Researchers at the Tampa-area Moffitt Cancer Center are utilizing a new AI-powered imaging system developed by nearby tech company Vu Technologies that creates highly detailed, three-dimensional views of tumors in real time, enabling doctors and scientists to examine cancer from multiple angles to give them a clear picture of how tumors grow, spread and interact with surrounding tissue.
The system runs on the NVIDIA DGX Spark, a powerful computing platform built to process enormous amounts of data at high speed. Vu says the technology replaces traditional image-rendering methods with AI-driven visualization tools that can generate images faster.
“We’re seeing a fundamental shift in the market, where customers are demanding faster, more intelligent visual interfaces that are both human-centric and real-time responsive,” Vu CEO Tim Moore said in a statement announcing the collaboration. “We’ve developed our technology on the NVIDIA DGX Spark because it allows our enterprise clients to work faster on the edge and more securely than ever before.”
At the center of the collaboration is a custom-built 3D microscopy visualizer created for Moffitt researchers studying patient-derived microtumors — tiny tumor samples grown from patient cells that allow scientists to test treatments and study how cancer behaves.
The system combines AI-powered visualization, real-time rendering and high-resolution biomedical imaging to help researchers analyze tumors more quickly and with greater detail than traditional imaging methods. That could have major implications for how cancer is treated.
Surgeons, for example, could use the technology to better map out complicated procedures before entering the operating room. Radiation specialists could more accurately target tumors while avoiding healthy tissue nearby. Researchers could also spot patterns or structures inside tumors that might otherwise be missed on standard scans.
And for patients, the technology could make cancer easier to understand. Instead of looking at flat black-and-white scans, doctors may eventually be able to show patients detailed 3D models of their tumors, helping explain where the cancer is located and how treatment will work.
The project also highlights Moffitt’s growing push into artificial intelligence and machine learning, areas many hospitals and research centers now see as the future of medicine.
According to Moffitt, its Machine Learning League was created to expand the use of AI, machine learning and deep learning across cancer research. The effort brings together scientists and clinicians to explore how new AI tools can help solve problems tied to diagnosis, treatment and research while also teaching researchers how to use emerging technologies.
The work with Vu is only the latest AI partnership for the cancer center. In 2024, Moffitt announced a separate collaboration with NVIDIA, Oracle and Deloitte focused on using AI and machine learning to improve cancer care delivery.
That project involved training a large language model using about 100,000 inpatient clinical notes from Moffitt. The goal was to improve medical documentation, streamline billing and coding processes, help match patients to clinical trials and support cancer research.
“This collaboration represents a significant step forward in our mission to provide our patients with the highest quality cancer care,” Dana Rollison, Ph.D., Moffitt’s vice president, chief data officer and associate center director of data science, said in the 2024 announcement. “By leveraging artificial intelligence and machine learning technologies, we can unlock new insights, streamline workflows and ultimately improve patient outcomes.”
The new imaging technology reflects a broader trend happening throughout medicine. AI is increasingly being used not just to sort through data, but also to help doctors actually see disease in more detailed and useful ways.
That matters because cancer is rarely simple. Tumors often twist around blood vessels, push against organs or spread into nearby tissue in ways that can be difficult to fully understand through traditional two-dimensional scans. Three-dimensional imaging gives doctors a more complete picture before surgery or treatment even begins.
“AI-driven visualization platforms are transforming how researchers interact with biomedical data,” said Greg Sawyer, Ph.D., chair of the Bioengineering Department at Moffitt, in the Vu announcement. “Technologies like this have the potential to make complex tumor imaging more interactive, accessible and actionable.”
Sawyer said Moffitt hopes to eventually share the technology more broadly with the research community to help speed up scientific discovery and deepen researchers’ understanding of cancer.
Vu says the tumor visualizer is already being used by researchers. The company also plans to expand similar AI-powered visualization tools on the DGX Spark platform in the future.
