Newswise — CLEVELAND—Case Western Reserve University (CWRU) researchers are using artificial intelligence (AI) and physics at the width of a single hair to peer into the surface of Renaissance paintings, offering new clues about how masterpieces were made—and by whom.

Researchers—from the physics and the art history departments—found a new way to read 400-year-old paintings, using AI to map microscopic surface textures. They uncovered hidden patterns in works by El Greco, revealing clues about authorship, collaboration and the artist’s distinctive hand.

In a new study published in the journal Science Advances, researchers led by Michael Hinczewski, associate professor of physics in the College of Arts and Sciences, scanned the surface of two Renaissance paintings to create ultra-detailed topographic maps, capturing tiny ridges and grooves left by brushstrokes.

The interdisciplinary team of researchers then trained an AI system to analyze the centimeter-scale areas—that they call “patches”—detecting patterns and relationships across the surface that are invisible to the human eye.

By treating each painting like a network of small, interconnected pieces, the algorithm could determine whether the surface patterns pointed to a single artist’s hand or multiple contributors.

The approach revealed a striking unity in one painting, “Christ on the Cross” housed at the  Cleveland Museum of Art (CMA). While examining another work, the “Baptism of Christ,” housed in Toledo, Spain, the team made a significant discovery: The work was long believed to have been finished posthumously by the master’s workshop, but the evidence in this study pointed to an underlying connection between regions of the painting previously attributed to different artists.

The research findings suggest a single set of materials, or even a single hand. If confirmed, that finding could reshape how scholars understand El Greco’s late work, researchers noted.

Hinczewski said the team’s new technique offered a novel, data-driven method to tackle long-standing questions of attribution and artistic practice.

“This is the first time we’ve been able to take surface texture at this scale and use it to say something meaningful about who made a painting,” he said. “When you can analyze details down to the width of a single paintbrush bristle, you start to uncover a kind of fingerprint—one that could eventually help us authenticate works and better understand how artists like El Greco actually painted.”

Connecting the dots

Hinczewski said the project itself began in an unlikely way: a conversation between two Case Western Reserve graduate students who were dating—one studying art history, the other physics. That connection sparked a collaboration that has now spanned seven years, ultimately bringing together scientists, art historians and partners at the CMA, Cleveland Institute of Art and the Factum Foundation in Madrid.

“This is what happens when science meets art,” said Andrew Van Horn, the Ross-Lynn postdoctoral fellow in the Department of Anthropology at Purdue University, who worked on the research as a postdoctoral fellow in both physics and art history at CWRU. “Applying computational methods to actual questions in art history was integral both to creating a new AI method and making a really cool discovery about El Greco’s art. Interdisciplinarity is going to be a key driver of innovation going forward, and the impressive team we put together at CWRU—along with other great schools and institutions—is proof of that.”

Looking ahead, Hinczewski said he envisions applying the technique across larger collections—comparing surface “fingerprints” from different works to more confidently attribute paintings, track how an artist’s style evolved over time and even resolve long-standing debates about disputed pieces. 

As the database grows, the approach could also help flag subtle inconsistencies that point to modern imitations, offering museums and collectors a powerful new tool for detecting counterfeits.

“This is just the beginning,” Hinczewski said. “We’re learning that even a few millimeters of paint can carry a wealth of information about how a work was made. As these tools evolve, they could transform how we study artists over time—and how cultural heritage is protected.”