be reached at firstname.lastname@example.org.