NVIDIA has launched a cloud service for medical imaging artificial intelligence (AI), which simplifies and accelerates the creation of real data and the training of professional AI models through a fully managed, cloud-based application programming interface (API). The NVIDIA MONAI Cloud API, announced at the Radiological Society of North America (RSNA) annual meeting in Chicago this week, provides developers and platform providers with a fast way to integrate AI into their medical imaging products using pre-trained base models and enterprise AI workflows. These APIs are built on the open-source MONAI project established by NVIDIA and King's College London. Medical imaging is crucial in the healthcare field, accounting for approximately 90% of medical data. It is used by radiologists and clinicians for screening, diagnosis, and intervention, by pharmaceutical researchers to assess patients' response to new drugs, and by medical device manufacturers to provide real-time decision support. The scale of work in each field requires a medical imaging-specific AI factory, an enterprise-level platform that provides large-scale data management, creates real annotations, accelerates model development, and establishes seamless AI application deployment. With the NVIDIA MONAI Cloud API, solution providers can more easily integrate AI into their medical imaging platforms, enabling them to provide powerful tools for radiologists, researchers, and clinical trial teams, and establish domain-specific AI factories. These APIs are available through the NVIDIA DGX Cloud AI supercomputing service for early access. The NVIDIA MONAI Cloud API has already been integrated into Flywheel, a leading medical imaging data and AI platform that supports end-to-end workflows for AI development. Medical imaging annotation companies, including RedBrick AI, and machine learning operations (MLOps) platform providers, including Dataiku, are also preparing to integrate the NVIDIA MONAI Cloud API into their products. Providing ready-to-use annotations and training for medical imaging Building efficient and cost-effective AI solutions requires a powerful, domain-specific development foundation, including software targeting, scalable multi-node systems, and state-of-the-art research optimization. It also requires high-quality real data, which can be challenging and time-consuming, especially for 3D medical images that require expert knowledge for annotation. The NVIDIA MONAI Cloud API provides interactive annotation driven by the VISTA-3D (Visual Imaging Segmentation and Annotation) base model. It is designed for continuous learning, which is the ability to improve AI model performance based on user feedback and new data. VISTA-3D has been trained on annotated image data sets from 3D CT scans of over 4,000 patients, covering various diseases and body parts. It accelerates the creation of 3D segmentation masks for medical image analysis. Through continuous learning, the annotation quality of the AI model improves over time. To further accelerate AI training, this version includes APIs that seamlessly enable the construction of custom models based on pre-trained MONAI models. The NVIDIA MONAI Cloud API also includes Auto3DSeg, which automates hyperparameter tuning and AI model selection for a given 3D segmentation task, simplifying the model development process. NVIDIA researchers recently won four challenges at the MICCAI Medical Image Computing and Computer Assisted Intervention conference using Auto3DSeg. These challenges include AI models for analyzing 3D CT scans of kidneys and hearts, brain MRIs, and 3D ultrasound of the heart. Solution providers and platform builders embrace NVIDIA MONAI Cloud API Medical imaging solution providers and machine learning platforms are using the NVIDIA MONAI Cloud API to provide critical AI insights and accelerate their work for their customers. Flywheel has integrated MONAI through NVIDIA AI Enterprise and now offers the NVIDIA MONAI Cloud API to accelerate the planning, annotation, analysis, and training of medical imaging. The Minneapolis-based company's centralized, cloud-based platform empowers biopharmaceutical companies, life science organizations, healthcare service providers, and academic medical centers to identify, plan, and train medical imaging data and develop trusted AI. "The NVIDIA MONAI Cloud API reduces the cost of building high-quality AI models for radiology, disease research, and clinical trial data evaluation," said Dan Marcus, Chief Scientist at Flywheel. "By adding interactive annotation and automated segmentation capabilities to the cloud API, customers of our medical imaging AI platform can accelerate AI model development and deliver innovative solutions faster." Annotation and viewer solution providers, including Redbrick AI, Radical Imaging, V7 Labs, and Centaur Labs, are also using the NVIDIA MONAI Cloud API to bring AI-assisted annotation and training capabilities to market faster without the need to host and manage AI infrastructure. RedBrick AI is integrating the VISTA-3D model provided by the NVIDIA MONAI Cloud API to offer cloud-based interactive annotation to its medical device customers, supporting distributed clinical teams. "VISTA-3D allows our customers to quickly build models across different modalities and conditions," said Shivam Sharma, CEO of RedBrick AI. "This base model has good generalization capabilities, making it easy to fine-tune for various clinical applications and obtain accurate and reliable segmentation results." To simplify enterprise AI model development, MLOps platform builders, including Dataiku, ClearML, and Weights & Biases, are also exploring the use of the NVIDIA MONAI Cloud API. Dataiku plans to integrate the NVIDIA MONAI Cloud API to further simplify AI model creation for medical imaging applications. "With the NVIDIA MONAI Cloud API, Dataiku users will be able to easily use Auto3DSeg, a low-code option that accelerates the development of state-of-the-art segmentation models, connected to a NVIDIA-hosted, GPU-accelerated service through Dataiku's web interface," said Kelci Miclaus, Global Healthcare and Life Sciences Solutions Lead at Dataiku. "This democratizes AI in medical imaging, extending the ability to create and apply AI-driven workflows to data and domain experts." Join the ranks of medical imaging innovators by registering for early access and accelerating AI development with the NVIDIA MONAI Cloud API.