Nvidia and Partners Release Open Source AI Framework for Medical Research
Nvidia and King's College London launch Project MONAI, an open source framework designed to improve reproducibility in healthcare AI research. The alpha release integrates with PyTorch and focuses on tasks like 3D organ segmentation and brain MRI classification.
Nvidia and King's College London release the open source alpha version of Project MONAI, a specialized AI framework for healthcare research. MONAI, which stands for Medical Open Network for AI, is now available on GitHub and is optimized to work seamlessly with deep learning frameworks like PyTorch and Ignite.
The primary goal of the MONAI framework is to ensure reproducibility in medical AI experiments so researchers can successfully build upon each other's work. The alpha release includes features like controlled data augmentation during training to guarantee consistent results, addressing a critical need for validation in medical imaging.
The framework currently provides examples for complex tasks such as 3D organ segmentation for abdominal CT scans and 2D classification for brain MRI imagery. Nvidia plans to integrate MONAI into its Clara medical imaging tools in the future, while global collaborators from institutions in China, Germany, and the U.S. continue to develop the platform.