We are proud to announce that Amazon has integrated the MONAI open source deep learning platform for medical image analysis into its AWS SageMaker system! This combination of outstanding tools paves the way for small and large businesses to more easily, seamlessly, and cost-effectively utilize the power of deep learning.
SageMaker is a cloud-based Amazon service for creating and deploying machine learning systems that can be easily scaled based on demand. It is intended to be used by data scientists and developers with any level of machine learning experience. It offers a variety of modules that can be easily combined to create scalable, end-to-end machine learning research systems and commercial products. Many of the major risks associated with exploring and deploying deep learning systems are mitigated by SageMaker, e.g., businesses can avoid expensive deep learning hardware purchases that may not be consistently utilized and that may quickly become obsolete.
MONAI is now available as a SageMaker module, enabling the rapid development and deployment of medical image analysis machine learning systems. It is intended for biological as well as pre-clinical and clinical research and application development. It provides numerous methods and examples for applying cutting-edge deep learning networks to common medical image analysis problems. These methods and examples have been curated to illustrate “best practices” for overcoming many of the unique challenges associated with applying deep learning to medical imaging: having limited training samples, needing to generate reproducible results, and having to smoothly integrate deep learning inference into existing medical research and clinical workflows.
Kitware is helping to guide the development of MONAI as a charter member of its advisory board and as a contributor to its code based via ITK. We can also help you design, develop, and deliver innovative deep learning solutions (powered by MONAI and running on SageMaker and other systems) to your customers, collaborators, and employees. Contact us for more information.