Dhananjay Bhaskar can’t hide his enthusiasm on the other end of the video call as he carries his laptop down the hallways of Yale University’s new Wu Tsai Institute building.
He stops in front of a striking piece of art he helped create: a lenticular print that shifts and shimmers as he moves. What makes it remarkable is not just the changing images, but how they were made. They come from brain scans of people imagining Yale’s campus, decoded with the help of artificial intelligence and transformed into moving images. Part artwork, part scientific experiment, the piece offers a glimpse into how thoughts themselves might one day be made visible.
While not tied to a scientific publication, the work is more than an artistic exercise: Bhaskar and his collaborators see it as a glimpse of future applications in neuroscience and medicine, from studying mental imagery in infants to aiding communication in patients with neurological conditions.
“That sort of project really gets me going,” Bhaskar says with a smile.
Bhaskar brings his passion for bridging science and the arts to the University of Wisconsin-Madison in fall 2025, joining the faculty as an assistant professor of biomedical engineering. Hired as part of the university’s RISE-AI Initiative, he uses a combination of topological data analysis, machine learning and mathematical modeling to better understand the “shape” of data and reveal insights that could inform drug discovery and design, cancer biology, neuroscience and more.
“I’m a very visual person,” says Bhaskar, whose wife is a manga artist. “I like to see the shapes in data and paint a picture of what they mean.”
The son of two chemistry professors in Kanpur, India, Bhaskar discovered the field of mathematical biology as an undergraduate student at the University of British Columbia. When he moved to Brown University to pursue his PhD, he looked deeper into cancer biology, using microscope data to build mathematical models and simulations of cancer progression. For his dissertation project, he applied insights from the work of M.C. Escher, a Dutch graphic artist known for infusing mathematical concepts into his art, to understanding how cells transition from healthy states to cancerous.
“Artists often pay attention to negative space, the blank areas that give a picture balance and meaning,” says Bhaskar. “In biology you can ask the same question: Where are cells not located? The key insight in my PhD work was realizing that the empty spaces in tissue matter just as much as the cells themselves. By studying those gaps, through the lens of geometry and topology, I could see how normal tissue patterns broke down as cancer emerged.”
As a postdoctoral researcher at the Yale School of Medicine, he incorporated machine learning, including deep learning, into his arsenal. One recent project involved teaching algorithms to recognize the structure of brain activity in individuals with schizophrenia, part of a broader research program on representation learning for neural data. By teaching machine learning algorithms to fundamentally understand the shape of such data, he says, they don’t require as much raw data—an important consideration for conditions like schizophrenia where patient data isn’t as plentiful.
“With just 20 patients, every data point counts,” he says. “That’s where understanding the shape of the data becomes powerful.”
In line with his research, Bhaskar is interested in creating new courses in machine learning for bioengineers and cell and systems modeling at UW-Madison. And before officially starting, he’d already begun collaborating with BME colleague Aviad Hai around calcium signaling patterns in the brain.
“I want to continue to develop new methods,” says Bhaskar. “And I think with my background, I can speak to biologists, to cancer biologists, I can speak to neuroscientists. We share the same language, I have the same vocabulary, so I find it easy to communicate with them. But I have a technical background that allows me to build methods using things like geometry and topology, which are things that I think I’m uniquely positioned to build on.”
He grins. “I love building tools that let us see what we couldn’t before—the hidden structures that make biology work.”