Nvidia sets the pace for medical digital twins

Mobility and modern telecommunication concept: macro view of tablet computer and touchscreen smartphones with colorful interfaces on laptop notebook PC

Missed a session at the Data Summit? View on demand here.

Follow along with VentureBeat’s coverage of Nvidia’s GTC 2022 event.

Digital twins play a vital role in Nvidia platforms for the development of robots and self-driving cars. But medical digital twins face several significant regulatory, technology and privacy hurdles. At this year’s GTC, Nvidia will showcase some key developments that could drive the adoption of digital twins in medicine.

The hardware that supports AI has become the stage where the AI ​​industry needs to be constantly improved, updated, tested and validated. “We won’t be able to scale that without digital twins,” Nvidia’s VP of healthcare, Kimberly Powell, said at a news conference.

Major healthcare advancements introduced at GTC include synthetic data generation, the commercial release of the Clara Medical AI platform, improved DNA sequencing workflow, enhanced pharmacovigilance capabilities, and improved discovery tools. medicines. Enhanced digital twin capabilities will ultimately take advantage of these advances to dramatically improve patient safety and support new business models in healthcare.

Improving robotic data platforms

Nvidia is an innovative leader when it comes to new AI platforms for autonomous driving (Drive), robot design (Isaac) and healthcare (Clara). These platforms simplify the development, implementation and continuous improvement of AI for industry domains. Each platform contains all the capabilities needed to bring data from various sources into a development environment, train new algorithms at scale, and then deploy them on new inference hardware.

Healthcare data accounts for 30% of global data requirements and is growing at a 36% CAGR. Nvidia Clara streamlines AI workflows with AI training; more than 40 pre-trained models, applications and platforms in data centers; on standalone servers; or integrated into medical devices. However, Nvidia has not yet announced a specific medical digital twin capability.

In contrast, both Drive and Isaac include digital twin capabilities that simplify product design workflows, AI training workflows, and performance testing of new combinations of hardware and software. For example, Drive supports complete simulation environments that can synthetically generate variations that reflect the impact of rain, snow, and darkness. These simulations can help train and develop models and predict when a model may not perform well enough. Likewise, Isaac helps build and test new robotic hardware and algorithms in these simulated environments.

Digital twin capabilities are more challenging in medicine due to privacy safeguards, medical regulations and security considerations. While companies today address these issues in one-time deployments, they are difficult to scale. The combination of Nvidia’s existing toolchain and the recent announcement could help address these challenges.

Healthcare Announcements at GTC

Ultrarapid nanopore analysis pipeline (UNAP) is a new DNA sequencing platform running on a single DGX A100 to reduce the computational cost of sequencing an entire genome from $568 to $183. It recently helped set the world record for sequencing an entire genome in four hours and ten minutes. Support for four startups developing AI transformers for decision making, therapies and drug discovery using the UK’s fastest supercomputers. Transformers help teams analyze unlabeled data, which represents the bulk of medical data. Early access to Megamolbart, a new training framework and generative model for chemistry developed in collaboration with Astrazeneca. a natural language processing (NLP) model that reads the text format of chemical compounds and uses AI to generate new molecules. The transformer chemistry model can train chemical language models with more than 1 billion parameters using the Nvidia Nemo Megatron framework. A new domain-specific NLP model from Janssen built on Biomegatron improves detection of adverse pharmaceutical events by 12%, for an overall detection rate of 88% .SynGatorTron, the world’s largest tool for generating clinical language models, in collaboration with the University of Florida. It automates the creation of synthetic data from healthcare data to improve AI models while protecting privacy. It could be used to create digital twins for patient records as a control group in clinical trials. An additional GatorTron model can also improve medical chatbots, biomedical research, clinical trial matching, and medical event detection. Nvidia Clara Holoscan MGC, a reference design for a medical-grade real-time AI computing platform, slated for early access in Q1 2023. It promises to industrialize AI development and comes with a 10-year software stack support agreement. Early hardware partners include ADLINK, Advantech, Dedicated Computing, Kontron, Leadtek, MBX Systems, Onyx Healthcare, Portwell, Prodrive Technologies, RYOYO Electro and Yuan High-Tech.

Enabling new business models

Powell expects that current breakthroughs in synthetic data generation and improved modeling tools will also help bring digital twins into healthcare. For example, previous work with the Cambridge One supercomputer helped King’s College London generate synthetic brains. This allowed them to test new algorithms and represent populations in different ways.

Healthcare could also adopt business models similar to those that innovators like Tesla have brought to the auto industry with SaaS autonomous driving offerings. In the traditional model, car companies would sell a car and make all their money from the direct sale of the vehicle. Tesla came along with a new programmable AI computing platform that is constantly updating new features and improving over time.

Likewise, healthcare companies could develop innovative new enhancements to existing tools such as MRI scanners or endoscopy robots that improve medical procedures and clinical workflows. Powell said, “We need these capabilities to be able to marry the new devices. The extra cost to build these is minimal compared to the cost of the services you can build on top of them.”

Powell predicts that current work on weaving digital twins into AI healthcare workflows is just the start of a journey of at least a decade. And furthermore, it will become a crucial part of AI and product development in healthcare.

“It will be synonymous with how AI is so ubiquitous in everything we do today. Digital twins will be ubiquitous in everything we do in every industry for the next decade,” said Powell.

VentureBeat’s mission is to be a digital city square for tech decision makers to learn about transformative business technology and transactions. Learn more

This post Nvidia sets the pace for medical digital twins

was original published at “https://venturebeat.com/2022/03/22/nvidia-sets-the-stage-for-medical-digital-twins/”