Automated processing of intraoral clinical photographs using deep learning techniques

Background: Intraoral clinical photographs are essential for diagnosis and treatment planning; however, capturing and managing high-quality images in pediatric dentistry is challenging. This study aimed to develop a deep-learning model for classifying and editing five types of intraoral clinical photographs (frontal, left buccal, right buccal, upper occlusal, and lower occlusal) of pediatric patients. Methods: A total of 3100 intraoral clinical photographs of 620 pediatric patients were used. The images were aligned to the normal occlusal plane and annotated into five categories. Two deep convolutional neural networks were implemented: Inception-ResNet-v2 for image orientation regression and Faster Region-based Convolutional Neural Network (R-CNN) for region detection. The models were trained and validated using five-fold cross-validation in Matrix Laboratory (MATLAB) 2024b, and their performance was assessed based on root mean squared error (RMSE), mean absolute error (MAE), classification accuracy, Intersection over Union (IoU), precision, recall, and average precision. Results: The image orientation correction network yielded a mean RMSE of 0.571◦ and an MAE of 0.407◦ in a five-fold cross-validation. The region detection network achieved a high classification accuracy across the five intraoral categories, with values ranging from 0.977 to 0.997. As the IoU threshold increased, average precision values decreased. Conclusions: The results show that our proposed method enables automated processing of intraoral photographs, particularly in terms of image rotation correction, cropping, and classification, with sufficiently high accuracy.

Artificial intelligence; Convolutional neural network; Deep learning; Image processing; Pediatric dentistry; Photography; Dental

[1] Li S, Guo Z, Lin J, Ying S. Artificial intelligence for classifying and archiving orthodontic images. BioMed Research International. 2022; 2022: 1473977.

[2] Ryu J, Kim YH, Kim TW, Jung SK. Evaluation of artificial intelligence model for crowding categorization and extraction diagnosis using intraoral photographs. Scientific Reports. 2023; 13: 5177.

[3] Ryu J, Lee YS, Mo SP, Lim K, Jung SK, Kim TW. Application of deep learning artificial intelligence technique to the classification of clinical orthodontic photos. BMC Oral Health. 2022; 22: 454.

[4] Schulz-Weidner N, Gruber M, Schraml EM, Wöstmann B, Krämer N, Schlenz MA. Improving the communication of dental findings in pediatric dentistry by using intraoral scans as a visual aid: a randomized clinical trial. Dentistry Journal. 2024; 12: 15.

[5] Alqadi A, O’Connell AC. Dental photography for children: a global survey. International Journal of Paediatric Dentistry. 2024; 34: 790–798.

[6] Selvakumar BG, Elavarasu PK, Vinodhini V, Ponnusamy P, Venkatachalapathy S, Abinaya R. Third eye of dentistry—digital dental photography: a literature review (part 1). Journal of Primary Care Dentistry and Oral Health. 2024; 5: 101–103.

[7] Al Omer H, Al Firdous RA, Al Shamrani SH, Al Anazi SS, Natto ZS, Al Dayel O. Assessment of dental photography imaging as a diagnostic tool for incipient pits and fissures caries in permanent posterior teeth. Edelweiss Applied Science and Technology. 2025; 9: 638–655.

[8] Ashtiani GH, Sabbagh S, Moradi S, Azimi S, Ravaghi V. Diagnostic accuracy of tele-dentistry in screening children for dental caries by community health workers in a lower-middle-income country. International Journal of Paediatric Dentistry. 2024; 34: 567–575.

[9] Werle SB, Piva F, Assunção C, Guimarães LF, de Araújo FB, Coelho-de-Souza FH. Photography in pediatric dentistry: basis and applications. Revista Odonto Ciencia. 2015; 30: 60–64.

[10] Sakr L, Abbas H, Thabet N, Abdelgawad F. Reliability of teledentistry mobile photos versus conventional clinical examination for dental caries diagnosis on occlusal surfaces in a group of school children: a diagnostic accuracy study. BMC Oral Health. 2025; 25: 545.

[11] Liu J, Chen Y, Li S, Zhao Z, Wu Z. Machine learning in orthodontics: challenges and perspectives. Advances in Clinical and Experimental Medicine. 2021; 30: 1065–1074.

[12] Kühnisch J, Meyer O, Hesenius M, Hickel R, Gruhn V. Caries detection on intraoral images using artificial intelligence. Journal of Dental Research. 2022; 101: 158–165.

[13] You W, Hao A, Li S, Wang Y, Xia B. Deep learning-based dental plaque detection on primary teeth: a comparison with clinical assessments. BMC Oral Health. 2020; 20: 141.

[14] Zhang X, Liang Y, Li W, Liu C, Gu D, Sun W, et al. Development and evaluation of deep learning for screening dental caries from oral photographs. Oral Diseases. 2022; 28: 173–181.

[15] Schwendicke F, Chaurasia A, Arsiwala L, Lee JH, Elhennawy K, Jost-Brinkmann PG, et al. Deep learning for cephalometric landmark detection: systematic review and meta-analysis. Clinical Oral Investigations. 2021; 25: 4299–4309.

[16] Rodriguez R, Dokladalova E, Dokládal P. ‘Rotation invariant CNN using scattering transform for image classification’, 2019 IEEE international conference on image processing. Taipei, Taiwan, 22–25 September 2019. IEEE: Piscataway, USA. 2019.

[17] Liu S, Johns E, Davison AJ. ‘End-to-end multi-task learning with attention’, proceedings of the IEEE/CVF conference on computer vision and pattern recognition. Long Beach, USA, 16–20 June 2019. IEEE: Piscataway, USA. 2019.

[18] Desai V, Bumb D. Digital dental photography: a contemporary revolution. International Journal of Clinical Pediatric Dentistry. 2013; 6: 193–196.

[19] Marcato L, Sandler J. The best choice of equipment to obtain high quality standardised results in intra-oral photography—a comparison between the common practice in the UK and the gold standard set by the literature. Journal of Visual Communication in Medicine. 2018; 41: 90–96.

[20] Kang S, Shon B, Park EY, Jeong S, Kim EK. Diagnostic accuracy of dental caries detection using ensemble techniques in deep learning with intraoral camera images. PLOS ONE. 2024; 19: e0310004.

[21] Girshick R. Fast R-CNN. 2015 IEEE international conference on computer vision. Santiago, Chile, 7–13 December 2015. IEEE: Piscataway, USA. 2015.

[22] Sumit SB, Joshi S, Rana U. Comprehensive review of R-CNN and its variant architectures. International Research Journal on Advanced Engineering Hub. 2024; 2: 959–966.

[23] Chai T, Draxler RR. Root mean square error (RMSE) or mean absolute error (MAE)?—Arguments against avoiding RMSE in the literature. Geoscientific Model Development. 2014; 7: 1247–1250.

[24] Hodson TO. Root-mean-square error (RMSE) or mean absolute error (MAE): when to use them or not. Geoscientific Model Development Discussions. 2022; 15: 5481–5487.

[25] Alhuwaish HA, Almoammar KA. Development and validation of an occlusal cant index. BMC Oral Health. 2022; 22: 127.

[26] Khan AS, Nagar P, Singh P, Bharti M. Changes in the sequence of eruption of permanent teeth; correlation between chronological and dental age and effects of body mass index of 5–15-year-old schoolchildren. International Journal of Clinical Pediatirc Dentistry. 2020; 13: 368–380.

[27] Pandey K, Ali M, Kumar Verma A, Ahmad N, Chaturvedi S, Deo K. Relating mandibular incisor to the lingual frenum in dentulous and edentulous (complete denture wearers) subjects: an in vitro study. British Journal of Medicine & Medical Research. 2016; 12: 1–8.

[28] Beuer F, Schweiger J, Edelhoff D. Digital dentistry: an overview of recent developments for CAD/CAM generated restorations. British Dental Journal. 2008; 204: 505–511.

[29] Alharbi SS, Alhasson HF. Exploring the applications of artificial intelligence in dental image detection: a systematic review. Diagnostics. 2024; 14: 2442.

[30] Singh V, Pencina M, Einstein AJ, Liang JX, Berman DS, Slomka P. Impact of train/test sample regimen on performance estimate stability of machine learning in cardiovascular imaging. Scientific Reports. 2021; 11: 14490.

[31] Schwendicke F, Samek W, Krois J. Artificial intelligence in dentistry: chances and challenges. Journal of Dental Research. 2020; 99: 769–774.

[32] Hartman H, Nurdin D, Akbar S, Cahyanto A, Setiawan AS. Exploring the potential of artificial intelligence in paediatric dentistry: a systematic review on deep learning algorithms for dental anomaly detection. International Journal of Paediatric Dentistry. 2024; 34: 639–652.

[33] Badano A, Revie C, Casertano A, Cheng WC, Green P, Kimpe T, et al. Consistency and standardization of color in medical imaging: a consensus report. Journal of Digital Imaging. 2015; 28: 41–52.

[34] Elyan E, Vuttipittayamongkol P, Johnston P, Martin K, McPherson K, Moreno-García CF, et al. Computer vision and machine learning for medical image analysis: recent advances, challenges, and way forward. Artificial Intelligence Surgery. 2022; 2: 24–45.

[35] Habib AR, Xu Y, Bock K, Mohanty S, Sederholm T, Weeks WB, et al. Evaluating the generalizability of deep learning image classification algorithms to detect middle ear disease using otoscopy. Scientific Reports. 2023; 13: 5368.

[36] Zhang JW, Fan J, Zhao FB, Ma BE, Shen XQ, Geng YM. Diagnostic accuracy of artificial intelligence-assisted caries detection: a clinical evaluation. BMC Oral Health. 2024; 24: 1095.

[37] Cejudo Grano de Oro JE, Koch PJ, Krois J, Garcia Cantu Ros A, Patel J, Meyer-Lueckel H, et al. Hyperparameter tuning and automatic image augmentation for deep learning-based angle classification on intraoral photographs—a retrospective study. Diagnostics. 2022; 12: 1526.