Artificial Intelligence-Driven Dental Age Estimation in Panoramic Radiographs via the Demirjian Method
Journal of Dental School,
Vol. 43 No. 3 (2025),
30 July 2025
,
Page 154-163
https://doi.org/10.22037/jds.v43i3.47402
Abstract
Objective(s): This research aimed to design and implement an artificial intelligence (AI) model for dental age estimation in panoramic radiographs using the Demirjian method. Accurate dental age estimation is crucial in forensic and clinical dentistry. Traditional methods for this purpose are typically time-consuming and require expertise, which increases the likelihood of human error. This study explored the potential of AI, particularly Vision Transformers (ViTs), to overcome these limitations. Methods: A number of 422 panoramic radiographs were analyzed, yielding 2,836 individual tooth images. The developmental stages of left mandibular teeth were determined according to the Demirjian method. Initially, 15% of the data was randomly separated for the test set. Subsequently, five-fold cross-validation was employed to partition the remaining data into training and validation sets. Three Convolutional Neural Network (CNN) models (ConvNeXt-Tiny, EfficientNet-V2-S, RegNet-Y-16GF) and three ViT models (ViT-B-16, Swin-V2-T, MaxVit-T) were trained using transfer learning. Model performance was evaluated using accuracy, precision, recall, and F1-score. The best-performing model, based on F1-score, was deployed in a web-based application. Mean Absolute Error (MAE) was used to assess the accuracy of AI-based age estimation compared to chronological age. Results: The Swin-V2-T model achieved the highest performance across all metrics (F1-score of 87%, Accuracy of 87.09%, Precision of 87.17%, and Recall of 87.09%). Analysis of AI-based age estimation accuracy revealed an MAE of 0.953 years. Conclusion: This study highlighted the applicability of AI, particularly ViT-based models, in automated dental age estimation using the Demirjian method. The developed AI model has the potential to streamline dental age assessments in both clinical and forensic settings, minimizing human error and enhancing efficiency.
- Dental Age Estimation
- Artificial Intelligence
- Vision Transformers
- Demirjian Method
How to Cite
References
Franco A, de Oliveira MN, Campos Vidigal MT, Blumenberg C, Pinheiro AA, Paranhos LR. Assessment of dental age estimation methods applied to Brazilian children: a systematic review and meta-analysis. Dentomaxillofac Radiol. 2021;50(2):20200128. doi: 10.1259/dmfr.20200128
Wen D, Ding Z, Tian Z, Wu W, Qu W, He W, et al. Comparing the accuracy of Demirjian and Nolla methods and establishing a new method for dental age estimation in northeastern Chinese children. Forensic Sci Res. 2022;7(4):685-93. doi: 10.1080/20961790.2021.2024655
Tyagi A, Srivastava N, Rana V, Kaushik N. Radiological and nonradiological methods of dental and skeletal age assessment: A narrative review. J Oral
Maxillofac Radiol. 2022;10(1):1-7. doi: 10.4103/jomr.jomr_5_22
Vila-Blanco N, Varas-Quintana P, Tomás I, Carreira MJ. A systematic overview of dental methods for age assessment in living individuals: from traditional to artificial intelligence-based approaches. Int J Legal Med. 2023;137(4):1117-46. doi: 10.1007/s00414-023-02960-z
Hostiuc S, Edison SE, Diaconescu I, Negoi I, Isaila OM. Accuracy of the Demirjian's method for assessing the age in children, from 1973 to 2020. A meta-analysis. Leg Med (Tokyo). 2021;52:101901. doi: 10.1016/j.legalmed.2021.101901
Verma M, Verma N, Sharma R, Sharma A. Dental age estimation methods in adult dentitions: An overview. J Forensic Dent Sci. 2019;11(2):57-63. doi: 10.4103/jfo.jfds_64_19
Nayyar AS, Babu BA, Krishnaveni B, Devi MV, Gayitri H. Age Estimation: Current State and Research Challenges. Journal of Medical Sciences. 2016;36(6):209-16. doi: 10.4103/1011-4564.196348
Shoukath A, Vidigal MTC, Vieira W, Paranhos LR, Mânica S, Franco A. Dental age estimation methods applied to Indian children and adolescents: A systematic review and meta-analysis. Morphologie. 2024;108(361):100758. doi: 10.1016/j.morpho.2023.100758
Demirjian A, Goldstein H, Tanner JM. A new system of dental age assessment. Hum Biol. 1973;45(2):211-27.
Vila-Blanco N, Carreira MJ, Varas-Quintana P, Balsa-Castro C, Tomas I. Deep Neural Networks for Chronological Age Estimation From OPG Images. IEEE Trans Med Imaging. 2020;39(7):2374-84. doi: 10.1109/TMI.2020.2968765
Kapoor P, Jain V. Comprehensive Chart for Dental Age Estimation (DAEcc8) based on Demirjian 8-teeth method: Simplified for operator ease. J Forensic Leg Med. 2018;59:45-9. doi: 10.1016/j.jflm.2018.07.014
Ong SH, Kim H, Song JS, Shin TJ, Hyun HK, Jang KT, Kim YJ. Fully automated deep learning approach to dental development assessment in panoramic radiographs. BMC Oral Health. 2024;24(1):426. doi: 10.1186/s12903-024-04160-6
Vila-Blanco N, Varas-Quintana P, Aneiros-Ardao Á, Tomás I, Carreira MJ. XAS: Automatic yet eXplainable Age and Sex determination by combining imprecise per-tooth predictions. Comput Biol Med. 2022;149:106072. doi: 10.1016/j.compbiomed.2022.106072
Aliyev R, Arslanoglu E, Yasa Y, Oktay AB, editors. Age Estimation from Pediatric Panoramic Dental Images with CNNs and LightGBM. 2022 Medical Technologies Congress (TIPTEKNO); 2022 31 Oct.-2 Nov. 2022. https://www.biyoklinikder.org/TIPTEKNO22_Bildiriler/025.pdf
Kurt A, Günaçar DN, Şılbır FY, Yeşil Z, Bayrakdar İ, Çelik Ö, et al. Evaluation of tooth development stages with deep learning-based artificial intelligence algorithm. BMC Oral Health. 2024;24(1):1034. doi: 10.1186/s12903-024-04786-6
Raghu M, Unterthiner T, Kornblith S, Zhang C, Dosovitskiy A. Do vision transformers see like convolutional neural networks? Advances in neural information processing systems. 2021;34:12116-28.
Takahashi S, Sakaguchi Y, Kouno N, Takasawa K, Ishizu K, Akagi Y, et al. Comparison of Vision Transformers and Convolutional Neural Networks in Medical Image Analysis: A Systematic Review. J Med Syst. 2024;48(1):84. doi: 10.1007/s10916-024-02105-8
Dosovitskiy A. An image is worth 16x16 words: Transformers for image recognition at scale. arXiv preprint arXiv:201011929. 2020. doi: 10.48550/arXiv.2010.11929
Vaswani A. Attention is all you need. Advances in Neural Information Processing Systems. 2017.
Liu Z, Mao H, Wu C-Y, Feichtenhofer C, Darrell T, Xie S, editors. A convnet for the 2020s. Proceedings of the IEEE/CVF conference on computer vision and pattern recognition; 2022. pp. 11976-86.
Tan M, Le Q, editors. Efficientnetv2: Smaller models and faster training. International conference on machine learning; 2021: PMLR. 139;P:10096-106.
Radosavovic I, Kosaraju RP, Girshick R, He K, Dollár P, editors. Designing network design spaces. Proceedings of the IEEE/CVF conference on computer vision and pattern recognition; 2020. P: 10428-36
Liu Z, Hu H, Lin Y, Yao Z, Xie Z, Wei Y, et al., editors. Swin transformer v2: Scaling up capacity and resolution. Proceedings of the IEEE/CVF conference on computer vision and pattern recognition; 2022. P: 12009-19.
Tu Z, Talebi H, Zhang H, Yang F, Milanfar P, Bovik A, Li Y, editors. Maxvit: Multi-axis vision transformer. European conference on computer vision; 2022: Springer.
Liu Z, Lin Y, Cao Y, Hu H, Wei Y, Zhang Z, et al., editors. Swin transformer: Hierarchical vision transformer using shifted windows. Proceedings of the IEEE/CVF international conference on computer vision; 2021; P:10012-22.
Tyagi K, Pathak G, Nijhawan R, Mittal A, editors. Detecting pneumonia using vision transformer and comparing with other techniques. 2021 5th international conference on electronics, communication and aerospace technology (ICECA); 2021: IEEE. doi: 10.1109/ICECA52323.2021.9676146
Wu Y, Qi S, Sun Y, Xia S, Yao Y, Qian W. A vision transformer for emphysema classification using CT images. Physics in Medicine & Biology. 2021;66(24):245016. doi: 10.1088/1361-6560/ac3dc8
Gheflati B, Rivaz H, editors. Vision transformers for classification of breast ultrasound images. 2022 44th Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC); 2022: IEEE. doi: 10.1109/EMBC48229.2022.9871809
- Abstract Viewed: 193 times
- PDF Downloaded: 182 times