By Danielle 
Artificial intelligence is already reshaping diagnostics in dentistry, but researchers at UT Health San Antonio and the University of Texas at San Antonio (UTSA) are now exploring how AI could help evaluate and optimize dental composite materials.
Their goal: to develop machine learning models that can accurately predict how commercially available dental composites—used in fillings and other restorations—will perform in clinical settings.
“Very few studies provide the kind of cross-comparable data that machine learning models need,” said Kyumin Whang, Barry K. Norling Endowed Professor in Comprehensive Dentistry at UT Health San Antonio. “Even though there are thousands of papers on dental composites, the vast majority focus on new or proprietary materials tested under specific lab conditions.”
Whang and co-lead investigator Yu Shin Kim, associate professor at the UT Health San Antonio School of Dentistry, collaborated with Mario Flores, professor in electrical and computer engineering and biomedical engineering at UTSA, to build a dataset of 240 commercially available dental composites. Their work, published in the Journal of Dental Research, represents a rare cross-disciplinary effort to apply artificial intelligence to restorative dental materials.
Read related article: Reshaping Dental Care with Artificial Intelligence
Read related story: To ‘revolutionize dental care’: U.S. institute gets $6.2M to develop AI-based dental filling to replace amalgam
Listen: Episode 2: Integrating Technology and Training Staff
Team filtered and standardized data
To build a usable dataset, the researchers reviewed more than 200 scientific studies and compiled data on 321 commercially available dental composites. These materials featured 28 types of composite additives—ingredients that influence factors like strength, polishability and bonding—and 17 distinct performance outcomes, including traits such as shrinkage, fracture resistance and overall durability.
Their initial analysis showed that AI could help identify the most important material properties that lead to clinical success. With more comprehensive and consistent data, they say AI models could one day recommend optimal formulations from thousands of potential combinations—dramatically accelerating the design and testing process.
“Once we make these models more accurate, we’ll be able to dial in the desired properties, and the AI model would recommend a formulation match,” Whang said. “This will narrow the field from thousands of possible combinations to a targeted few, dramatically reducing the time from concept to clinical use.”
As a next step, the researchers hope to create an open-access platform where companies and research institutions can input formulation data and receive predictive performance insights—paving the way for faster development of customized dental composites.