3.6 billion. That’s about how much medical imaging exams are carried out each year around the world to diagnose, monitor and treat various conditions.
Speeding up the processing and evaluation of all those X-rays, CT scans, MRIs and ultrasounds is essential to help doctors manage their workload and improve health outcomes.
This is why NVIDIA introduced MONAIwhich serves as an open source research and development platform for AI applications used in medical imaging and beyond. MONAI brings together doctors and data scientists to unleash the power of medical data to create deep learning models and deployable applications for medical AI workflows.
This week at the annual meeting of RSNA, the Radiological Society of North America, NVIDIA announced that Siemens Healthineers has adopted MONAI Deploy, a module within MONAI that bridges the gap between research and clinical production, to increase the speed and efficiency of AI workflow integration. for medical imaging in clinical deployments.
With more than 15,000 medical device installations worldwide, Siemens Healthineers Syngo Carbon And syngo.via Enterprise imaging platforms help clinicians better read and extract information from medical images from many sources.
Developers typically use a variety of frameworks when creating AI applications. This makes it difficult to deploy their applications in clinical environments.
With just a few lines of code, MONAI Deploy creates AI applications that can run anywhere. It is a tool for developing, packaging, testing, deploying and running medical AI applications in clinical production. Its use streamlines the process of developing and integrating medical imaging AI applications into clinical workflows.
MONAI Deploy on the Siemens Healthineers Platform has significantly accelerated the AI onboarding process, enabling users to transfer trained AI models into real-world clinical environments in just a few clicks, whereas previously it took months. This helps researchers, entrepreneurs and startups get their applications to radiologists faster.
“By accelerating the deployment of AI models, we are enabling healthcare organizations to harness and benefit from the latest advances in AI-based medical imaging faster than ever before,” said Axel Heitland, head of digital technologies and research at Siemens Healthineers. “With MONAI Deploy, researchers can quickly adapt AI models and move innovations from the laboratory to clinical practice, providing thousands of clinical researchers around the world with access to AI-based advances directly to their syngo.via and Syngo Carbon imaging platforms. »
Enhanced with applications developed by MONAI, these platforms can significantly streamline AI integration. These applications can be easily provided and used on the Siemens Healthineers Digital Marketwhere users can browse, select and seamlessly integrate them into their clinical workflows.
The MONAI ecosystem drives innovation and adoption
Now celebrating its fifth anniversary, MONAI has seen over 3.5 million downloads, 220 contributors from around the world, acknowledgments in over 3,000 publications, 17 MICCAI Challenge wins, and use in numerous clinical products.
The latest version of MONAI — v1.4 — includes updates that provide researchers and clinicians with even more opportunities to benefit from MONAI innovations and contribute to Siemens Healthineers Syngo Carbon, syngo.via and Siemens Healthineers Digital Marketplace.
Updates to MONAI v1.4 and related NVIDIA products include new base models for medical imaging, which can be customized in MONAI and deployed as NVIDIA NIM Microservices. The following models are now generally available as NIM microservices:
- BUTSI (Medical AI for Synthetic Imaging) is a basic latent diffusion generative AI model that can simulate high-resolution full-frame 3D CT images and their anatomical segmentations.
- VISTA-3D is a core model for CT image segmentation that provides accurate, out-of-the-box performance covering more than 120 major organ classes. It also provides efficient zero-shot adaptation capabilities for learning to segment new structures.
Alongside the main features of MONAI 1.4, the new MONAI multimodal model, or M3, is now accessible via MONAI VLM GitHub repository. M3 is a framework that extends any multimodal LLM with medical AI experts such as trained AI models from MONAI’s Model Zoo. The power of this new framework is demonstrated by the VILA-M3 base model now available on Hugging Face, delivering industry-leading X-ray image co-pilot performance.
MONAI connects hospitals, healthcare startups and research institutes
Leading healthcare institutions, academic medical centers, startups and software providers around the world are adopting and advancing MONAI, including:
- German Cancer Research Center leads MONAI’s Benchmarks and Metrics Working Group, which provides metrics for measuring AI performance and guidelines for how and when to use these metrics.
- Nadeem Laboratory from Memorial Sloan Kettering Cancer Center (MSK) has pioneered the cloud deployment of multiple AI-assisted annotation pipelines and inference modules for pathology data using MONAI.
- University of Colorado School of Medicine the faculty has developed MONAI-based ophthalmological tools to detect retinal diseases using various imaging modalities. The university is also leading some of the original federated learning developments and clinical demonstrations using MONAI.
- Mathematical Work has integrated MONAI Label into its medical imaging toolkit, bringing medical imaging AI and AI-assisted annotation capabilities to thousands of MATLAB users engaged in medical and biomedical applications in academia and industrial.
- GSK explores MONAI base models such as VISTA-3D and VISTA-2D for image segmentation.
- Flying offers a platform, which includes MONAI, to streamline imaging data management, automate research workflows, and enable AI development and analysis, which scales to the needs of research institutions and life sciences organizations.
- Alara Imaging published his work on integrating MONAI core models such as VISTA-3D with LLMs such as Llama 3 at the 2024 Society for Imaging Informatics in Medicine conference.
- RadImageNet explores the use of MONAI’s M3 framework to develop cutting-edge vision language models that use MONAI’s expert image AI models to generate high-quality radiology reports.
- Kitchen utensils provides professional software development services around MONAI, helping to integrate MONAI into custom workflows for device manufacturers as well as regulatory approved products.
Researchers and enterprises also use MONAI on cloud service providers to run and deploy scalable AI applications. Cloud platforms providing access to MONAI include AWS HealthImaging, Google Cloud, Precision Imaging Network, part of Microsoft Cloud for Healthcare, and Oracle Cloud Infrastructure.
See information statements on syngo.via, Syngo Carbon and products in the Digital Market.