Seeing the unseen: How AI and medical imaging are changing health care

As a third grader with a love for technology, Bennett Landman decided he wanted to upgrade the operating system on his father's old MS-DOS computer. In the process, he accidentally wiped the hard drive. Instead of scolding him, his father, a physician with a passion for research, simply said, "Great, now let's figure out how to fix it." That moment-equal parts curiosity and problem-solving-set the tone for Landman's career.

Professor Bennett Landman

Now a professor at Vanderbilt University and director of the Vanderbilt Lab for Immersive AI Translation, Landman, who holds a Stevenson Chair, is still doing what he loves: figuring things out. But instead of tinkering with MS-DOS, he's revolutionizing the way artificial intelligence interacts with medical imaging. His work is helping to refine how we diagnose disease, understand the human body and even reshape what's possible in health care.

Cracking the code of medical images
By Landman's count, he has more than 2,000 collaborations globally that look at all sorts of different diseases, from the head down to the shins. But the heart of his work remains medicine and the promise of a world where AI-driven imaging leads to earlier diagnoses and better patient outcomes.

Medical imaging-MRI scans, CT scans, ultrasounds-is one of the most powerful tools in modern medicine. But Landman sees a fundamental problem: "When you go in for an MRI, that scan is not a test on its own. It's only useful because a radiologist interprets it," he explains. "We don't have a standard, quantitative way of saying 'yes, this is disease' the same way we do with a blood test."

Landman's research aims to bridge that gap. By training AI models to "read" medical images with the same (or better) accuracy as human experts, he's working toward a future where AI can provide a second opinion, reduce human error and even uncover hidden patterns in disease progression. "We want to take imaging beyond the one-time diagnosis," he says. "Every scan should contribute to a broader, more complete understanding of a patient's health over time."

"Right now, we're wasting so much information," he says. "You get a scan, it's read for one thing, and then it's done. But what if that scan could be used to track everything about your health? Your lung function, heart function, muscle quality-things you might not even know to look for yet."

In talking about what's possible, Landman compares it to the evolution of medical records from stacks of physical paperwork, to digital records, to now as those disparate pieces of digital data are combined in new, even unimaginable, ways.

By creating models that analyze imaging data alongside other medical records, Landman and his team are working toward a future where every piece of medical data contributes to a deeper understanding of a patient's health. "If we can start making sense of all this information, we can make medicine proactive instead of reactive," he says. "It's not just about catching disease-it's about optimizing health."

The research world is taking note. Recently, Landman was elected as a fellow of the Institute of Electrical and Electronics Engineers, the world's largest and most prestigious society for electrical and computer engineering. With more than 460,000 members in more than 190 countries, IEEE is the leading authority on a wide variety of areas in electrical and computer sciences, engineering and related disciplines.

A culture of curiosity
Landman credits Vanderbilt's culture of collaboration for making this work possible. "The hardest thing about being here is that it's really difficult to get someone to tell you an idea is bad," he says with a laugh. "From the moment I arrived, the answer was always, 'Yes, let's try it.' That spirit of curiosity fuels everything we do."

Bennett A. Landman, Ph.D. is Stevenson Chair of Engineering, Professor of Electrical and Computer Engineering at Vanderbilt University. He is seen here working with students at Featheringill Hall.
Photos by Joe Howell

It's that same spirit that has kept Landman in academia instead of industry, despite offers. "I love working with students," he says. "Every day, someone walks in and says, 'I wonder if we could…' and I get to say, 'Let's find out.' That's why I do this."

"Bennett is one of those super-energy researchers who make you scratch your head and wonder, 'how does he do it all?'" said Krish Roy, the Bruce and Bridgitt Evans Dean of the School of Engineering and University Distinguished Professor. "With a total of seven academic appointments at Vanderbilt, ranging from computer engineering and computer science to neurology and radiology, Bennett represents the very definition of an interdisciplinary scholar and radical collaboration. In addition to his wide-ranging research brilliance, he has an enviable track record of leadership with real-world impact and achievement."

Whether it's training AI to diagnose disease, helping students crash and rebuild their own systems, or applying medical imaging techniques to ancient history, one thing is clear: Bennett Landman still resonates with that third grader who just wants to figure things out. And in doing so, he's helping to reshape the future of medicine-and beyond.

"We've always opened new ideas, and we've created new discoveries," Landman said. "Each thing we find just makes me more curious."

Grand Opening of Vanderbilt Lab for Immersive Artificial-Intelligence Translation (VALIANT)
Photo: Harrison McClary/Vanderbilt University