Journal of the korean academy of Pediatric Dentistry (대한소아치과학회지)

Korean Academy of Pediatric Dentistry (대한소아치과학회)
권호 표지
DOI QR코드

DOI QR Code

Identification of Mesiodens Using Machine Learning Application in Panoramic Images

기계 학습 어플리케이션을 활용한 파노라마 영상에서의 정중 과잉치 식별

  • Seung, Jaegook (Department of Pediatric Dentistry and Institute of Oral Bioscience, School of Dentistry, Jeonbuk National University) ;
  • Kim, Jaegon (Department of Pediatric Dentistry and Institute of Oral Bioscience, School of Dentistry, Jeonbuk National University) ;
  • Yang, Yeonmi (Department of Pediatric Dentistry and Institute of Oral Bioscience, School of Dentistry, Jeonbuk National University) ;
  • Lim, Hyungbin (Department of Pediatric Dentistry and Institute of Oral Bioscience, School of Dentistry, Jeonbuk National University) ;
  • Le, Van Nhat Thang (Department of Pediatric Dentistry and Institute of Oral Bioscience, School of Dentistry, Jeonbuk National University) ;
  • Lee, Daewoo (Department of Pediatric Dentistry and Institute of Oral Bioscience, School of Dentistry, Jeonbuk National University)
  • 승재국 (전북대학교 치과대학 소아치과학교실 및 구강생체과학연구소) ;
  • 김재곤 (전북대학교 치과대학 소아치과학교실 및 구강생체과학연구소) ;
  • 양연미 (전북대학교 치과대학 소아치과학교실 및 구강생체과학연구소) ;
  • 임형빈 (전북대학교 치과대학 소아치과학교실 및 구강생체과학연구소) ;
  • 레반낫탕 (전북대학교 치과대학 소아치과학교실 및 구강생체과학연구소) ;
  • 이대우 (전북대학교 치과대학 소아치과학교실 및 구강생체과학연구소)
  • Received : 2021.04.01
  • Accepted : 2021.04.20
  • Published : 2021.05.31
Download PDF
⟨ Previous Next ⟩

Abstract

The aim of this study was to evaluate the use of easily accessible machine learning application to identify mesiodens, and to compare the ability to identify mesiodens between trained model and human. A total of 1604 panoramic images (805 images with mesiodens, 799 images without mesiodens) of patients aged 5 - 7 years were used for this study. The model used for machine learning was Google's teachable machine. Data set 1 was used to train model and to verify the model. Data set 2 was used to compare the ability between the learning model and human group. As a result of data set 1, the average accuracy of the model was 0.82. After testing data set 2, the accuracy of the model was 0.78. From the resident group and the student group, the accuracy was 0.82, 0.69. This study developed a model for identifying mesiodens using panoramic radiographs of children in primary and early mixed dentition. The classification accuracy of the model was lower than that of the resident group. However, the classification accuracy (0.78) was higher than that of dental students (0.69), so it could be used to assist the diagnosis of mesiodens for non-expert students or general dentists.

이번 연구는 손쉽게 접근 가능한 웹사이트 기반 기계 학습 어플리케이션을 활용하여 파노라마 방사선 영상에서 과잉치 식별 모델을 학습시키고, 학습된 모델의 과잉치를 식별하는 성능을 평가하고자 하였으며, 인간 집단과의 성능을 비교하기 위한 연구를 진행하였다. 총 1604장의 5 - 7세 환자의 파노라마 이미지가 이번 연구에서 사용되었다. 연구에 사용된 모델은 Google에서 개발한 기계학습 모델인 Teachable Machine을 사용하였다. 과잉치 식별 모델을 훈련시키고 성능을 평가하기 위해 data set 1을 설정하였다. Data set 2는 학습모델과 인간 집단 간의 정확도 비교를 위해 설정하였다. 학습모델 및 인간 집단의 과잉치 식별 능력을 평가하기 위해 정확도(accuracy), 민감도(sensitivity), 특이도(specificity) 값을 사용하였다. Data set 1의 검증 결과, 평균 0.82의 분류 정확도를 얻었다. Data set 2의 테스트 결과, 모델의 정확도는 0.78이었다. 전공의군과 학생군의 평균 정확도는 각각 0.82, 0.69였다. 이번 연구는 유치열기 및 초기 혼합치열기 어린이의 파노라마 방사선 영상과 웹 기반 기계 학습 어플리케이션 이용하여 과잉치 식별 모델을 개발하였고 학습된 모델과 인간 의사 집단(전공의 및 학생) 간의 과잉치 식별 정도를 비교 연구하였다. 훈련모델의 분류 정확도는 전공의군과 비교 시 낮았지만 훈련받지 않은 치과 대학 학생군보다 분류 정확도가 높아 비전문가 학생 또는 일반의사에게 과잉치 진단 정확도를 높이는 데 활용될 가능성이 있음을 확인하였다.

Keywords

Acknowledgement

This work was supported by the National Research Foundation of Korea(NRF) grant funded by the Korea government(MSIT) (No. 2020R1F1A1072484).

References

  1. Chen Y W, Stanley K, Att W : Artificial intelligence in dentistry: current applications and future perspectives. Quintessence Int, 51:248-257, 2020.
  2. Hwang J J, Jung Y H, Cho B H, et al. : An overview of deep learning in the field of dentistry. Imaging Sci Dent, 49:1-7, 2019. https://doi.org/10.5624/isd.2019.49.1.1
  3. Ariji Y, Yanashita Y, Kutsuna S, et al. : Automatic detection and classification of radiolucent lesions in the mandible on panoramic radiographs using a deep learning object detection technique. Oral Surg Oral Med Oral Pathol Oral Radiol, 128:424-430, 2019. https://doi.org/10.1016/j.oooo.2019.05.014
  4. Schwendicke F, Samek W, Krois J : Artificial Intelligence in Dentistry: Chances and Challenges. J Dent Res, 99:769-774, 2020. https://doi.org/10.1177/0022034520915714
  5. Anthonappa RP, King NM, Rabie AB : Diagnostic tools used to predict the prevalence of supernumerary teeth: a meta-analysis. Dentomaxillofac Radiol, 41:444-449, 2012. https://doi.org/10.1259/dmfr/19442214
  6. Tsiklakis K, Mitsea A, Tsichlaki A, et al. : A systematic review of relative indications and contra-indications for prescribing panoramic radiographs in dental paediatric patients. Eur Arch Paediatr Dent, 21:387-406, 2020. https://doi.org/10.1007/s40368-019-00478-w
  7. Shah A, Gill D S, Tredwin C, et al. : Diagnosis and management of supernumerary teeth. Dent Update, 35:510-512, 514-516, 519-520, 2008. https://doi.org/10.12968/denu.2008.35.8.510
  8. Rajab LD, Hamdan MA : Supernumerary teeth: review of the literature and a survey of 152 cases. Int J Paediatr Dent, 12:244-254, 2002. https://doi.org/10.1046/j.1365-263X.2002.00366.x
  9. Omer RS, Anthonappa RP, King NM : Determination of the optimum time for surgical removal of unerupted anterior supernumerary teeth. Pediatr Dent, 32:14-20, 2010.
  10. Ata-Ali F, Ata-Ali J, Penarrocha-Oltra D, et al. : Prevalence, etiology, diagnosis, treatment and complications of supernumerary teeth. J Clin Exp Dent, 6:414-418, 2014. https://doi.org/10.4317/jced.51499
  11. Park K, Lee D, Kim J, et al. : Timing for Removal of Mesiodens in Relation to the Maxillary Cental Incisors. J Korean Acad Pediatr Dent, 43:246-253, 2016.
  12. Katheria BC, Kau CH, Tate R, et al. : Effectiveness of impacted and supernumerary tooth diagnosis from traditional radiography versus cone beam computed tomography. Pediatr Dent, 32:304-309, 2010.
  13. Anthonappa RP, King NM, Rabie AB, et al. : Reliability of panoramic radiographs for identifying supernumerary teeth in children. Int J Paediatr Dent, 22:37-43, 2012. https://doi.org/10.1111/j.1365-263X.2011.01155.x
  14. Gavala S, Donta C, Tsiklakis K, et al. : Radiation dose reduction in direct digital panoramic radiography. Eur J Radiol, 71:42-48, 2009. https://doi.org/10.1016/j.ejrad.2008.03.018
  15. Kuwada C, Ariji Y, Fukuda M, et al. : Deep learning systems for detecting and classifying the presence of impacted supernumerary teeth in the maxillary incisor region on panoramic radiographs. Oral Surg Oral Med Oral Pathol Oral Radiol, 130:464-469, 2020. https://doi.org/10.1016/j.oooo.2020.04.813
  16. Ryu G, Song JS, Shin TJ, et al. : Retrospective study on three-dimensional characteristics of mesiodens using CBCT in pediatric dentistry. J Korean Acad Pediatr Dent, 48:77-94, 2021.
  17. Kilic MC, Bayrakdar IS, Celik O, et al. : Artificial intelligence system for automatic deciduous tooth detection and numbering in panoramic radiographs. Dentomaxillofac Radiol, 20200172, 2021.
  18. Abdalla-Aslan R, Yeshua T, Kabla D, et al. : An artificial intelligence system using machine-learning for automatic detection and classification of dental restorations in panoramic radiography. Oral Surg Oral Med Oral Pathol Oral Radiol, 130:593-602, 2020. https://doi.org/10.1016/j.oooo.2020.05.012
  19. Ekert T, Krois J, Meinhold L, et al. : Deep Learning for the Radiographic Detection of Apical Lesions. J Endod, 45:917-922.e5, 2019. https://doi.org/10.1016/j.joen.2019.03.016
  20. Lee J H, Kim D H, Jeong S N, et al. : Detection and diagnosis of dental caries using a deep learning-based convolutional neural network algorithm. J Dent, 77:106-111, 2018. https://doi.org/10.1016/j.jdent.2018.07.015
  21. Teachable Machine. Available from URL: https://teachablemachine.withgoogle.com (Accessed on March 30, 2021).