Three-Dimensional Alveolar Bone Resorption Assessment and Periodontitis Staging Based on Deep Learning

doi:10.13701/j.cnki.kqyxyj.2025.08.010

Abstract

Abstract: Objective: To develop a multi-stage deep learning-based model for automatic quantitative assessment of three-dimensional alveolar bone resorption, providing intelligent quantitative indicators for periodontitis staging. Methods: Cone beam computed tomography (CBCT) scans from 100 periodontitis patients were collected. A multi-task segmentation strategy combining fully Convolutional Neural Network for volumetric medical image segmentation (V-Net) and transformer-based encoder-decoder network for efficient and accurate 3D medical image (UNETR++) was proposed to achieve high-precision segmentation of teeth, alveolar bone, and the cemento-enamel junction (CEJ). By combining dilation operations and contact-area ratio calculations, complemented by clustering in the CEJ direction, the three-dimensional alveolar bone resorption was quantitatively measured. Preliminary determination of periodontitis staging was acquired based on the 2018 new classification of periodontal diseases. Results: The model automatically segmented the teeth, alveolar bone, and CEJ, achieving Dice similarity coefficient (DSC) of 95.7%, 91.5%, and 87.9%. The model also demonstrated high accuracy and stability in terms of average surface distance (ASD), hausdorff distance (HD), and sensitivity (SEN). The model enabled multidimensional analysis of bone resorption in the mesiodistal and buccolingual directions. Compared with periodontists’ evaluations, the overall accuracy of periodontitis staging by this model was 85%, with a high level of agreement (Kappa=0.773). The model achieved an F1 score of 1.00 in identifying severe alveolar bone resorption (Stage Ⅲ or Ⅳ). A paired t-test (P<0.01) indicated a significant improvement in efficiency over manual assessment. Conclusion: This method can measure three-dimensional alveolar bone resorption quickly and accurately, providing a valuable reference for precise periodontitis staging and clinical diagnosis.

Key words: periodontitis, alveolar bone resorption rate, CBCT, deep learning, new classification of periodontal diseases

QIU Yue, HAN Yangping, DENG Guanghong, LI Jing. Three-Dimensional Alveolar Bone Resorption Assessment and Periodontitis Staging Based on Deep Learning[J]. Journal of Oral Science Research, 2025, 41(8): 695-699.

References

[1] Kassebaum NJ, Bernabe E, Dahiya M, et al. Global burden ofsevere periodontitis in 1990-2010: a systematic review and meta-regression [J]. J Dent Res, 2014, 93(11):1045-1053.
[2] Tonetti MS, Eickholz P, Loos BG, et al. Principles in prevention of periodontaldiseases: Consensus report of group 1 of the11th European Workshop on Periodontology on effective prevention of periodontal and peri-implant diseases [J]. J Clin Periodontol, 2015, 42(Suppl 16):S5-S11.
[3] Tonetti MS, Greenwell H, Kornman KS. Staging and gradingof periodontitis: Framework and proposal of a new classification and case definition [J]. J Clin Periodontol, 2018, 89(Suppl 1):S159-S172.
[4] Vandenberghe B, Jacobs R, Yang J. Detection of periodontalbone loss using digital intraoral and cone beam computed tomography images: an invitro assessment of bony and/or infrabony defects [J]. Dentomaxillofac Radiol, 2008, 37(5):252-260.
[5] Bornstein MM, Scarfe WC, Vaughn VM, et al. Cone beamcomputed tomography in implant dentistry: a systematic reviewfocusing on guidelines, indications, and radiation dose risks [J]. Int J Oral Maxillofac Implants, 2014, 29 Suppl:55-77.
[6] Singha A, Thakur RS, Patel T. Deep learning applications in medical image analysis.In: Dash S, Pani S K. Biomedical data mining for information retrieval: methodologies, techniques, and applications[M]. John Wiley & Sons, 2021: 293-350.
[7] Ni FD, Xu ZN, Liu MQ, et al. Towards clinically applicable automated mandibular canal segmentation on CBCT [J]. J Dent, 2024, 144:104931.
[8] Ying S, Wang B, Zhu H, et al. Caries segmentation on tooth X-ray images with a deep network [J]. J Dent, 2022, 119:104076.
[9] Fujima N, Andreu-Arasa VC, Meibom SK, et al. Deep learning analysis using FDG-PET to predict treatment outcome in patients with oral cavity squamous cellcarcinoma [J]. Eur Radiol, 2020, 30(11):6322-6330.
[10] Yu HJ, Cho SR, Kim MJ, et al. Automated skeletal classification with lateral cephalometry based on artificial intelligence [J]. J Dent Res, 2020, 99(3):249-256.
[11] Chang HJ, Lee SJ, Yong TH, et al. Deep learning hybrid method to automatically diagnose periodontal bone loss and stage periodontitis [J]. Sci Rep, 2020, 10(1):7531.
[12] Dujic H, Meyer O, Hoss P, et al. Automatized detection of periodontal bone losson periapical radiographs by vision transformer networks [J]. Diagnostics, 2023, 13(23):3562.
[13] Kim J, Lee HS, Song IS, et al. DeNTNet: Deep Neural Transfer Network for the detection of periodontal bone loss using panoramic dental radiographs[J]. Sci Rep, 2019, 9(1):17615.
[14] Krois J, Ekert T, Meinhold L, et al. Deep learning for the radiographic detection of periodontal bone loss [J]. Sci Rep, 2019, 9(1): 8495.
[15] Xue T, Chen L, Sun Q. Deep learning method to automatically diagnose periodontal bone loss and periodontitis stage in dental panoramic radiograph [J]. J Dent, 2024, 150: 105373.
[16] Zhang Y, Luo Z, Li S. A Multi-stage network with self-attention for tooth instance segmentation [C]. // Computer supported cooperative work and social computing. Singapore: Springer Nature Singapore, 2023: 438-450.
[17] Lang NP, Bartold PM. Periodontal health [J]. J Periodontol, 2018, 89 Suppl 1:S9-S16.
[18] Shaker AM, Maaz M, Rasheed H, et al. UNETR++: Delving into efficient and accurate 3D medical image segmentation [J]. IEEE Trans Med Imaging, 2024, 43(9):3377-3390.