Monday, March 4, 2024

Top 5 This Week

Related Posts

Unlock Efficiency NVIDIA’s MONAI Cloud Simplifies Imaging AI

Through fully controlled, cloud-based application programming interfaces, NVIDIA created a cloud service for medical imaging AI to further expedite and automate the generation of ground-truth data and training of specialized AI models.

Using pretrained foundation models and AI workflows for enterprises, the NVIDIA MONAI cloud APIs, which were unveiled at the Radiological Society of North America’s annual meeting this week in Chicago, offer developers and platform providers a quicker way to incorporate AI into their medical imaging offerings. The open-source MONAI project, started by King’s College London and NVIDIA, serves as the foundation for the APIs.

With medical imaging accounting for almost 90% of healthcare data, it is essential to the industry. Medical device manufacturers utilize it to give real-time decision assistance; radiologists and physicians use it for screening, diagnosis, and intervention; biopharma researchers use it to assess how clinical trial patients react to novel medications.

Due to the volume of work in each of these fields, a medical imaging-specific AI factory is needed, as well as an enterprise-grade platform that manages enormous amounts of data, produces ground-truth annotations, expedites the construction of models, and ensures the smooth deployment of AI applications.

Solution providers may more readily incorporate AI into their medical imaging platforms with NVIDIA MONAI cloud APIs, giving them the ability to offer radiologists, researchers, and clinical trial teams enhanced resources to create AI factories that are domain-specific. Early access to the APIs is offered via the NVIDIA DGX Cloud AI supercomputing service.

Flywheel, a top medical imaging data and AI platform that facilitates end-to-end workflows for AI development, is connected with the NVIDIA MONAI cloud API. Developers at firms that provide machine learning operations (MLOps) platforms, such as Dataiku, and medical image annotation companies, like RedBrick AI, are well-positioned to incorporate NVIDIA MONAI cloud APIs into their products.

Annotation & Training for Medical Imaging that is Ready to Run

A strong, domain-specific development foundation, including of state-of-the-art research, scalable multi-node systems, and full-stack software optimizations, is necessary to build effective and economical AI solutions. Additionally, it needs high-quality ground-truth data, which can be difficult and time-consuming to collect, especially for 3D medical images that need to be annotated by experts.

The VISTA-3D (Vision Imaging Segmentation and Annotation) foundation model powers interactive annotation in the NVIDIA MONAI cloud APIs. It was designed with continuous learning in mind, a feature that enhances the performance of AI models in response to human input and fresh data.

VISTA-3D is trained on a collection of annotated pictures from over 4,000 3D CT scans of different body parts and disorders, which speeds up the process of creating 3D segmentation masks for medical image analysis. The AI model’s annotation quality becomes better with continued learning.

This release provides APIs that enable it simple to create bespoke models based on MONAI pretrained models, which will speed up AI training even further. Auto3DSeg is another feature of the NVIDIA MONAI cloud APIs that streamlines the model creation process by automating hyperparameter tuning and AI model selection for a specific 3D segmentation assignment.

Recently, NVIDIA researchers used Auto3DSeg to win four challenges at the MICCAI medical imaging conference. These included artificial intelligence models to evaluate 3D cardiac and kidney CT scans, brain MRIs, and 3D ultrasounds of the heart.

Platform builders and solution providers adopt NVIDIA MONAI Cloud APIs

NVIDIA MONAI cloud APIs are being used by machine learning platforms and suppliers of medical imaging solutions to supply their clients with extremely valuable AI insights that expedite their job.

In order to expedite medical image curation, labeling analysis, and training, Flywheel has integrated MONAI with NVIDIA AI Enterprise and is currently providing NVIDIA MONAI cloud APIs. The Minneapolis-based company uses a centralized cloud-based platform to identify, collect, and train medical imaging data for the creation of reliable artificial intelligence. This platform is used by biopharma companies, life science organizations, healthcare providers, and academic medical institutes.

Dan Marcus, chief scientific officer of Flywheel, stated that “NVIDIA MONAI cloud APIs lower the cost of building high-quality AI models for radiology, disease research, and the evaluation of clinical trial data.” The integration of cloud APIs for automated segmentation and interactive annotation enables our medical imaging AI platform’s clients to construct AI models more quickly and produce creative solutions.

NVIDIA MONAI cloud APIs will also be used by annotation and viewer solution providers, such as Redbrick AI, Radical Imaging, V7 Labs, and Centaur Labs, to expedite the release of AI-assisted annotation and training capabilities, all without the need to host and maintain the AI infrastructure in-house.

RedBrick AI is providing interactive cloud annotation for its medical device customers that support distributed teams of physicians by integrating the VISTA-3D model made available through NVIDIA MONAI cloud APIs.

RedBrick AI CEO Shivam Sharma stated, “VISTA-3D enables our clients to quickly build models across various modalities and conditions.” “With accurate and dependable segmentation results, the foundation model can be easily adjusted for a range of clinical applications due to its generalizability.”

MLOps platform builders like Dataiku, ClearML, and Weight & Biases are also looking into using NVIDIA MONAI cloud APIs to speed up the construction of enterprise AI models.

To make the process of creating AI models for medical imaging applications even easier, Dataiku intends to use the cloud APIs for NVIDIA MONAI.

Through Dataiku’s web interface connected to an NVIDIA-hosted, GPU-accelerated service, “Auto3DSeg, a low-code option to accelerate the development of state-of-the-art segmentation models, would be easily used by Dataiku users with NVIDIA MONAI cloud APIs,” stated Kelci Miclaus, global head of AI health and life sciences solutions at Dataiku. “By giving data and domain experts the ability to create and implement AI-driven workflows, this democratizes AI in biomedical imaging.”

Register for early access to join the medical imaging pioneers who are using NVIDIA MONAI cloud APIs to accelerate AI research.

agarapuramesh
agarapurameshhttps://govindhtech.com
Agarapu Ramesh was founder of the Govindhtech and Computer Hardware enthusiast. He interested in writing Technews articles. Working as an Editor of Govindhtech for one Year and previously working as a Computer Assembling Technician in G Traders from 2018 in India. His Education Qualification MSc.

2 Comments

  1. […] Nvidia released the Kepler-powered GTX 680 three months after AMD released the HD 7970. AMD’s HD 4000- and 5000-series GPUs were great, but the 680 was faster, more efficient, and smaller than the 7970. Nvidia only led these metrics slightly, but it led all three, which was rare or unprecedented. […]

Leave a reply

Please enter your comment!
Please enter your name here

Govindhtech.com Would you like to receive notifications on latest updates? No Yes

Discover more from Govindhtech

Subscribe now to keep reading and get access to the full archive.

Continue reading