A Deep Learning Approach to Segment and Classify C-Shaped Canal Morphologies in Mandibular Second Molars Using Cone-beam Computed Tomography

Published:September 23, 2021DOI:



      The identification of C-shaped root canal anatomy on radiographic images affects clinical decision making and treatment. The aims of this study were to develop a deep learning (DL) model to classify C-shaped canal anatomy in mandibular second molars from cone-beam computed tomographic (CBCT) volumes and to compare the performance of 3 different architectures.


      U-Net, residual U-Net, and Xception U-Net architectures were used for image segmentation and classification of C-shaped anatomies. Model training and validation were performed on 100 of a total of 135 available limited field of view CBCT images containing mandibular molars with C-shaped anatomy. Thirty-five CBCT images were used for testing. Voxel-matching accuracy of the automated labeling of the C-shaped anatomy was assessed with the Dice index. The mean sensitivity of predicting the correct C-shape subcategory was calculated based on detection accuracy. One-way analysis of variance and post hoc Tukey honestly significant difference tests were used for statistical evaluation.


      The mean Dice coefficients were 0.768 ± 0.0349 for Xception U-Net, 0.736 ± 0.0297 for residual U-Net, and 0.660 ± 0.0354 for U-Net on the test data set. The performance of the 3 models was significantly different overall (analysis of variance, P = .000779). Both Xception U-Net (Q = 7.23, P = .00070) and residual U-Net (Q = 5.09, P = .00951) performed significantly better than U-Net (post hoc Tukey honestly significant difference test). The mean sensitivity values were 0.786 ± 0.0378 for Xception U-Net, 0.746 ± 0.0391 for residual U-Net, and 0.720 ± 0.0495 for U-Net. The mean positive predictive values were 77.6% ± 0.1998% for U-Net, 78.2% ± 0.0.1971% for residual U-Net, and 80.0% ± 0.1098% for Xception U-Net. The addition of contrast-limited adaptive histogram equalization had improved overall architecture efficacy by a mean of 4.6% (P < .0001).


      DL may aid in the detection and classification of C-shaped canal anatomy.

      Key Words

      To read this article in full you will need to make a payment

      Purchase one-time access:

      Academic & Personal: 24 hour online accessCorporate R&D Professionals: 24 hour online access
      One-time access price info
      • For academic or personal research use, select 'Academic and Personal'
      • For corporate R&D use, select 'Corporate R&D Professionals'


      Subscribe to Journal of Endodontics
      Already a print subscriber? Claim online access
      Already an online subscriber? Sign in
      Institutional Access: Sign in to ScienceDirect


        • Schwendicke F.
        • Samek W.
        • Krois J.
        Artificial intelligence in dentistry: chances and challenges.
        J Dent Res. 2020; 99: 769-774
        • Orhan K.
        • Bayrakdar I.S.
        • Ezhov M.
        • et al.
        Evaluation of artificial intelligence for detecting periapical pathosis on cone-beam computed tomography scans.
        Int Endod J. 2020; 53: 680-689
        • Setzer F.C.
        • Shi K.J.
        • Zhang Z.
        • et al.
        Artificial intelligence for the computer-aided detection of periapical lesions in cone-beam computed tomographic images.
        J Endod. 2020; 46: 987-993
        • Endres M.G.
        • Hillen F.
        • Salloumis M.
        • et al.
        Development of a deep learning algorithm for periapical disease detection in dental radiographs.
        Diagnostics (Basel). 2020; 10: 430
        • Zheng Z.
        • Yan H.
        • Setzer F.C.
        • et al.
        Anatomically constrained deep learning for automating dental CBCT segmentation and lesion detection.
        IEEE Trans Autom Sci Eng. 2021; 18: 603-614
        • Pauwels R.
        • Brasil D.M.
        • Yamasaki M.C.
        • et al.
        Artificial intelligence for detection of periapical lesions on intraoral radiographs: comparison between convolutional neural networks and human observers.
        Oral Surg Oral Med Oral Pathol Oral Radiol. 2021; 131: 610-616
        • Ekert T.
        • Krois J.
        • Meinhold L.
        • et al.
        Deep learning for the radiographic detection of apical lesions.
        J Endod. 2019; 45: 917-922.e5
        • Lee K.S.
        • Kwak H.J.
        • Oh J.M.
        • et al.
        Automated detection of TMJ osteoarthritis based on artificial intelligence.
        J Dent Res. 2020; 99: 1363-1367
        • Fukuda M.
        • Inamoto K.
        • Shibata N.
        • et al.
        Evaluation of an artificial intelligence system for detecting vertical root fracture on panoramic radiography.
        Oral Radiol. 2020; 36: 337-343
        • Johari M.
        • Esmaeili F.
        • Andalib A.
        • et al.
        Detection of vertical root fractures in intact and endodontically treated premolar teeth by designing a probabilistic neural network: an ex vivo study.
        Dentomaxillofac Radiol. 2017; 46: 20160107
        • Shah H.
        • Hernandez P.
        • Budin F.
        • et al.
        Automatic quantification framework to detect cracks in teeth.
        Proc SPIE Int Soc Opt Eng. 2018; 10578: 105781K
        • Vicory J.
        • Chandradevan R.
        • Hernandez-Cerdan P.
        • et al.
        Dental microfracture detection using wavelet features and machine learning.
        in: Isgum I. Landman B.A. Medical Imaging 2021: Image Processing. International Society for Optics and Photonics, Washington, DC2021: 115961R
        • Saghiri M.A.
        • Asgar K.
        • Boukani K.K.
        • et al.
        A new approach for locating the minor apical foramen using an artificial neural network.
        Int Endod J. 2012; 45: 257-265
        • Saghiri M.A.
        • Garcia-Godoy F.
        • Gutmann J.L.
        • et al.
        The reliability of artificial neural network in locating minor apical foramen: a cadaver study.
        J Endod. 2012; 38: 1130-1134
        • Campo L.
        • Aliaga I.J.
        • De Paz J.F.
        • et al.
        Retreatment predictions in odontology by means of CBR systems.
        Comput Intell Neurosci. 2016; 2016: 7485250
        • Leite A.F.
        • Gerven A.V.
        • Willems H.
        • et al.
        Artificial intelligence-driven novel tool for tooth detection and segmentation on panoramic radiographs.
        Clin Oral Investig. 2021; 25: 2257-2267
        • Hiraiwa T.
        • Ariji Y.
        • Fukuda M.
        • et al.
        A deep-learning artificial intelligence system for assessment of root morphology of the mandibular first molar on panoramic radiography.
        Dentomaxillofac Radiol. 2019; 48: 20180218
        • Kwak G.H.
        • Kwak E.J.
        • Song J.M.
        • et al.
        Automatic mandibular canal detection using a deep convolutional neural network.
        Sci Rep. 2020; 10: 5711
        • Li Q.
        • Chen K.
        • Han L.
        • et al.
        Automatic tooth roots segmentation of cone beam computed tomography image sequences using U-net and RNN.
        J Xray Sci Technol. 2020; 28: 905-922
        • Lahoud P.
        • EzEldeen M.
        • Beznik T.
        • et al.
        Artificial intelligence for fast and accurate 3-dimensional tooth segmentation on cone-beam computed tomography.
        J Endod. 2021; 47: 827-835
        • Sherwood I.A.
        Pre-operative diagnostic radiograph interpretation by general dental practitioners for root canal treatment.
        Dentomaxillofac Radiol. 2012; 41: 43-54
        • Sherwood A.
        • Gutmann J.L.
        • Kumar S.
        • et al.
        CBCT analysis of the anatomy of C-shaped root canals in mandibular second molars from a southern Indian population in Tamil Nadu.
        Endod Pract Today. 2019; 13: 61-70
        • Fernandes M.
        • de Ataide I.
        • Wagle R.
        C-shaped root canal configuration: a review of literature.
        J Conserv Dent. 2014; 17: 312-319
        • Fan B.
        • Cheung G.S.
        • Fan M.
        • et al.
        C-shaped canal system in mandibular second molars: part II--radiographic features.
        J Endod. 2004; 30: 904-908
        • Jeon S.J.
        • Yun J.P.
        • Yeom H.G.
        • et al.
        Deep-learning for predicting C-shaped canals in mandibular second molars on panoramic radiographs.
        Dentomaxillofac Radiol. 2021; 50: 20200513
        • Yushkevich P.A.
        • Piven J.
        • Hazlett H.C.
        • et al.
        User-guided 3D active contour segmentation of anatomical structures: significantly improved efficiency and reliability.
        Neuroimage. 2006; 31: 1116-1128
        • Schloss T.
        • Sonntag D.
        • Kohli M.R.
        • Setzer F.C.
        A comparison of 2- and 3-dimensional healing assessment after endodontic surgery using cone-beam computed tomographic volumes or periapical radiographs.
        J Endod. 2017; 43: 1072-1079
        • Buslaev A.
        • Iglovikov V.I.
        • Khvedchenya E.
        • et al.
        Albumentations: fast and flexible image augmentations.
        Information. 2020; 11: 125
      1. Itseez. Open-source computer vision library.
        (Available at:) (Accessed June 1, 2021)
        • Ronneberger O.
        • Fischer P.
        • Brox T.
        U-net: convolutional networks for biomedical image segmentation.
        in: International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer, New York2015: 234-241
        • Khanna A.
        • Narendra
        • Londhe D.
        • et al.
        A deep residual U-net convolutional neural network for automated lung segmentation in computed tomography images.
        Biocybern Biomed Eng. 2020; 40: 1314-1327
        • Chollet F.
        Xception: Deep Learning with Depthwise Separable Convolutions.
        in: 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). IEEE, Honolulu, HI2017: 1800-1807
        • Kaiming H.
        • Xiangyu Z.
        • Shaoqing R.
        • Jian S.
        Deep residual learning for image recognition.
        2015 (Available at:)
        Date accessed: May 27, 2021
        • Paszke A.
        • Gross S.
        • Massa F.
        • et al.
        PyTorch: an imperative style, high-performance deep learning library.
        in: Wallach H. Larochelle H. Beygelzimer A. Advances in Neural information Processing Systems. Curran Associates, Inc, Red Hook, NY2019: 8026-8037
        • Wightman R.
        PyTorch image models. Asia-Pacific Signal and Information Processing Association (APSIPA). Online Conference Proceedings APSIPA Annual Summit and Conference (ASC). 2010 APSIPA Annual Summit and Conference, Biopolis, Singapore, December 14-17, 2010.
        (Available at:) (Accessed May 27, 2021)
        • Leonardi Dutra K.
        • Haas L.
        • Porporatti A.L.
        • et al.
        Diagnostic accuracy of cone-beam computed tomography and conventional radiography on apical periodontitis: a systematic review and meta- analysis.
        J Endod. 2016; 42: 356-364
        • Setzer F.C.
        • Hinckley N.
        • Kohli M.R.
        • Karabucak B.
        A survey of CBCT use amongst endodontic practitioners in the United States.
        J Endod. 2017; 43: 699-704
        • Special Committee to Revise the Joint AAE/AAOMR Position Statement on use of CBCT in Endodontics
        AAE and AAOMR joint position statement: use of cone beam computed tomography in endodontics 2015 update.
        J Endod. 2015; 41: 1393-1396
        • Kida S.
        • Nakamoto T.
        • Nakano M.
        • et al.
        Cone beam computed tomography image quality improvement using a deep convolutional neural network.
        Cureus. 2018; 10: e2548
        • Duan W.
        • Chen Y.
        • Zhang Q.
        • et al.
        Refined tooth and pulp segmentation using U-Net in CBCT image.
        Dentomaxillofac Radiol. 2021; 50: 20200251
        • Shan T.
        • Tay F.R.
        • Gu L.
        Application of artificial intelligence in dentistry.
        J Dent Res. 2021; 100: 232-244
        • Abdolali F.
        • Zoroofi R.A.
        • Otake Y.
        • Sato Y.
        Automatic segmentation of maxillofacial cysts in cone beam CT images.
        Comput Biol Med. 2016; 72: 108-119
        • Li S.
        • Fevens T.
        • Krzyzak A.
        • et al.
        Toward automatic computer-aided dental X-ray analysis using level set method.
        Med Image Comput Comput Assist Interv. 2005; 8: 670-678
        • Okada K.
        • Rysavy S.
        • Flores A.
        • Linguraru M.G.
        Noninvasive differential diagnosis of dental periapical lesions in cone-beam CT scans.
        Med Phys. 2015; 42: 1653-1665
        • Hiew L.T.
        • Ong S.H.
        • Foong K.W.
        • Weng C.
        Tooth segmentation from cone-beam CT using graph cut.
        in: Proceedings of the Second APSIPA Annual Summit and Conference. 2010: 272-275
        • Yang L.
        • Tonghe W.
        • Joseph H.
        • et al.
        Image quality improvement in cone-beam CT using deep learning.
        in: Proceedings of SPIE-The International Society for Optical Engineering. SPIE Medical Imaging, San Diego, California2019
        • Yang H.
        • Liang K.
        • Zhang L.
        • et al.
        Improve 3D cone-beam CT reconstruction by slice-wise deep learning.
        in: 2018 IEEE Nuclear Science Symposium and Medical Imaging Conference, Sydney, Australia. Institute of Electrical & Electronics Engineers, 2018: 1-3
        • Pei Y.
        • Ai X.
        • Zha H.
        • et al.
        3D exemplar-based random walks for tooth segmentation from cone-beam computed tomography images.
        Med Phys. 2016; 43: 5040