Journal of Technologic Dentistry (대한치과기공학회지)

Korean Academy of Dental Technology (대한치과기공학회)
권호 표지
DOI QR코드

DOI QR Code

Morphological analysis of virtual teeth generated by deep learning

딥러닝으로 생성된 가상 치아의 형태학적 분석 연구

  • Eun-Jeong Bae (Department of Dental Technology, Bucheon University)
  • 배은정 (부천대학교 치기공과)
  • Received : 2024.08.22
  • Accepted : 2024.09.12
  • Published : 2024.09.30
Download PDF
⟨ Previous Next ⟩

Abstract

Purpose: This study aimed to generate virtual mandibular first molars using deep learning technology, specifically deep convolutional generative adversarial network (DCGAN), and evaluate the accuracy and reliability of these virtual teeth by analyzing their morphological characteristics. These morphological characteristics were classified based on various evaluation criteria, facilitating the assessment of deep learning-based dental prosthesis production's practical applicability. Methods: Based on our previous research, 1,000 virtual mandibular first molars were generated, and based on morphological criteria, categorized as matching, non-matching, and partially matching. The generated first molars or the categorization of the generated molars were analyzed through the expert judgment of dental technicians. Results: Among the 1,000 generated virtual teeth, 143 (14.3%) met all five evaluation criteria, whereas 76 (7.6%) were judged as completely non-matching. The most frequent issue, with 781 (78.1%) instances, including some overlapping instances, was related to occlusal buccal cusp discrepancies. Conclusion: The study reveals the potential of DCGAN-generated virtual teeth as substitutes for real teeth; however, additional research and improvements in data quality are necessary to enhance accuracy. Continued data collection and refinement of generation methods can maximize the practicality and utility of deep learning-based dental prosthesis production.

Keywords

References

  1. Tsoi JKH, Ding H, Burrow M, Fok A, Wang W. Premolar and molar crown generation using 3D-GAN method. Dent Mater. 2023;39(Suppl 1):e76.
  2. Sivari E, Senirkentli GB, Bostanci E, Guzel MS, Acici K, Asuroglu T. Deep learning in diagnosis of dental anomalies and diseases: a systematic review. Diagnostics (Basel). 2023;13:2512.
  3. Kim M, Bae HJ. Data augmentation techniques for deep learning-based medical image analyses. J Korean Soc Radiol. 2020;81:1290-1304.
  4. Zhao D, Zhu D, Lu J, Luo Y, Zhang G. Synthetic medical images using F&BGAN for improved lung nodules classification by multi-scale VGG16. Symmetry. 2018;10:519.
  5. Ding H, Cui Z, Maghami E, Chen Y, Matinlinna JP, Pow EHN, et al. Morphology and mechanical performance of dental crown designed by 3D-DCGAN. Dent Mater. 2023;39:320-332.
  6. Nguyen TT, Larrivee N, Lee A, Bilaniuk O, Durand R. Use of artificial intelligence in dentistry: current clinical trends and research advances. J Can Dent Assoc. 2021;87:l7.
  7. Wang H, Minnema J, Batenburg KJ, Forouzanfar T, Hu FJ, Wu G. Multiclass CBCT image segmentation for orthodontics with deep learning. J Dent Res. 2021;100:943-949.
  8. Dobrzanski LA, Dobrzanski LB. Dentistry 4.0 concept in the design and manufacturing of prosthetic dental restorations. Processes. 2020;8:525.
  9. Khanna NN, Maindarkar MA, Viswanathan V, Fernandes JFE, Paul S, Bhagawati M, et al. Economics of artificial intelligence in healthcare: diagnosis vs. treatment. Healthcare (Basel). 2022;10:2493.
  10. Goodfellow I, Pouget-Abadie J, Mirza M, Xu B, Warde-Farley D, Ozair S, et al. Generative adversarial nets. In: Ghahramani Z, Welling M, Cortes C, Lawrence N, Weinberger KQ, eds. Advances in neural information processing systems 27. Curran Associates, Inc., 2014. p. 2672-2680.
  11. Bae EJ, Ihm SY. Generation of virtual mandibular first molar teeth and accuracy analysis using deep convolutional generative adversarial network. J Tech Dent. 2024;46:36-41.
  12. Lee S, Kwon D, Ji Y. A study on the deep learning-based interpolation for time-series data with continuous missing values. Paper presented at: Korea Institute of Communications and Information Sciences (KICS) Fall Conference 2023; 2023 Nov 22-24; Gyeongju, Korea. Seoul: KICS, 2023. p. 765-766.
  13. Jang Y, Yoo J, Hong H. Assessment and analysis of fidelity and diversity for GAN-based medical image generative model. J Korea Comput Graph Soc. 2022;28:11-19.
  14. Fang W, Zhang F, Sheng VS, Ding Y. A method for improving CNN-based image recognition using DCGAN. Comput Mater Continua. 2018;57:167-178.
  15. Gulrajani I, Ahmed F, Arjovsky M, Dumoulin V, Courville A. Improved training of wasserstein GANs. Paper presented at: 31st International Conference on Neural Information Processing Systems; 2017 Dec 4-9; Long Beach, CA, USA. Red Hook, NY: ACM, 2017. p. 5769-5779.