Researchers use AI to predict dental composite performance

Artificial intelligence (AI) is fast becoming a key driver of innovation in dentistry—and now, researchers at UT Health San Antonio and the University of Texas at San Antonio (UTSA) are training machine learning models to predict how dental composites will perform in real-world clinical settings.

This pioneering project, published in the Journal of Dental Research, aims to optimize the selection and development of composite materials used in dental restorations, including fillings, by leveraging AI-driven predictive modeling.

“Very few studies provide the kind of cross-comparable data that machine learning models need,” explained Dr. Kyumin Whang, Barry K. Norling Endowed Professor at UT Health San Antonio.

Despite thousands of academic studies on dental composites, most focus on newly developed or proprietary materials, often evaluated in narrowly defined lab settings. This limits their usefulness for AI, which requires large, standardized datasets for accurate modeling.

Cross-disciplinary collaboration and comprehensive data gathering

The team, co-led by Dr. Yu Shin Kim, associate professor at UT Health San Antonio School of Dentistry, and Dr. Mario Flores, professor of electrical and computer engineering at UTSA, reviewed over 200 scientific publications to compile performance data on 321 commercially available dental composites.

From these, they refined a dataset of 240 composites, including:

• 28 types of additives affecting properties like strength, polishability, and bonding
• 17 measurable performance outcomes, including shrinkage, fracture resistance, and durability

This extensive dataset laid the foundation for AI models capable of recognizing patterns that indicate superior clinical performance.

Toward precision material recommendations with AI

Initial analysis has demonstrated the potential of AI in isolating which material attributes contribute most to long-term success in dental treatments.

“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,” said Dr. Whang.

This breakthrough could shorten the product development lifecycle, narrowing thousands of possible material combinations to a handful of high-potential options. The result: faster, more efficient design of custom dental composites tailored to clinical needs.

Next step: open-access AI platform for dental material innovation

The researchers envision creating an open-access AI platform where dental manufacturers, clinicians, and research institutions can input formulation data and receive predictive insights into how a material is likely to perform.

Such a tool could:

• Accelerate product development
• Improve clinical outcomes
• Enable data-driven innovation in dental material science

By combining domain expertise from dentistry, biomedical engineering, and AI, this project is setting the stage for a more precise, reliable, and personalized future in restorative dentistry.