Dental News - Researchers use generative artificial intelligence to design realistic dental crowns

Search Dental Tribune

Researchers use generative artificial intelligence to design realistic dental crowns

A recent study has demonstrated that using a newly developed algorithm for generating new shapes in 3D space based on training sets could replace existing conventional methods of designing dental crowns. (Image: Andrey Zhernovoy/Shutterstock)

HONG KONG: CAD/CAM technology has greatly improved the productivity of dental prostheses but still has its drawbacks when it comes to the design of crowns. Using an artificial intelligence (AI) learning method, researchers from the Faculty of Dentistry at the University of Hong Kong have created an algorithm for personalised dental crown design of high accuracy to produce crowns that resemble the morphology and biomechanics of natural teeth.

According to the researchers, the CAD/CAM workflow has significantly improved dentistry, but is still labour-intensive and time-consuming, particularly because of the need for customisation for each patient, despite CAD software using a tooth library to assist in generating prosthetic designs, and it generates health and environmental hazards during the 3D-printing and milling processes. CAD/CAM remakes are often necessary because of marginal misfits, and design can affect the biomechanical performance and thus the fatigue lifetime of the crown. For this reason, a means of crown design that addresses these issues is needed.

Lead author Dr Hao Ding. (Image: University of Hong Kong)

In the study, the researchers trained the algorithm on 600 sets of digital casts of mandibular second premolars and their adjacent and antagonist teeth and tested it on an additional 12 sets of data to generate 12 crowns. They then compared the natural second premolars with the designs created by their algorithm, using CEREC software and by a technician using a CAD program. They looked at morphological parameters of 3D similarity, cusp angle, and number and area of occlusal contact points. They also subjected the designs realised on computer in lithium disilicate to biomechanical fatigue simulations based on physiological occlusal force.

During the training process, natural teeth morphological features were learned by the algorithm, so that it can design dental crowns comparable to a natural tooth—both morphologically and functionally,” lead author Dr Hao Ding, a postdoctoral fellow in applied oral sciences and community dental care at the university, said in a press release.

The generative AI-designed crowns had the lowest 3D discrepancy, closest cusp angle and similar occlusal contacts compared with natural teeth. In lithium silicate, the AI-designed crowns were found to have an expected lifespan similar to that of natural teeth. According to the researchers, the other two methods of designing dental crowns produced crowns that were either too large or too thin and failed to reach the same lifespan as that of natural teeth.

“This demonstrates that [the algorithm] could be utilised to design personalised dental crowns with high accuracy that can not only mimic both the morphology and biomechanics of natural teeth, but also operate without any additional human fine-tuning, thus saving additional costs in the production process,” added senior researcher Dr James Tsoi, an associate professor in dental materials science at the university.

Many AI approaches design a ‘lookalike’ product, but I believe this is the first project that functionalises data-driven AI into real dental application
—Dr James Tsoi, University of Hong Kong

“Many AI approaches design a ‘lookalike’ product, but I believe this is the first project that functionalises data-driven AI into real dental application. We hope this smart manufacturing technology will be the stepping stone for driving Industry 4.0 in dentistry, which is vital to meet the challenges of an ageing society and lack of dental personnel in Hong Kong,” Dr Tsoi stated. 

Clinical trials for using the generative AI for dental crowns are underway. Additionally, the researchers are working to expand the applicability of the tool to other dental prostheses, such as partial and complete dentures.

The study, titled “Morphology and mechanical performance of dental crown designed by 3D-DCGAN”, was published online in the March 2023 issue of Dental Materials.

To post a reply please login or register