NVIDIA Healthcare

At #BIO2026, Kimberly Powell sat down with the founders of Benchling, Chai Discovery, and Edison Scientific to discuss what #AgenticAI for life sciences looks like in practice.   Learn what’s possible in the space via our latest newsletter ⬇️

What does it actually take to turn an AI agent into a life sciences agent? NVIDIA BioNeMo Agent Toolkit combines open models, recipes, accelerated libraries, and NIM microservices so agents can run complex scientific workflows without managing the infrastructure themselves.

🚨 NVIDIA Medical Physics Simulation 🚨 A new open source framework within Isaac for healthcare, helps medical robotics developers model anatomy-device interaction, generate hard-to-capture scenarios, test in silico, and train or evaluate robot policies before hardware-heavy testing. By unifying anatomy-device physics, sensor simulation, and robot learning in reusable environments, it turns ad hoc simulations into scalable infrastructure for the next generation of surgical and interventional robots. Developers have full access to the framework, models and weights to reproduce results, evaluate performance across anatomies, and build regulatory evidence on a GPU-accelerated foundation. 📰 https://nvda.ws/3RcbeoV

To identify the most promising drug candidates, we need to understand not only whether a molecule is likely to bind to a target, but how strongly. Today, we’re excited to introduce Nesso-1, an open-source, coarse-grained co-folding model that significantly accelerates binding-affinity predictions. Compared with leading open-source models, Nesso-1 offers: ✅ Higher accuracy across public benchmarks and internal biochemical assays, including challenging, heavily out-of-distribution, medicinal chemistry datasets ✅ More than 10× speedup, enabling state-of-the-art virtual screening of even larger chemical spaces By leveraging NVIDIA cuEquivariance, we’ve been able to improve training and inference efficiency by a further factor of 2-3x. We look forward to continuing to improve Nesso-1 in collaboration with NVIDIA. By open-sourcing the model, we hope to accelerate innovation across the drug discovery community and, ultimately, help bring new therapies to patients faster. While zero-shot generalization remains challenging, Nesso-1’s speed makes continued iteration, evaluation, and improvement significantly faster and more cost-efficient. The technical report and GitHub repository are linked in the comments.

https://lnkd.in/gKynEcCn A new era for surgery with real-time AI. We're proud to unveil Touch Surgery™ Aide — our next-generation advanced compute platform for the operating room, built in collaboration with NVIDIA — at #SRS2026. For years, the Touch Surgery™ ecosystem has brought insight to surgical teams in more than 1,500 operating rooms worldwide, connecting pre-op planning and training, intra-op tele-mentoring, and post-op analysis. Touch Surgery™ Aide is the next step: real-time AI*, running during the procedure itself. Powered by the NVIDIA Holoscan SDK alongside CUDA and TensorRT, the platform is capable of running multiple AI applications simultaneously in a live case, processing surgical data and surfacing insight the moment it matters. Read the full announcement: https://lnkd.in/gKynEcCn #Medtronic #TouchSurgery #DigitalSurgery #AI #SurgicalInnovation -- The Medtronic Touch Surgery™ ecosystem is not intended to direct surgery, or aid in diagnosis or treatment of a disease or condition.

🎉TWO HUNDRED FIFTY THOUSAND DOWNLOADS 🎉 This is a meaningful milestone for Open-H-Embodiment and a testament to what is possible when the robotics and physical AI community comes together. TLDR: It’s the first large‑scale, multi‑institution, multi‑robot open dataset for medical robot learning - 780 hours of synchronized surgical video and robot motion data to power physical AI for healthcare robotics. Developed with steering committee members from The Johns Hopkins University and the Technical University of Munich, and shaped by contributions from more than 50 organizations worldwide, Open-H-Embodiment is helping researchers and engineers push the boundaries of embodied intelligence with real-world robotic data. Every download represents new experiments, new ideas, and progress toward more capable real‑world robotic systems. Thank you for your feedback, contributions, and support. 🔍 Explore the dataset https://lnkd.in/g7Q4B7GD 📈 See the results https://lnkd.in/gszyz2vx 📖 Read the story https://lnkd.in/dY7vimti

🚀 Bristol Myers Squibb is fundamentally changing how we approach drug discovery. They’ve just announced the deployment of their second AI factory - the most powerful in the life sciences industry, built on NVIDIA Vera Rubin architecture. By integrating eight NVIDIA DGX Vera Rubin NVL72 systems, BMS is achieving up to 10x the performance per megawatt compared to previous systems. A game-changer for healthcare: 🧬 Democratized Compute: Opening unified supercomputing access to BMS scientists globally 🤖 Agentic AI: Utilizing the NVIDIA BioNeMo Agent Toolkit for complex biological predictions and model training ⚡ Streamlined Discovery: Eliminating traditional research bottlenecks to drastically accelerate the journey from lab to market We are moving past isolated AI projects and entering an era where AI factories are the backbone of biomedical research. 📃https://nvda.ws/4vAoMsh

Japan is laying the foundation for the future of physical AI. Partnering with Noetra and supported by METI, the NVIDIA Vera Rubin AI factory built on the NVIDIA DSX platform will bring world-class AI infrastructure to manufacturing, robotics, healthcare, and beyond. ✅ 27,500 NVIDIA Rubin GPUs ✅ 13,750 NVIDIA Vera CPUs ✅ Connected and scaled with NVIDIA Spectrum-X Ethernet networking Read the announcement ➡️ https://nvda.ws/44BXWVI