Proceedings of the
9th International Conference of Asian Society for Precision Engineering and Nanotechnology (ASPEN2022)
15 – 18 November 2022, Singapore
doi:10.3850/978-981-18-6021-8_OR-08-0183
Big Data and Artificial Intelligence (AI) Driven Dental Prostheses Design
1Division of Applied Oral Sciences & Community Dental Care, Faculty of Dentistry, The University of Hong Kong, 34 Hospital Road, Sai Ying Pun, Hong Kong
2Department of Computer Science, Faculty of Engineering, The University of Hong Kong, Pok Fu Lam, Hong Kong
3Department of Mechanical Engineering and Mechanics, Drexel University, 3141 Chestnut Street, Philadelphia, PA 19104, USA
4Restorative Dental Sciences, Faculty of Dentistry, The University of Hong Kong, 34 Hospital Road, Sai Ying Pun, Hong Kong
5Minnesota Dental Research Center for Biomaterials and Biomechanics, School of Dentistry, University of Minnesota, 515 Delaware St. SE, Minneapolis, MN 55455, USA
6Department of Visualization, College of Architecture, Texas A&M University, 789 Ross Street, 3137 TAMU College Station, TX 77843-3137, USA
ABSTRACT
OBJECTIVES:In prosthetic dentistry, Computer-Aided Design and Computer-Aided Manufacturing (CAD/CAM) has been widely used in the design and manufacturing process of dental crowns with high accuracy and efficiency compared with traditional way. Large amount of digitalised dental crown models have been created with the help of CAD. However, the digitalised prosthetic data is only used to assist CAM process. With the development of AI, big data and digital technologies, these data can be anticipated to guide the dental crown design process, thus achieving the transformation from knowledge-based design to big data driven design. In this work, a fully automatic dental crown design method by utilising AI and big data is presented with the potential of improving current partially digitalised dental crown design workflow.
METHODS: 500 sets of mandibular second premolars, their adjacent and antagonist teeth from healthy and young adults (19-22y.o.) were collected digitally, and machine learned with 3D-Deep Convolution Generative Adversarial Network (3D-DCGAN) approach. 12 sets of data were randomly selected as test dataset. The 12 natural teeth in the test dataset were compared with (1) our 3D-DCGAN design, (2) knowledge-based design (commercially available as CEREC), and (3) technician's design individually in parameters of 3D similarity, cusp angle, occlusal contact point number and area, and Finite Element (FE) static and fatigue simulation using Lithium disilicate ceramic as crown materials. The data were statistically analysed by SPSS 22.0 (IBM) at α=0.05.
RESULTS: 3D-DCGAN design and natural tooth had lowest discrepancy in morphology compared with other groups. Knowledge-based design showed a statistically significant (p<0.05) higher cusp angle compared with our 3D-DCGAN design and natural tooth. No significant difference was observed regarding the occlusal contact point number and area among all four groups. FE analysis results showed 3D-DCGAN design had a comparable performance with natural teeth regarding the stress distribution in crown, adhesive layer and dentine; the two groups also showed similar fatigue lifetimes under simulated cyclic loadings of 100-400 N.
CONCLUSION: Dental crowns designed by the big-data 3D-DCGAN method in this study showed no statistical differences among morphological, occlusal and mechanical parameters compared with natural teeth. This study demonstrated suitable AI can be utilised to design personalised dental crowns with high accuracy.
Keywords: 3D-GAN, Artificial Intelligence, Dental crown design, CAD/CAM
1Division of Applied Oral Sciences & Community Dental Care, Faculty of Dentistry, The University of Hong Kong, 34 Hospital Road, Sai Ying Pun, Hong Kong
2Department of Computer Science, Faculty of Engineering, The University of Hong Kong, Pok Fu Lam, Hong Kong
3Department of Mechanical Engineering and Mechanics, Drexel University, 3141 Chestnut Street, Philadelphia, PA 19104, USA
4Restorative Dental Sciences, Faculty of Dentistry, The University of Hong Kong, 34 Hospital Road, Sai Ying Pun, Hong Kong
5Minnesota Dental Research Center for Biomaterials and Biomechanics, School of Dentistry, University of Minnesota, 515 Delaware St. SE, Minneapolis, MN 55455, USA
6Department of Visualization, College of Architecture, Texas A&M University, 789 Ross Street, 3137 TAMU College Station, TX 77843-3137, USA
ABSTRACT
OBJECTIVES:In prosthetic dentistry, Computer-Aided Design and Computer-Aided Manufacturing (CAD/CAM) has been widely used in the design and manufacturing process of dental crowns with high accuracy and efficiency compared with traditional way. Large amount of digitalised dental crown models have been created with the help of CAD. However, the digitalised prosthetic data is only used to assist CAM process. With the development of AI, big data and digital technologies, these data can be anticipated to guide the dental crown design process, thus achieving the transformation from knowledge-based design to big data driven design. In this work, a fully automatic dental crown design method by utilising AI and big data is presented with the potential of improving current partially digitalised dental crown design workflow.
METHODS: 500 sets of mandibular second premolars, their adjacent and antagonist teeth from healthy and young adults (19-22y.o.) were collected digitally, and machine learned with 3D-Deep Convolution Generative Adversarial Network (3D-DCGAN) approach. 12 sets of data were randomly selected as test dataset. The 12 natural teeth in the test dataset were compared with (1) our 3D-DCGAN design, (2) knowledge-based design (commercially available as CEREC), and (3) technician's design individually in parameters of 3D similarity, cusp angle, occlusal contact point number and area, and Finite Element (FE) static and fatigue simulation using Lithium disilicate ceramic as crown materials. The data were statistically analysed by SPSS 22.0 (IBM) at α=0.05.
RESULTS: 3D-DCGAN design and natural tooth had lowest discrepancy in morphology compared with other groups. Knowledge-based design showed a statistically significant (p<0.05) higher cusp angle compared with our 3D-DCGAN design and natural tooth. No significant difference was observed regarding the occlusal contact point number and area among all four groups. FE analysis results showed 3D-DCGAN design had a comparable performance with natural teeth regarding the stress distribution in crown, adhesive layer and dentine; the two groups also showed similar fatigue lifetimes under simulated cyclic loadings of 100-400 N.
CONCLUSION: Dental crowns designed by the big-data 3D-DCGAN method in this study showed no statistical differences among morphological, occlusal and mechanical parameters compared with natural teeth. This study demonstrated suitable AI can be utilised to design personalised dental crowns with high accuracy.
Keywords: 3D-GAN, Artificial Intelligence, Dental crown design, CAD/CAM
