NVIDIA expands healthcare AI and protein design tools

NVIDIA is partnering with healthcare and research organizations to expand its artificial intelligence capabilities in protein design and cancer treatment. The announcements focus on open models and advanced tools to support personalized medicine.

Open Protein Design Model

The company introduced RFD4-Proteina, a model within its BioNeMo platform. It was developed with the Institute for Protein Design at the University of Washington School of Medicine. The tool generates targeted protein designs, allowing scientists to search and refine molecules based on specific inputs and constraints.

According to NVIDIA, the model achieved a higher computational design success rate than evaluated baselines across protein binding, small-molecule binding, enzyme scaffolding and DNA binding. It is now available through the company’s partners, including Amazon Bio Discovery, Anthropic, Benchling, Databricks, Google Cloud, Hugging Face, Microsoft Azure, Nebius, OpenAI and Tamarind Bio.

Siemens Healthineers and Adaptive Radiotherapy

NVIDIA also announced a collaboration with Siemens Healthineers. The partners aim to develop adaptive radiotherapy using agentic and robotic intelligence. The goal is to simplify workflows and expand access to cancer care.

The system will use AI-assisted treatment planning, patient-specific simulation and real-time computing. The companies will combine this with NVIDIA’s Holoscan platform to connect live sensor data with AI and simulation. They will also explore integrating NVIDIA’s Nemotron open foundation models to help oncologists evaluate and enhance personalized treatment.

Arthur Kaindl, head of Varian, the Siemens Healthineers cancer-care business, emphasized the goal of adapting treatment to the patient. “Adaptive radiotherapy is built on a single principle: treatment should adapt to the patient, not the other way around,” Kaindl said in a statement. “We’ve made important advances in precision and adaptation – most recently, with the launch of the Accela system. Our collaboration with NVIDIA builds on that progress, exploring how AI and real-time computing could give clinicians greater speed and insight to make informed decisions, adapt treatment with greater confidence and deliver personalized care to more patients.”

Industry Expansion

Recent partnerships with major pharmaceutical companies illustrate the broader trend. In January, Eli Lilly and Co. and NVIDIA announced a partnership to develop a San Francisco-based AI co-innovation lab. The lab will house biologists, scientists and experts in medicine along with AI model builders and engineers from NVIDIA. The investment includes up to $1 billion in infrastructure, workforce and compute power over five years.

Roche announced plans in March to extend its partnership with NVIDIA to deploy a hybrid-cloud AI factory. The company will use the chip maker’s computing and AI to enhance drug discovery, clinical trial efficiency and unlock data insights. In July, NVIDIA and Bristol Myers Squibb announced plans to build what they call the most powerful AI factory in the life sciences industry. Bristol Myers Squibb will deploy NVIDIA DGX SuperPOD with DGX Vera Rubin NVL72 systems.

More stories like this: Q&A: NVIDIA on the future of healthcare AI, Part 2 Abridge partners with NVIDIA, expands beyond clinical documentation Intel spin-out RealSense secures $50M funding, partners with NVIDIA Jessica Hagen is the executive editor of MobiHealthNews. Email: [email protected] MobiHealthNews is a HIMSS Media publication. Follow Jessica’s digital health coverage on LinkedIn: Jessica Hagen. HIMSS is hosting the one-day AI Executive Leadership Summit in San Diego on Oct. 21, 2026, followed by its AI in Healthcare Forum Oct. 22–23. Register separately for the two events here and here. WATCH NOW: Is Medicare Advantage bonus spending achieving the desired results?