TLDR: Mayo Clinic has deployed NVIDIA’s advanced Blackwell infrastructure, including DGX SuperPOD with DGX B200 systems, to accelerate generative AI solutions in medicine. This initiative aims to transform healthcare through faster disease diagnosis, drug discovery, and precision medicine, leveraging AI to analyze vast amounts of medical data, particularly in digital pathology.
Mayo Clinic has taken a significant leap forward in integrating artificial intelligence into clinical practice with the deployment of NVIDIA’s cutting-edge Blackwell infrastructure. This includes the NVIDIA DGX SuperPOD, equipped with NVIDIA DGX B200 systems, which provides state-of-the-art AI compute capabilities. This collaboration between Mayo Clinic and NVIDIA is designed to foster rapid innovation and the development of foundation models, aligning with Mayo Clinic’s ‘Bold. Forward.’ strategy and its broader platform approach to healthcare.
The advanced computing infrastructure will initially support the development of foundation models in critical areas such as pathomics, drug discovery, and precision medicine. The NVIDIA Blackwell-powered DGX SuperPOD is specifically engineered to efficiently process large, high-resolution imaging data, which is crucial for training AI foundation models. This technology is expected to dramatically accelerate pathology slide analysis and foundation model development, potentially reducing a four-week workload to just one week, thereby improving patient outcomes.
Dr. Matthew Callstrom, M.D., Ph.D., medical director of the Department of Strategy and leader of Mayo Clinic’s Generative Artificial Intelligence Program, emphasized the transformative potential. “Our aspiration for AI is to meaningfully improve patient outcomes by detecting disease early enough to intervene. What was once a hypothetical — ‘If only we had the right data’ — is now becoming reality thanks to AI and advanced computing,” he stated.
This deployment marks the first time NVIDIA’s latest AI chipmaker technology will be used on such a large scale in a healthcare setting. Jim Rogers, CEO of Mayo Clinic Digital Pathology, highlighted the profound impact, stating, “The confluence of the [AI computing power] that we’re talking about, the data, and the clinical knowledge coming together could be kind of a once-in-a-generation opportunity to truly transform how medicine is practiced.”
The system, located in Brooklyn Park, utilizes a large computing cluster with 128 high-powered Blackwell graphics processing units (GPUs). This infrastructure is described as “rocket fuel” for accelerating innovation, significantly enhancing Mayo’s previous technology by three to four times. It will enable Mayo Clinic to fully digitize its extensive pathology archives, aiming to uncover new insights from aggregated patient data.
One notable achievement already in progress is the development of Atlas, a leading pathology foundation model created by Mayo Clinic in partnership with Aignostics. Atlas was trained on over 1.2 million histopathology whole-slide images, empowering clinicians and researchers with improved diagnostic capabilities.
Also Read:
- GenSeg: Revolutionizing Medical Image Segmentation with Generative AI in Data-Scarce Environments
- OpenAI and Penda Health Collaborate on AI Clinical Copilot for Healthcare Advancement
This investment is part of Mayo Clinic’s broader AI strategy, which also includes partnerships with technology giants like Google and Microsoft, as well as genomics-focused AI company Cerebras. The overarching goal is to alleviate administrative burdens and facilitate groundbreaking medical discoveries, ultimately enhancing patient care worldwide.
