HKU Dentistry develops core technologies using intelligent generative AI

Leading researchers from the Faculty of Dentistry at the University of Hong Kong (HKU) have developed a new intelligent manufacture of dental crowns using generative artificial intelligence (AI) that leverages tooth manufacturing technology.

The team led by Dr James Tsoi, Associate Professor in Dental Materials Science collaborated with colleagues from HKU’s School of Engineering Computer Science Department to take a step forward for the production workflow of the next generation of AI-designed dental prostheses.

The researchers developed a generative AI algorithm that uses a true three-dimensional (3D) deep learning approach, generating high-accuracy personalized dental crowns that mimic the morphology and match materials required for natural tooth biomechanics. Biomechanical finite element analysis revealed that by using lithium silicate, AI-designed crowns can achieve the expected natural tooth age. In contrast, the two existing dental crown design methods result in crowns that are either too large or too thin, and do not conform to the same service life as natural teeth.

The results have been published in leading academic journals Dental Materials in an article entitled ‘Morphology and mechanical performance of dental crown designed by 3D-DCGAN’.

Currently, digital Computer-Aided Design and Manufacturing (CAD/CAM) workflows have significantly improved dentistry but still pose challenges. From design to fabrication of dental prostheses, the process is labor intensive, time consuming and poses health and environmental hazards during the 3D printing and milling process. The software uses a ‘dental library’ which contains predefined crown templates to help generate a prosthetic design but further customization is still required by the operator to suit individual conditions.

The intelligent manufacturing method developed by the research team can rise to the challenge and help replace conventional approaches to designing personalized dental crowns.

“We used a 3D-DCGAN (3D-Deep Convolutional Generative Adversarial Network) approach to ‘teach’ a ‘good’ AI algorithm design by feeding the algorithm with more than 600 cases of natural, healthy teeth. The algorithm improves design quality through internal competition between generators and discriminators,” said Dr Hao Ding, one of the researchers on the project.

“During the training process, the morphological features of natural teeth are studied by the algorithm, so that it can design dental crowns that are comparable to natural teeth — both morphologically and functionally.” Dr Ding added.

3D-DCGAN AI crowns were compared with natural teeth and with two other conventional CAD crown design methods. The results revealed that the AI-designed generative crowns had the lowest 3D differences, closest cusp angles (morphological features), and similar occlusal contacts (functional features) compared to natural teeth.

“This demonstrates that 3D-DCGAN can be used to design high-accuracy personalized dental crowns that can not only mimic the morphology and biomechanics of natural teeth, but also operate without additional adjustments by humans, thereby saving additional costs in production. process,” said lead researcher Dr James Tsoi.

“Many AI approaches design ‘look-alike’ products, but I believe this is the first project to translate data-driven AI into true dental applications. We hope that this intelligent manufacturing technology will be a stepping stone to propel Industry 4.0 in dentistry, which is critical to addressing the challenges of Hong Kong’s aging society and shortage of dental hygienists.” he added.

Dr Tsoi said the breakthrough marked an important step towards boosting the dental industry in the Great Bay Area, which generates USD 3.3 billion in annual revenue to manufacture 25-30% of dental prostheses globally, and to align with the National 14th Five Year Development plan. new forms of industrialization and informatization viz. smart smart manufacturing.

Clinical trials to use this generative AI for dental crowns are underway. The team is also working on applying this device to other dental prostheses such as bridges and dentures.

This study was supported by the General Research Fund (GRF), the Innovation and Technology Fund Mainland-Hong Kong Joint Funding Scheme (ITF-MHKJFS), and the Health and Medical Research Fund (HMRF). Initial results were presented by Dr Hao Ding at the 35th Annual Scientific Meeting of the International Association of Dental Research (IADR) Southeast Asia (SEA) and was awarded the leading IADR-SEA Research Category Award (Dental Materials and Biomaterials Category) in 2021.

Articles on Dental Materials entitled ‘Morphology and mechanical performance of dental crown designed by 3D-DCGAN’ published in Dental Materials, accessible via this link.

Research team
Faculty of Dentistry, University of Hong Kong
Principal Investigator
Dr James Kit Hon Tsoi, Associate Professor in Dental Materials Science

Co-Investigator
Dr Hao Ding, Postdoctoral Fellow in Dental Materials Science
Ms Yanning Chen, PhD Candidate in Dental Materials Science
Prof. Jukka Pekka Matinlinna, Honorary Professor of Dental Materials Science
Dr Edmond Ho Nang Pow, Clinical Associate Professor in Prosthodontics
Prof Michael Francis Burrow, Clinical Professor of Prosthodontics

Faculty of Engineering, University of Hong Kong
Dr Zhiming Cui, PhD graduate in Computer Science Department
Prof Wenping Wang, Professor Emeritus in the Department of Computer Science

Drexel University
Mr. Ebrahim Maghami, PhD Candidate in the Department of Mechanical and Mechanical Engineering

University of Minnesota Twin Cities
Prof Alex Siu Lun Fok, Professor at the Minnesota Dental Research Center for Biomaterials and Biomechanics

Media Inquiries:
Miss Melody Tang, Senior Communications Officer
Faculty of Dentistry, University of Hong Kong
Tel: 2859 0494 / Email: (email protected)