The aim of this work was to assemble a large annotated dataset of bitewing radiographs and to use convolutional neural networks to automate the detection of dental caries in bitewing radiographs with human-level performance.
Materials and methods
A dataset of 3989 bitewing radiographs was created, and 7257 carious lesions were annotated using minimal bounding boxes. The dataset was then divided into 3 parts for the training (70%), validation (15%), and testing (15%) of multiple object detection convolutional neural networks (CNN). The tested CNN architectures included YOLOv5, Faster R-CNN, RetinaNet, and EfficientDet. To further improve the detection performance, model ensembling was used, and nested predictions were removed during post-processing. The models were compared in terms of the \(F_1\) score and average precision (AP) with various thresholds of the intersection over union (IoU).
The twelve tested architectures had \(F_1\) scores of 0.72–0.76. Their performance was improved by ensembling which increased the \(F_1\) score to 0.79–0.80. The best-performing ensemble detected caries with the precision of 0.83, recall of 0.77, \(F_1=0.80\), and AP of 0.86 at IoU=0.5. Small carious lesions were predicted with slightly lower accuracy (AP 0.82) than medium or large lesions (AP 0.88).
The trained ensemble of object detection CNNs detected caries with satisfactory accuracy and performed at least as well as experienced dentists (see companion paper, Part II). The performance on small lesions was likely limited by inconsistencies in the training dataset.
Caries can be automatically detected using convolutional neural networks. However, detecting incipient carious lesions remains challenging.
This is a preview of subscription content, access via your institution.
Buy single article
Instant access to the full article PDF.
Price excludes VAT (USA)
Tax calculation will be finalised during checkout.
Bayrakdar IS, Orhan K, Akarsu S, et al (2021) Deep-learning approach for caries detection and segmentation on dental bitewing radiographs. Oral Radiol 38(4). https://doi.org/10.1007/s11282-021-00577-9
Bayraktar Y, Ayan E (2021) Diagnosis of interproximal caries lesions with deep convolutional neural network in digital bitewing radiographs. Clin Oral Investig 26(1). https://doi.org/10.1007/s00784-021-04040-1
Bochkovskiy A, Wang C, Liao HM (2020) YOLOv4: optimal speed and accuracy of object detection. CoRR abs/2004.10934. https://doi.org/10.48550/arXiv.2004.10934
Bodla N, Singh B, Chellappa R, et al (2017) Soft-NMS – improving object detection with one line of code. In: International conference on computer vision (ICCV), pp 5561–5569. https://doi.org/10.48550/ARXIV.1704.04503
Cantu AG, Gehrung S, Krois J et al (2020) Detecting caries lesions of different radiographic extension on bitewings using deep learning. J Dent 100:103425. https://doi.org/10.1016/j.jdent.2020.103425
Chen L, Li S, Bai Q et al (2021) Review of image classification algorithms based on convolutional neural networks. Remote Sens 13(22):4712. https://doi.org/10.3390/rs13224712
Chen X, Guo J, Ye J et al (2023) Detection of proximal caries lesions on bitewing radiographs using deep learning method. Caries Res 56(5–6):455–463. https://doi.org/10.1159/000527418
Estai M, Tennant M, Gebauer D et al (2023) Evaluation of a deep learning system for automatic detection of proximal surface dental caries on bitewing radiographs. Oral Surg Oral Med Oral Pathol Oral Radiol 134(2):262–270. https://doi.org/10.1016/j.oooo.2022.03.008
García-Cañas A, Bonfanti-Gris M, Paraíso-Medina S et al (2022) Diagnosis of interproximal caries lesions in bitewing radiographs using a deep convolutional neural network-based software. Caries Res 56(5–6):503–511. https://doi.org/10.1159/000527491
He K, Zhang X, Ren S, et al (2015) Delving deep into rectifiers: surpassing human-level performance on ImageNet classification. In: IEEE International conference on computer vision (ICCV), pp 1026–1034. https://doi.org/10.1109/ICCV.2015.123
He K, Zhang X, Ren S, et al (2016) Deep residual learning for image recognition. In: IEEE Conference on computer vision and pattern recognition (CVPR), pp 770–778. https://doi.org/10.1109/CVPR.2016.90
Jocher G, Chaurasia A, Stoken A, et al (2022) YOLOv5 SOTA realtime instance segmentation. https://doi.org/10.5281/zenodo.7347926
Khanagar SB, Al-ehaideb A, Maganur PC et al (2021) Developments, application, and performance of artificial intelligence in dentistry–a systematic review. J Dent Sci 16(1):508–522. https://doi.org/10.1016/j.jds.2020.06.019
Kuang W, Ye W, (2008) A kernel-modified SVM based computer-aided diagnosis system in initial caries. In, (2008) Second international symposium on intelligent information technology application. IEEE. https://doi.org/10.1109/iita.2008.206
Kumar P, Srivastava MM (2018) Example mining for incremental learning in medical imaging. In: IEEE Symposium Series on Computational Intelligence (SSCI). https://doi.org/10.1109/SSCI.2018.8628895
Lee JH, Kim DH, Jeong SN et al (2018) Detection and diagnosis of dental caries using a deep learning-based convolutional neural network algorithm. J Dent 77:106–111. https://doi.org/10.1016/j.jdent.2018.07.015
Lee S, Oh S, Jo J, et al (2021) Deep learning for early dental caries detection in bitewing radiographs. Sci Reports 11(1). https://doi.org/10.1038/s41598-021-96368-7
Lian L, Zhu T, Zhu F et al (2021) Deep learning for caries detection and classification. Diagnostics 11(9):1672. https://doi.org/10.3390/diagnostics11091672
Lin TY, Goyal P, Girshick R, et al (2017) Focal loss for dense object detection. In: International conference on computer vision (ICCV), pp 2999–300. https://doi.org/10.1109/ICCV.2017.324
Liu Z, Lin Y, Cao Y, et al (2021) Swin transformer: hierarchical vision transformer using shifted windows. In: IEEE International conference on computer vision (ICCV), pp 10012–10022. https://doi.org/10.1109/ICCV48922.2021.00986
Loshchilov I, Hutter F (2017) SGDR: stochastic gradient descent with warm restarts. In: International conference on learning representations (ICLR). https://doi.org/10.48550/arXiv.1608.03983
Loshchilov I, Hutter F (2019) Decoupled weight decay regularization. In: International conference on learning representations (ICLR). https://doi.org/10.48550/ARXIV.1711.05101
Mao YC, Chen TY, Chou HS et al (2021) Caries and restoration detection using bitewing film based on transfer learning with CNNs. Sensors 21. https://doi.org/10.3390/s21134613
Mohammad-Rahimi H, Motamedian SR, Rohban MH et al (2022) Deep learning for caries detection: a systematic review. J Dent 122. https://doi.org/10.1016/j.jdent.2022.104115
Moran M, Faria M, Giraldi G et al (2021) Classification of approximal caries in bitewing radiographs using convolutional neural networks. Sensors 21(15):5192. https://doi.org/10.3390/s21155192
Padilla R, Passos WL, Dias TLB et al (2021) A comparative analysis of object detection metrics with a companion open-source toolkit. Electronics 10(3):279. https://doi.org/10.3390/electronics10030279
Panyarak W, Suttapak W, Wantanajittikul K et al (2023) Assessment of YOLOv3 for caries detection in bitewing radiographs based on the ICCMS radiographic scoring system. Clin Oral Investig 27:1731–1742. https://doi.org/10.1007/s00784-022-04801-6
Panyarak W, Wantanajittikul K, Suttapak W et al (2023) Feasibility of deep learning for dental caries classification in bitewing radiographs based on the ICCMS radiographic scoring system. Oral Surg Oral Med Oral Pathol Oral Radiol 135(2):272–281. https://doi.org/10.1016/j.oooo.2022.06.012
Prados-Privado M, Villalón JG, Martínez-Martínez CH et al (2020) Dental caries diagnosis and detection using neural networks: a systematic review. J Clin Med 9(11):3579. https://doi.org/10.3390/jcm9113579
Ren S, He K, Girshick RB, et al (2015) Faster R-CNN: towards real-time object detection with region proposal networks. In: Neural information processing systems (NIPS). https://doi.org/10.48550/arXiv.1506.01497
Rodriguez-Ruiz A, Lång K, Gubern-Merida A, et al (2019) Stand-alone artificial intelligence for breast cancer detection in mammography: comparison with 101 radiologists. JNCI: J National Cancer Inst 111(9):916–922. https://doi.org/10.1093/jnci/djy222
Solovyev R, Wang W, Gabruseva T (2019) Weighted boxes fusion: ensembling boxes from different object detection models. Image Vis Comput. https://doi.org/10.48550/ARXIV.1910.13302
Srivastava MM, Kumar P, Pradhan L, et al (2017) Detection of tooth caries in bitewing radiographs using deep learning. In: NIPS workshop on machine learning for health. https://doi.org/10.48550/arXiv.1711.07312
Tan M, Pang R, Le QV (2020) EfficientDet: scalable and efficient object detection. In: Computer vision and pattern recognition conference (CVPR). https://doi.org/10.48550/arXiv.1711.07312
Tichý A, Kunt L, Kybic J (2023a) Dental caries in bitewing radiographs. Mendeley Data. https://doi.org/10.17632/4fbdxs7s7w.1
Tichý A, Kunt L, Nagyová V, et al (2023b) Automatic caries detection in bitewing radiographs. part II: Experimental comparison. Clin Oral Investig. https://doi.org/10.1007/s00784-023-05335-1
Wang CW, Huang CT, Lee JH et al (2016) A benchmark for comparison of dental radiography analysis algorithms. Med Image Anal 31:63–76. https://doi.org/10.1016/j.media.2016.02.004
Yasa Y, Çelik O, Bayrakdar IS et al (2020) An artificial intelligence proposal to automatic teeth detection and numbering in dental bite-wing radiographs. Acta Odontol Scand 79(4):275–281. https://doi.org/10.1080/00016357.2020.1840624
Zhou H, Li Z, Ning C, et al (2017) CAD: scale invariant framework for real-time object detection. In: 2017 EEE International conference on computer vision workshops (ICCVW). https://doi.org/10.1109/iccvw.2017.95
This work was supported by the General University Hospital in Prague (project GIP-21-SL-01-232) and by the OP VVV funded project CZ.02.1.01/0.0/0.0/16_019/0000765 “Research Center for Informatics”. The study sponsors had no involvement in the study design, analysis, interpretation of the data, writing, or choosing the publication venue.
This research was approved by the Ethics Committee of the General University Hospital in Prague, protocol number 82/21. The patients signed a written informed consent, agreeing with the use of their data in anonymized form for research purposes.
Conflict of interest
The authors declare no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
About this article
Cite this article
Kunt, L., Kybic, J., Nagyová, V. et al. Automatic caries detection in bitewing radiographs: part I—deep learning. Clin Oral Invest (2023). https://doi.org/10.1007/s00784-023-05335-1