SAN ANTONIO, JULY 22, 2025 — Scientists from The University of Texas at San Antonio (UTSA) and UT Health Science Center at San Antonio are leveraging artificial intelligence (AI) to revolutionize how dental care is delivered. Their goal: to streamline the development of dental materials and better tailor treatments to individual patients.
The research team’s findings, recently published in the Journal of Dental Research, demonstrate a significant step forward in dental science. By using AI to predict how dental composite materials will perform, the researchers aim to improve long-term outcomes while cutting down on the traditional trial-and-error approach.
A Smarter Way to Choose Dental Materials
Dental composites—used in procedures like fillings and sealants—must be strong, durable, and reliable. Yet choosing the right formulation can be challenging due to the overwhelming number of available materials and the lack of standardized testing.
“There are thousands of studies on dental composites, but most focus on specific materials tested under unique lab conditions,” said Dr. Kyumin Whang, Barry K. Norling Endowed Professor in Comprehensive Dentistry at UT Health San Antonio. “Machine learning requires cross-comparable data, and very few studies offer that.”
To address this gap, researchers compiled a dataset of 240 commercially available dental composites drawn from scientific literature. The collaborative team included Whang and Dr. Yu Shin Kim from UT Health San Antonio, and Dr. Mario Flores from UTSA’s Departments of Electrical and Computer Engineering and Biomedical Engineering.
Training AI to Predict Performance
With the data in hand, the team trained machine learning models to predict key properties of dental composites—such as strength, shrinkage, and viscosity—that are critical for success in clinical settings. These models were designed to estimate how well a material would perform under real-life conditions, including its ability to resist fracture and maintain integrity over time.
“We took a rigorous, data-driven approach,” said Flores. “By identifying correlations between properties, we could create models that learn patterns and predict outcomes.”
Although the size of the dataset limited the full potential of the predictions, the study confirmed that AI has real promise—especially if more standardized data becomes available in the future.
A Vision for the Future of Dental Innovation
Looking ahead, the researchers hope to develop an open-access platform that allows scientists and manufacturers to input material formulations and receive AI-generated performance predictions. This could drastically cut the time it takes to move a new dental material from the lab to the clinic.
“Once we fine-tune these models, we’ll be able to input the desired characteristics, and the AI will suggest the best formulation,” said Whang. “This could narrow down thousands of combinations to just a few, making the development process faster and more focused.”
The UTSA–UT Health San Antonio partnership not only highlights how AI can transform dental research but also serves as a model for interdisciplinary collaboration in health care innovation.
