Skip to main content


Log in

Accurate age classification using manual method and deep convolutional neural network based on orthopantomogram images

  • Original Article
  • Published:
International Journal of Legal Medicine Aims and scope Submit manuscript


Age estimation is an important challenge in many fields, including immigrant identification, legal requirements, and clinical treatments. Deep learning techniques have been applied for age estimation recently but lacking performance comparison between manual and machine learning methods based on a large sample of dental orthopantomograms (OPGs). In total, we collected 10,257 orthopantomograms for the study. We derived logistic regression linear models for each legal age threshold (14, 16, and 18 years old) for manual method and developed the end-to-end convolutional neural network (CNN) which classified the dental age directly to compare with the manual method. Both methods are based on left mandibular eight permanent teeth or the third molar separately. Our results show that compared with the manual methods (92.5%, 91.3%, and 91.8% for age thresholds of 14, 16, and 18, respectively), the end-to-end CNN models perform better (95.9%, 95.4%, and 92.3% for age thresholds of 14, 16, and 18, respectively). This work proves that CNN models can surpass humans in age classification, and the features extracted by machines may be different from that defined by human.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Fig. 1
Fig. 2
Fig. 3

Similar content being viewed by others


  1. Schmeling A, Olze A, Reisinger W, Geserick G (2001) Age estimation of living people undergoing criminal proceedings. Lancet 358:89–90.

    Article  CAS  PubMed  Google Scholar 

  2. Hagen M, Schmidt S, Schulz R, Vieth V, Ottow C, Olze A, Pfeiffer H, Schmeling A (2020) Forensic age assessment of living adolescents and young adults at the Institute of Legal Medicine, Munster, from 2009 to 2018. Int J Legal Med 134:745–751.

    Article  CAS  PubMed  Google Scholar 

  3. Thevissen PW, Kvaal SI, Willems G (2012) Ethics in age estimation of unaccompanied minors. J Forensic Odontostomatol 30(Suppl 1):84–102

    PubMed  Google Scholar 

  4. Befurt L, Kirchhoff G, Rudolf E, Schmeling A (2020) Legal aspects of forensic age diagnostics on the basis of 42f German Social Code (SGB) VIII. Rechtsmedizin 30:241–248.

    Article  Google Scholar 

  5. Olze A, Peschke C, Schulz R, Schmeling A (2008) Studies of the chronological course of wisdom tooth eruption in a German population. J Forensic Leg Med 15:426–429.

    Article  CAS  PubMed  Google Scholar 

  6. Bassed RB, Briggs C, Drummer OH (2011) Age estimation and the developing third molar tooth: an analysis of an Australian population using computed tomography. J Forensic Sci 56:1185–1191.

    Article  PubMed  Google Scholar 

  7. Kasper KA, Austin D, Kvanli AH, Rios TR, Senn DR (2009) Reliability of third molar development for age estimation in a Texas Hispanic population: a comparison study. J Forensic Sci 54:651–657.

    Article  PubMed  Google Scholar 

  8. Lengfu Z (2006) On minors’ civil liability ability. J Shangluo Teach Coll 20:59–63

    Google Scholar 

  9. Jun L (2005) Investigation on disputed issues about criminal responsibility of criminal minor in China. J Liaoning Univ (Phil Soc Sci) 33:34–39.

    Article  Google Scholar 

  10. Mincer HH, Harris EF, Berryman HE (1993) The A.B.F.O. study of third molar development and its use as an estimator of chronological age. J Forensic Sci 38:379–390

    Article  CAS  PubMed  Google Scholar 

  11. Cameriere R, Ferrante L, De Angelis D, Scarpino F, Galli F (2008) The comparison between measurement of open apices of third molars and Demirjian stages to test chronological age of over 18 year olds in living subjects. Int J Legal Med 122:493–497.

    Article  CAS  PubMed  Google Scholar 

  12. Olze A, Solheim T, Schulz R, Kupfer M, Pfeiffer H, Schmeling A (2010) Assessment of the radiographic visibility of the periodontal ligament in the lower third molars for the purpose of forensic age estimation in living individuals. Int J Legal Med 124:445–448.

    Article  PubMed  Google Scholar 

  13. Cameriere R, Brkic H, Ermenc B, Ferrante L, Ovsenik M, Cingolani M (2008) The measurement of open apices of teeth to test chronological age of over 14-year olds in living subjects. Forensic Sci Int 174:217–221.

    Article  PubMed  Google Scholar 

  14. Cameriere R, Velandia Palacio LA, Pinares J, Bestetti F, Paba R, Coccia E, Ferrante L (2018) Assessment of second (I2M) and third (I3M) molar indices for establishing 14 and 16 legal ages and validation of the Cameriere’s I3M cut-off for 18 years old in Chilean population. Forensic Sci Int 285(205):e1–e5.

    Article  Google Scholar 

  15. Lee JH, Han SS, Kim YH, Lee C, Kim I (2019) Application of a fully deep convolutional neural network to the automation of tooth segmentation on panoramic radiographs. Oral Surg Oral Med Oral Pathol Oral Radiol. 129:635–642.

    Article  PubMed  Google Scholar 

  16. Banar N, Bertels J, Laurent F, Boedi RM, de Tobel J, Thevissen P, Vandermeulen D (2020) Towards fully automated third molar development staging in panoramic radiographs. Int J Legal Med 134:1831–1841.

    Article  PubMed  Google Scholar 

  17. Merdietio Boedi R, Banar N, De Tobel J, Bertels J, Vandermeulen D, Thevissen PW (2020) Effect of lower third molar segmentations on automated tooth development staging using a convolutional neural network. J Forensic Sci 65:481–486.

    Article  PubMed  Google Scholar 

  18. Zheng Q, Ge Z, Du H, Li G (2021) Age estimation based on 3D pulp chamber segmentation of first molars from cone-beam-computed tomography by integrated deep learning and level set. Int J Legal Med 135:365–373.

    Article  PubMed  Google Scholar 

  19. Stern D, Payer C, Giuliani N, Urschler M (2019) Automatic age estimation and majority age classification from multi-factorial MRI data. IEEE J Biomed Health Inform 23:1392–1403.

    Article  PubMed  Google Scholar 

  20. Ekert T, Krois J, Meinhold L, Elhennawy K, Emara R, Golla T, Schwendicke F (2019) Deep learning for the radiographic detection of apical Lesions. J Endod 45:917–22 e5.

    Article  PubMed  Google Scholar 

  21. Lee JH, Kim DH, Jeong SN, Choi SH (2018) Diagnosis and prediction of periodontally compromised teeth using a deep learning-based convolutional neural network algorithm. J Periodontal Implant Sci 48:114–123.

    Article  PubMed  PubMed Central  Google Scholar 

  22. Chu P, Bo C, Liang X et al (2018) Using octuplet siamese network for osteoporosis analysis on dental panoramic radiographs. Conf Proc IEEE Eng Med Biol Soc 2018:2579–2582.

    Article  Google Scholar 

  23. Galibourg A, Cussat-Blanc S, Dumoncel J, Telmon N, Monsarrat P, Maret D (2021) Comparison of different machine learning approaches to predict dental age using Demirjian’s staging approach. Int J Legal Med. 135:665–675.

    Article  PubMed  Google Scholar 

  24. Demirjian A, Goldstein H, Tanner JM (1973) A new system of dental age assessment. Hum Biol 45:211–227

    CAS  PubMed  Google Scholar 

  25. Tan M, Le QV (2019) Efficientnet: rethinking model scaling for convolutional neural networks. arXiv:1905.11946

  26. Hu J, Shen L, Sun G (2020) Squeeze-and-excitation networks.  IEEE Transactions on Pattern Analysis and Machine Intelligence 42(8):2011–2023.

  27. Franklin D, Karkhanis S, Flavel A, Collini F, DeLuca S, Cameriere R (2016) Accuracy of a cut-off value based on the third molar index: Validation in an Australian population. Forensic Sci Int 266(575):e1–e6.

    Article  Google Scholar 

  28. Galic I, Vodanovic M, Cameriere R et al (2011) Accuracy of Cameriere, Haavikko, and Willems radiographic methods on age estimation on Bosnian-Herzegovian children age groups 6–13. Int J Legal Med 125:315–321.

    Article  PubMed  Google Scholar 

  29. Melo M, Ata-Ali J (2017) Accuracy of the estimation of dental age in comparison with chronological age in a Spanish sample of 2641 living subjects using the Demirjian and Nolla methods. Forensic Sci Int 270(276):e1–e7.

    Article  Google Scholar 

  30. Athanasiadou E, Geradts Z, Van Eijk E (2018) Camera recognition with deep learning. Forensic Sci Res 3:210–218.

    Article  PubMed  PubMed Central  Google Scholar 

  31. Bedeli M, Geradts Z, van Eijk E (2018) Clothing identification via deep learning: forensic applications. Forensic Sci Res 3:219–229.

    Article  PubMed  PubMed Central  Google Scholar 

  32. Yin W , Kann K , Yu M,  Schutze H (2017) Comparative study of CNN and RNN for natural language processing. arXiv:1702.01923

  33. Li Y, Huang C, Ding L, Li Z, Pan Y, Gao X (2019) Deep learning in bioinformatics: introduction, application, and perspective in the big data era. Methods 166:4–21.

    Article  CAS  PubMed  Google Scholar 

  34. Redmon J, Divvala S, Girshick R, Farhadi A (2016) You only look once: unified, Real-time object detection. Conf Proc IEEE Comput Vis Pattern Recongnit (CVPR) 2016:779–788.

  35. Liu W, Anguelov D, Erhan D, Szegedy C, Berg AC (2016) SSD: single shot multibox detector. Conf Proc Springer Cham European Conf on Comput Vis (ECCV) 2016:21–37.

  36. Cheng CT, Ho TY, Lee TY, Chang CC, Chou CC, Chen CC, Chung IF, Liao CH (2019) Application of a deep learning algorithm for detection and visualization of hip fractures on plain pelvic radiographs. Eur Radiol 29:5469–5477.

    Article  PubMed  PubMed Central  Google Scholar 

  37. He T, Guo J, Chen N, Xu X, Wang Z, Fu K, Liu L, Yi Z (2020) MediMLP: using Grad-CAM to extract crucial variables for lung cancer postoperative complication prediction. IEEE J Biomed Health Inform 24:1762–1771.

    Article  PubMed  Google Scholar 

  38. Iizuka T, Fukasawa M, Kameyama M (2019) Deep-learning-based imaging-classification identified cingulate island sign in dementia with Lewy bodies. Sci Rep 9:8944.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  39. Jonas S, Rossetti AO, Oddo M, Jenni S, Favaro P, Zubler F (2019) EEG-based outcome prediction after cardiac arrest with convolutional neural networks: performance and visualization of discriminative features. Hum Brain Mapp 40:4606–4617.

    Article  PubMed  PubMed Central  Google Scholar 

  40. Biuki N, Razi T, Faramarzi M (2017) Relationship between pulp-tooth volume ratios and chronological age in different anterior teeth on CBCT. J Clin Exp Dent 9:e688–ee93.

    Article  PubMed  PubMed Central  Google Scholar 

  41. Cameriere R, Ferrante L, Belcastro MG, Bonfiglioli B, Rastelli E, Cingolani M (2007) Age estimation by pulp/tooth ratio in canines by peri-apical X-rays. J Forensic Sci 52:166–170.

    Article  PubMed  Google Scholar 

  42. Kvaal SI, Kolltveit KM, Thomsen IO, Solheim T (1995) Age estimation of adults from dental radiographs. Forensic Sci Int 74:175–185.

    Article  CAS  PubMed  Google Scholar 

  43. Timme M, Borkert J, Nagelmann N, Schmeling A (2020) Evaluation of secondary dentin formation for forensic age assessment by means of semi-automatic segmented ultrahigh field 9.4 T UTE MRI datasets. Int J Legal Med 134:2283–2288.

    Article  PubMed  PubMed Central  Google Scholar 

  44. Molina A, Bravo M, Fonseca GM, Marquez-Grant N, Martin-de-Las-Heras S (2021) Dental age estimation based on pulp chamber/crown volume ratio measured on CBCT images in a Spanish population. Int J Legal Med 135:359–364.

    Article  PubMed  Google Scholar 

  45. Olze A, Solheim T, Schulz R, Kupfer M, Schmeling A (2010) Evaluation of the radiographic visibility of the root pulp in the lower third molars for the purpose of forensic age estimation in living individuals. Int J Legal Med 124:183–186.

    Article  PubMed  Google Scholar 

  46. Olze A, Hertel J, Schulz R, Wierer T, Schmeling A (2012) Radiographic evaluation of Gustafson’s criteria for the purpose of forensic age diagnostics. Int J Legal Med 126:615–621.

    Article  PubMed  Google Scholar 

  47. Guo YC, Li MJ, Olze A, Schmidt S, Schulz R, Zhou H, Pfeiffer H, Chen T, Schmeling A (2018) Studies on the radiographic visibility of the periodontal ligament in lower third molars: can the Olze method be used in the Chinese population? Int J Legal Med 132:617–622.

    Article  PubMed  Google Scholar 

Download references


This work was supported by the National Key Research and Development Program of China under Grant No. 2017YFA0700800, the National Natural Science Foundation of China under Grant Nos. 81701869, 61971343 and 61627811, and the Key Research and Development Program of Shaanxi Province of China under Grant No. 2020GXLH-Y-008.

Author information

Authors and Affiliations


Corresponding author

Correspondence to Shaoyi Du.

Ethics declarations

Ethical approval was granted by the ethics committee of Stomatological Hospital of Xi’an Jiaotong University.

Conflict of interest

The authors declare no competing interests.

Ethics approval

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. For this type of study, formal consent is not required.

Consent to participate

For this type of study, formal consent is not required.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Guo, Yc., Han, M., Chi, Y. et al. Accurate age classification using manual method and deep convolutional neural network based on orthopantomogram images. Int J Legal Med 135, 1589–1597 (2021).

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: