NVIDIA Launches NIM Microservices to Empower Generative AI Applications in Healthcare

San Jose, California, USA—GTC—March 18, 2024, Pacific Time——NVIDIA today launched more than 20 new microservices, enabling healthcare enterprises worldwide to fully leverage the latest advancements in generative AI, regardless of location or cloud platform.

The all-new NVIDIA Healthcare Microservices Suite includes optimized NVIDIA NIM™ AI models and workflows, and provides industry-standard application programming interfaces (APIs) for creating and deploying cloud-native applications. They offer advanced capabilities in medical imaging, natural language and speech recognition, as well as digital biology generation, prediction, and simulation.

Additionally, NVIDIA-accelerated software development kits and tools, including Parabricks^®, MONAI, NeMo™, Riva, and Metropolis are now accessible via NVIDIA CUDA-X™ microservices to accelerate healthcare workflows such as drug discovery, medical imaging, and genomics analysis.

Of these microservices, 25 were launched today to accelerate the digital transformation of healthcare enterprises, as generative AI presents numerous opportunities for pharmaceutical companies, physicians, and hospitals. These include screening trillions of drug compounds to advance medical science, collecting more comprehensive patient data to improve early disease detection, and enabling smarter digital assistants.

Researchers, developers, and healthcare professionals can leverage these microservices to seamlessly integrate AI into both new and existing applications, deploying them across environments ranging from the cloud to on-premises infrastructure, thereby enhancing their life-saving work.

Kimberly Powell, Vice President of Healthcare at NVIDIA, stated, “For the first time in history, we are representing the worlds of biology and chemistry within computers, making computer-aided drug discovery possible. With our support, healthcare enterprises can easily build and manage AI solutions to fully leverage the capabilities and potential of generative AI.”

Within the All-New Medical Microservices SuiteNVIDIA NIMOptimized inference can be provided for the growing number of models in fields such as medical imaging, health technology, drug discovery, and digital health. These models can be used for generative biology and chemistry, as well as molecular prediction tasks. NIM microservices are now available viaNVIDIA AI Enterprise 5.0Software platform provided.

This suite of microservices also includes a range of models for drug discovery, such as the generative chemistry model MolMIM, the protein structure prediction model ESMFold, and DiffDock, which helps researchers understand how drug molecules interact with targets. The VISTA 3D microservice accelerates the creation of 3D segmentation models. Compared to standard DeepVariant running on CPUs, the Universal DeepVariant microservice increases variant identification speed in genomic analysis workflows by more than 50-fold.

Leading computational software company Cadence is integrating NVIDIA BioNeMo microservices, used for AI-guided molecular discovery and lead compound optimization, into its Orion platform for accelerating drug development.^®in the molecular design platform.

With Orion, researchers at pharmaceutical companies can generate and search databases containing hundreds of billions of compounds, as well as build relevant models. BioNeMo microservices, such as the MolMIM generative chemistry model and the AlphaFold-2 protein folding model, significantly enhance Orion’s design capabilities.

Anthony Nicholls, Vice President at Cadence, stated, “Our pharmaceutical and biotechnology customers require access to accelerated resources for molecular simulation. By leveraging BioNeMo microservices, researchers can generate molecules optimized to meet scientists’ specific needs.”

Nearly 50 application providers and multiple biotechnology and pharmaceutical companies are using medical microservices on the platform, including Amgen, Astellas, DNA Nexus, Iambic Therapeutics, Recursion, and Terray, as well asV7medical imaging software manufacturers.

David M. Reese, Executive Vice President and Chief Technology Officer at Amgen, stated, “Generative AI is transforming drug discovery, enabling us to build advanced models and seamlessly integrate AI into the antibody design process. Our team is leveraging this technology to develop next-generation therapeutics that will deliver maximum value to patients.”

Generative AI Is Transforming the Future of Patient Care. Hippocratic AI is developing task-specific generative AI healthcare agents, powered by the company’s safety-focused medical large language models (LLMs), integrated with NVIDIA Avatar Cloud Engine (ACE) microservices, and leveraging NVIDIA NIM for low-latency inference and speech recognition.

These AI healthcare agents converse with patients to handle tasks such as appointment scheduling, preoperative reminders, and post-discharge follow-ups.

Munjal Shah, Co-founder and CEO of Hippocratic AI, stated, “With generative AI, we can address some of the healthcare industry’s most pressing needs, helping to alleviate widespread staffing shortages, deliver more high-quality care, and improve patient outcomes. NVIDIA’s technology stack is critical to achieving the speed and fluency in conversations that are essential for building natural emotional connections between patients and Hippocratic’s generative AI healthcare agents.”

Abridge is building an AI-driven clinical conversation platform. The platform generates draft clinical notes, saving clinicians up to three hours per day. Transforming raw audio from noisy environments into transcripts requires seamless collaboration among multiple AI technologies: speech recognition, transcription, correction, and speaker diarization must all be completed within seconds. Conversations must be structured based on the types of medical information contained in each sentence, and robust language models must be applied to synthesize relevant evidence into summaries. The system converts clinical conversations into high-quality post-visit documentation in real time.

Models created by Flywheel can be converted into microservices. The company’s cloud-based centralized platform supports biopharmaceutical companies, life sciences organizations, healthcare providers, and specialized medical centers in identifying, curating, and training medical imaging data, thereby accelerating the time to insights.

Trent Norris, Chief Product Officer at Flywheel, stated, “In an era of rapid advancement in medical technology, the integration of NVIDIA’s generative AI microservices with the Flywheel platform represents a transformative leap. Leveraging these advanced tools, we have not only enhanced our capabilities in medical imaging and data management but also significantly accelerated medical research and patient care outcomes. Powered by cutting-edge AI solutions from NVIDIA, Flywheel’s AI Factory meets the needs of healthcare customers anytime, anywhere, helping to drive progress in digital health and biopharmaceuticals.”

Developers can accessai.nvidia.comTry NVIDIA AI microservices, and byNVIDIA Certified SystemsDeploy production-grade NIM microservices on NVIDIA AI Enterprise 5.0, with providers including Dell Technologies and Hewlett Packard Enterprise,Associationand Supermicro, as well as leading public cloud platforms such as AWS, Google Cloud, Microsoft Azure, and Oracle Cloud Infrastructure. In addition, it can also beNVIDIA DGX CloudTrial use.

VCBeat believes that NVIDIA has been steadily advancing the application of AI in the healthcare industry. From initially providing hardware computing power, to launching a series of medical AI service components such as NVIDIA Clara, and now to NVIDIA NIM microservices, these iterative advancements will enhance the implementation of AI in the healthcare sector and further solidify NVIDIA’s position within the AI healthcare ecosystem.