DOI QR코드

DOI QR Code

Generation of virtual mandibular first molar teeth and accuracy analysis using deep convolutional generative adversarial network

심층 합성곱 생성적 적대 신경망을 활용한 하악 제1대구치 가상 치아 생성 및 정확도 분석

  • Eun-Jeong Bae (Department of Dental Technology, Bucheon University) ;
  • Sun-Young Ihm (Department of Computer Engineering, Pai Chai University)
  • Received : 2024.06.12
  • Accepted : 2024.06.18
  • Published : 2024.06.30

Abstract

Purpose: This study aimed to generate virtual mandibular left first molar teeth using deep convolutional generative adversarial networks (DCGANs) and analyze their matching accuracy with actual tooth morphology to propose a new paradigm for using medical data. Methods: Occlusal surface images of the mandibular left first molar scanned using a dental model scanner were analyzed using DCGANs. Overall, 100 training sets comprising 50 original and 50 background-removed images were created, thus generating 1,000 virtual teeth. These virtual teeth were classified based on the number of cusps and occlusal surface ratio, and subsequently, were analyzed for consistency by expert dental technicians over three rounds of examination. Statistical analysis was conducted using IBM SPSS Statistics ver. 23.0 (IBM), including intraclass correlation coefficient for intrarater reliability, one-way ANOVA, and Tukey's post-hoc analysis. Results: Virtual mandibular left first molars exhibited high consistency in the occlusal surface ratio but varied in other criteria. Moreover, consistency was the highest in the occlusal buccal lingual criteria at 91.9%, whereas discrepancies were observed most in the occusal buccal cusp criteria at 85.5%. Significant differences were observed among all groups (p<0.05). Conclusion: Based on the classification of the virtually generated left mandibular first molar according to several criteria, DCGANs can generate virtual data highly similar to real data. Thus, subsequent research in the dental field, including the development of improved neural network structures, is necessary.

Keywords

Acknowledgement

This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (No. 2021R1C1C2011105) and Innovative Human Resource Development for Local Intellectualization program through the Institute of Information & Communications Technology Planning & Evaluation (IITP) grant funded by the Korea government (MSIT)(IITP-2024-RS-2022-00156334).

References

  1. Liu Z, Zhang Z, Lei Z, Omura M, Wang RL, Gao S. Dendritic deep learning for medical segmentation. IEEE/CAA J Autom Sin. 2024;11:803-805. 
  2. Kebaili A, Lapuyade-Lahorgue J, Ruan S. Deep learning approaches for data augmentation in medical imaging: a review. J Imaging. 2023;9:81. 
  3. Bouali R, Mahboub O, Lazaar M. Review of dental diagnosis by deep learning models: trends, applications and challenges. Procedia Comput Sci. 2024;231:221-228.  https://doi.org/10.1016/j.procs.2023.12.196
  4. Bae EJ, Jeong J, Son Y, Lim J. A study on virtual tooth image generation using deep learning - based on the number of learning. J Korean Acad Dent Technol. 2020;42:1-8.  https://doi.org/10.14347/kadt.2020.42.1.1
  5. Imangaliyev S, van der Veen MH, Volgenant CMC, Keijser BJF, Crielaard W, Levin E. Deep learning for classification of dental plaque images. In: Pardalos P, Conca P, Giuffrida G, Nicosia G, eds. Machine learning, optimization, and big data. MOD 2016. Cham: Springer, 2016. p. 407-410. 
  6. Patil HK, Seshadri R. Big data security and privacy issues in healthcare. Paper presented at: 2014 IEEE International Congress on Big Data; 2014 Jun 27-Jul 2; Anchorage, AK, USA. Piscataway, NJ: IEEE, 2014. pp. 762-765. 
  7. Zhang Q, Wang H, Lu H, Won D, Yoon SW. Medical image synthesis with generative adversarial networks for tissue recognition. Paper presented at: 2018 IEEE International Conference on Healthcare Informatics; 2018 Jun 4-7; New York, NY, USA. Piscataway, NJ: IEEE, 2018. pp. 199-207. 
  8. Yu S. Big privacy: challenges and opportunities of privacy study in the age of big data. IEEE Access. 2016;4:2751-2763.  https://doi.org/10.1109/ACCESS.2016.2577036
  9. Radford A, Metz L, Chintala S. Unsupervised representation learning with deep convolutional generative adversarial networks. arXiv 2015. https://doi.org/10.48550/arXiv.1511.06434 [Epub ahead of print] 
  10. 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.  https://doi.org/10.1016/j.dental.2023.02.001
  11. 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. 
  12. Yang S, Kim KD, Ariji E, Kise Y. Generative adversarial networks in dental imaging: a systematic review. Oral Radiol. 2024;40:93-108.  https://doi.org/10.1007/s11282-023-00719-1
  13. Broll A, Goldhacker M, Hahnel S, Rosentritt M. Generative deep learning approaches for the design of dental restorations: a narrative review. J Dent. 2024;145:104988. 
  14. Meredith HV, Hixon EH. Frequency, size, and bilateralism of Carabelli's tubercle. J Dent Res. 1954;33:435-440.  https://doi.org/10.1177/00220345540330031601
  15. Stuart CE. Good occlusion for natural teeth. J Prosthet Dent. 1964;14:716-724.  https://doi.org/10.1016/0022-3913(64)90207-0
  16. Hellman M. Racial characters in human dentition part I. A racial distribution of the Dryopithecus pattern and its modifications in the lower molar teeth of man. Proc Am Philos Soc. 1928;67:157-174. 
  17. Andrews LF. The six keys to normal occlusion. Am J Orthod. 1972;62:296-309.  https://doi.org/10.1016/S0002-9416(72)90268-0
  18. Christie TE. Cephalometric patterns of adults with normal occlusion. Angle Orthod. 1977;47:128-135. 
  19. Koo TK, Li MY. A guideline of selecting and reporting intraclass correlation coefficients for reliability research. J Chiropr Med. 2016;15:155-163.  https://doi.org/10.1016/j.jcm.2016.02.012
  20. Portney LG, Watkins MP. Foundations of clinical research: applications to practice. 3rd ed. Pearson/Prentice Hall, 2009. p. 892. 
  21. Lucic M, Kurach K, Michalski M, Bousquet O, Gelly S. Are GANs created equal? A large-scale study. Paper presented at: 32nd International Conference on Neural Information Processing Systems; 2018 Dec 3-8; Montreal, Quebec, Canada. Red Hook, NJ: Curran Associates Inc., 2018. pp. 698-707. 
  22. Kim HY. Statistical notes for clinical researchers: evaluation of measurement error 1: using intraclass correlation coefficients. Restor Dent Endod. 2013;38:98-102. https://doi.org/10.5395/rde.2013.38.2.98