NVIDIA integrates its AI microservices with AWS

As part of NVIDIA’s AI Enterprise software available on AWS Marketplace, developers will gain access to an expanding AI model library through standard APIs.

This integration with Amazon SageMaker and AWS ParallelCluster is designed to streamline deployment and management of machine learning models and high-performance computing clusters on AWS.

Additionally, NIMs can be orchestrated using AWS HealthOmics, tailored for biological data analysis.

In a blog post announcing the integration, NVIDIA noted NIM incorporates genomics models derived from Parabricks, a software platform developed by NVIDIA that specializes in accelerating and optimizing genomic data analysis.

These workflows are also accessible on AWS HealthOmics as Ready2Run workflows, enabling customers to deploy pre-built pipelines.

NIM microservices provide optimized AI models for decoding proteins and genomic sequences, as well as for conversational AI and visual GenAI for avatars and digital humans.

Trained on organization-specific data utilizing techniques like Retrieval-Augmented Generation, these assistants could tap into internal data sources to synthesize research.

The company said these applications could enhance patient support and clinician assistance, by leveraging organization-specific data to synthesize research and improve productivity.

WHY THIS MATTERS
Streamlined development of GenAI tools could have profound implications across the healthcare and life sciences industries.

Multiple startups are already leveraging the potential to accelerate drug discovery workflows, train generative models for protein design, and help researchers with cloud-based data analysis.

Despite persistent concerns over the accuracy of AI and the broader implications of integrating a relatively new technology into all aspects of healthcare, an April survey by Wolters Kluwer Health found nearly 70% of physicians enthusiastically embrace the benefits of GenAI.

THE LARGER TREND
In March, NVIDIA Healthcare introduced 25 new cloud-agnostic microservices, enabling healthcare developers to integrate GenAI into their applications across various use cases and digital health initiatives, whether deployed in the cloud or on-premises.

A partnership with Johnson & Johnson MedTech will see integration of NVIDIA’s AI technology in surgical procedures to enhance real-time analysis and broaden the application of AI algorithms in surgical decision-making, education, and collaboration within operating rooms.

Roughly a year ago, the company integrated its AI Enterprise platform with Microsoft Azure machine learning, offering healthcare customers expanded support for building, deploying, and managing customized AI applications using more than 100 NVIDIA AI frameworks and tools.

ON THE RECORD
“Easy access to NIM will enable the thousands of healthcare and life sciences companies already using AWS to deploy generative AI more quickly, without the complexities of model development and packaging for production,” said NVIDIA in the blog post. “It’ll also help developers build workflows that combine AI models across different modalities, such as amino acid sequences, MRI images and plain-text patient health records.”