Abstract
In this paper, we address the problem of automatic three-dimensional cephalometric analysis. Cephalometric analysis performed on lateral radiographs doesn’t fully exploit the structure of 3D objects due to projection onto the lateral plane. With the development of three-dimensional imaging techniques such as CT, several analysis methods have been proposed that extend to the 3D case. The analysis based on these methods is invariant to rotations and translations and can describe difficult skull deformation, where 2D cephalometry has no use. In this paper, we provide a wide overview of existing approaches for cephalometric landmark regression. Moreover, we perform a series of experiments with state of the art 3D convolutional neural network (CNN) based methods for keypoint regression: direct regression with CNN, heatmap regression and Softargmax regression. For the first time, we extensively evaluate the described methods and demonstrate their effectiveness in the estimation of Frankfort Horizontal and cephalometric points locations for patients with severe skull deformations. We demonstrate that Heatmap and Softargmax regression models provide sufficient regression error for medical applications (less than 4 mm). Moreover, the Softargmax model achieves 1.15° inclination error for the Frankfort horizontal. For the fair comparison with the prior art, we also report results projected on the lateral plane.
Similar content being viewed by others
REFERENCES
A. Jacobson (ed.), Radiographic Cephalometry: From Basics to Videoimaging (Quintessence Publishing, Chicago, 1995).
A. Katsumata, M. Fujishita, M. Maeda, Y. Ariji, E. Ariji, and R. P. Langlais, “3D-CT evaluation of facial asymmetry,” Oral Surg. Oral Med. Oral Pathol. Oral Radiol. Endodontol. 99 (2), 212–220 (2005).
M. G. Cavalcanti, J. W. Haller, and M. W. Vannier, “Three-dimensional computed tomography landmark measurement in craniofacial surgical planning: Experimental validation in vitro,” J. Oral Maxillofac. Surg. 57 (6), 690–694 (1999).
R. Olszewski, F. Zech, G. Cosnard, V. Nicolas, B. Macq, and H. Reychler, “Three-dimensional computed tomography cephalometric craniofacial analysis: Experimental validation in vitro,” Int. J. Oral Maxillofac. Surg. 36 (9), 828–833 (2007).
M. J. Troulis, P. Everett, E. B. Seldin, R. Kikinis, and L. B. Kaban, “Development of a three-dimensional treatment planning system based on computed tomographic data,” Int. J. Oral Maxillofac. Surg. 31, 349–357 (2002).
T. Smektała, M. Jędrzejewski, J. Szyndel, K. Sporniak-Tutak, and R. Olszewski, “Experimental and clinical assessment of three-dimensional cephalometry: A systematic review,” J. Cranio-Maxillofac. Surg. 42 (8), 1795–1801 (2014).
J. Gateno, J. J. Xia, and J. F. Teichgraeber, “New 3-dimensional cephalometric analysis for orthognathic surgery,” J. Oral Maxillofac. Surg. 69 (3), 606–622 (2011).
H.-L. Kim, B. C. Kim, J.-G. Kim, P. Zhengguo, S. H. Kang, and S.-H. Lee, “Construction and validation of the midsagittal reference plane based on the skull base symmetry for three-dimensional cephalometric craniofacial analysis,” J. Craniofac. Surg. 25 (2), 338–342 (2014).
M.-H. Chen, J. Z.-C. Chang, S.-H. Kok, Y.-J. Chen, Y.-D. Huang, K.-Y. Cheng, and C.-P. Lin, “Intraobserver reliability of landmark identification in cone-beam computed tomography-synthesized two-dimensional cephalograms versus conventional cephalometric radiography: A preliminary study,” J. Dent. Sci. 9 (1), 56–62 (2014).
M. B. da Neiva, Á. C. Soares, C. de Oliveira Lisboa, O. de Vasconcellos Vilella, and A. T. Motta, “Evaluation of cephalometric landmark identification on CBCT multiplanar and 3D reconstructions,” Angle Orthod. 85 (1), 11–17 (2015).
D. Grauer, L. S. H. Cevidanes, M. A. Styner, I. Heulfe, E. T. Harmon, H. Zhu, and W. R. Proffit, “Accuracy and landmark error calculation using cone-beam computed tomography-generated cephalograms,” Angle Orthod. 80 (2), 286–294 (2010).
J. J. Xia, J. Gateno, J. F. Teichgraeber, P. Yuan, J. Li, K.-C. Chen, A. Jajoo, M. Nicol, and D. M. Alfi, “Algorithm for planning a double-jaw orthognathic surgery using a computer-aided surgical simulation (CASS) protocol. Part 2: Three-dimensional cephalometry,” Int. J. Oral Maxillofac. Surg. 44 (12), 1441–1450 (2015).
I. Ono, T. Ohura, E. Narumi, K. Kawashima, I. Matsuno, S. Nakamura, N. Ohhata, Y. Uchiyama, Y. Watanabe, F. Tanaka, and T. Kishinami, “Three-dimensional analysis of craniofacial bones using three-dimensional computer tomography,” J. Cranio-Maxillofac. Surg. 20 (2), 49–60 (1992).
I. Hayashi, “Morphological relationship between the cranial base and dentofacial complex obtained by reconstructive computer tomographic images,” Eur. J. Orthod. 25 (4), 385–391 (2003).
B. B. Tuncer, M. S. Ataç, and S. Yüksel, “A case report comparing 3-D evaluation in the diagnosis and treatment planning of hemimandibular hyperplasia with conventional radiography,” J. Cranio-Maxillofac. Surg. 37 (6), 312–319 (2009).
D. Zhang, S. Wang, J. Li, and Y. Zhou, “Novel method of constructing a stable reference frame for 3-dimensional cephalometric analysis,” Am. J. Orthod. Dentofac. Orthop. 154 (3), 397–404 (2018).
O. J. C. van Vlijmen, T. Maal, S. J. Bergé, E. M. Bron-khorst, C. Katsaros, and A. M. Kuijpers-Jagtman, “A comparison between 2D and 3D cephalometry on CBCT scans of human skulls,” Int. J. Oral Maxillofac. Surg. 39 (2), 156–160 (2010).
R. Olszewski, O. Tanesy, G. Cosnard, F. Zech, and H. Reychler, “Reproducibility of osseous landmarks used for computed tomography based three-dimensional cephalometric analyses,” J. Cranio-Maxillofac. Surg. 38 (3), 214–221 (2010).
R. Olszewski, G. Cosnard, B. Macq, P. Mahy, and H. Reychler, “3D CT-based cephalometric analysis: 3D cephalometric theoretical concept and software,” Neuroradiology 48 (11), 853–862 (2006).
G. R. J. Swennen, F. Schutyser, and J.-E. Hausamen, Three-Dimensional Cephalometry. A Color Atlas and Manual (Springer, Berlin, Heidelberg, 2006).
G. J. Edwards, C. J. Taylor, and T. F. Cootes, “Interpreting face images using active appearance models,” in Proc. Third IEEE Int. Conf. on Automatic Face and Gesture Recognition (Nara, Japan, 1998), pp. 300–305.
Y. J. Chen, S. K. Chen, H. F. Chang, and K. C. Chen, “Comparison of landmark identification in traditional versus computer-aided digital cephalometry,” Angle Orthod. 70 (5), 387–392 (2000).
I. El-Feghi, M. A. Sid-Ahmed, and M. Ahmadi, “Automatic localization of craniofacial landmarks for assisted cephalometry,” in Proc. 2003 Int. Symposium on Circuits and Systems (ISCAS 2003) (Bangkok, Thailand, 2003), IEEE, Vol. 3, pp. III-630–III-633; Pattern Recogn. 37 (3), 609–621 (2004).
A. A. Pouyan and M. Farshbaf, “Cephalometric landmarks localization based on histograms of oriented gradients,” in Proc. 2010 Int. Conf. on Signal and Image Processing (ICSIP 2010) (Chennai, India, 2010), IEEE, pp. 1–6.
M. Dantone, J. Gall, G. Fanelli, and L. Van Gool, “Real-time facial feature detection using conditional regression forests,” in Proc. 2012 IEEE Conference on Computer Vision and Pattern Recognition (Providence, RI, 2012), pp. 2578–2585.
M. Osadchy, Y. Le Cun, and M. L. Miller, “Synergistic face detection and pose estimation with energy-based models,” in Toward Category-Level Object Recognition, Ed. by J. Ponce, M. Hebert, C. Schmid, and A. Zisserman, Lecture Notes in Computer Science (Springer, Berlin, Heidelberg, 2006), Vol. 4170, pp. 196–206.
J. Tompson, R. Goroshin, A. Jain, Y. LeCun, and C. Bregler, “Efficient object localization using Convolutional Networks,” in Proc. 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (Boston, MA, 2015), pp. 648–656.
A. Newell, K. Yang, and J. Deng, “Stacked hourglass networks for human pose estimation,” in Computer Vision–ECCV 2016, Proc. 14th European Conf., Part VIII, Ed. by B. Leibe et al., Lecture Notes in Computer Science (Springer, Cham, 2016), Vol. 9912, pp. 483–499.
A. Nibali, Z. He, S. Morgan, and L. Prendergast, “Numerical coordinate regression with convolutional neural networks,” arXiv preprint arXiv:1801.07372 (2018). https://arxiv.org/abs/1801.07372
W. Yue, D. Yin, C. Li, G. Wang, and T. Xu, “Automated 2-D cephalometric analysis on X-ray images by a model-based approach,” IEEE Trans. Biomed. Eng. 53 (8), 1615– 1623 (2006).
D. A. Lachinov, A. A. Getmanskaya, and V. E. Turlapov, “Refinement of the coherent point drift registration results by the example of cephalometry problems,” Program. Comput. Software 44 (4), 248–257 (2018).
A. Myronenko and X. B. Song, “Point set registration: Coherent point drift,” IEEE Trans. Pattern Anal. Mach. Intell. 32 (12), 2262–2275 (2010).
C. Lee, C. Tanikawa, J.-Y. Lim, and T. Yamashiro, “Deep learning based cephalometric landmark identification using landmark-dependent multi-scale patches,” arXiv preprint arXiv:1906.02961 (2019). https://arxiv.org/abs/1906.02961
R. Chen, Y. Ma, N. Chen, D. Lee, and W. Wang, “Cephalometric landmark detection by attentive feature pyramid fusion and regression-voting,” in Medical Image Computing and Computer Assisted Intervention–MICCAI 2019, Ed. by D. Shen, T. Liu, T. M. Peters, L. H. Staib, C. Essert, S. Zhou, P.-T. Yap, and A. Khan, Lecture Notes in Computer Science (Springer, Cham, 2019), Vol. 11766, pp. 873–881.
H.-W. Hwang, J.-H. Park, J.-H. Moon, Y. Yu, H. Kim, S.-B. Her, G. Srinivasan, M. N. A. Aljanabi, R. E. Donatelli, and S.-J. Lee, “Automated identification of cephalometric landmarks: Part 2–Might it be better than human?” Angle Orthod. 90 (1), 69–76 (2020).
J. Redmon and A. Farhadi, “YOLOv3: An incremental improvement,” arXiv preprint arXiv:1804.02767 (2018). https://arxiv.org/abs/1804.02767
S. M. Lee, H. P. Kim, K. Jeon, S.-H. Lee, and J. K. Seo, “Automatic 3D cephalometric annotation system using shadowed 2D image-based machine learning,” Phys. Med. Biol. 64 (5), 055002 (2019).
K. Simonyan and A. Zisserman, “Very deep convolutional networks for large-scale image recognition,” arXiv preprint arXiv:1409.1556 (2014). https://arxiv.org/abs/1409.1556
S. H. Kang, K. Jeon, H.-J. Kim, J. K. Seo, and S.‑H. Lee, “Automatic three-dimensional cephalometric annotation system using three-dimensional convolutional neural networks: A developmental trial,” Comput. Methods Biomech. Biomed. Eng.: Imaging & Visualization 8 (2), 210–218 (2020).
D. Ulyanov, A. Vedaldi, and V. Lempitsky, “Instance normalization: The missing ingredient for fast stylization,” arXiv preprint arXiv:1607.08022 (2016).
S. Xie, R. Girshick, P. Dollár, Z. Tu, and K. He, “Aggregated residual transformations for deep neural networks,” in Proc. 2017 30th IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2017) (Honolulu, HI, 2017), pp. 5987–5995.
Y. Wu and K. He, “Group normalization,” in Computer Vision−ECCV 2018, Proc. 15th European Conference, Part XIII, Ed. by V. Ferrari, M. Hebert, C. Sminchisescu, and Y. Weiss, Lecture Notes in Computer Science (Springer, Cham, 2018), Vol. 11217, pp. 3–19; Int. J. Comput. Vision 128 (3), 742–755 (2020).
D. C. Luvizon, H. Tabia, and D. Picard, “Human pose regression by combining indirect part detection and contextual information,” Comput. Graphics 85, 15–22 (2017).
C. Goch, J. Metzger, and M. Nolden, “Tutorial: Medical image processing with MITK introduction and new developments,” in Bildverarbeitung für die Medizin 2017, Ed. by K. H. Majer-Hein, T. M. Deserno, H. Handels, and T. Tolxdorff, Informatik aktuell. Springer, Berlin, Heidelberg, 2017), p. 10. http://mitk.org/wiki/MITK
A. Paszke, S. Gross, F. Massa, A. Lerer, J. Bradbury, G. Chanan, T. Killeen, Z. Lin, N. Gimelshein, L. Antiga, A. Desmaison, A. Köpf, E. Yang, Z. DeVito, M. Raison, A. Tejani, S. Chilamkurthy, B. Steiner, L. Fang, J. Bai, and S. Chintala, “PyTorch: An imperative style, high-performance deep learning library,” in Advances in Neural Information Processing Systems 32: Proc. NIPS 2019 Conf. (Vancouver, Canada, 2019), pp. 8024–8035.
ACKNOWLEDGMENTS
This work was supported by the Russian Foundation for Basic Research, project no. 18-37-00383.
Author information
Authors and Affiliations
Corresponding authors
Ethics declarations
The authors declare they have no conflicts of interest.
Additional information
Dmitrii Lachinov. Born in 1993. MSc degree received from Lobachevsky State University, Institute of Information Technologies, Mathematics, and Mechanics in 2018. In the 2018–2019 academic year, he was a graduate student at Lobachevsky University Currently pursuing a PhD degree at the Medical University of Vienna, Department of Ophthalmology and Optometry. Research Assistant at the Medical University of Vienna. Area of research: medical imaging and image processing.
Aleksandra Getmanskaya. Born in 1986. MSc degree received from Lobachevsky State University, Institute of Information Technologies, Mathematics, and Mechanics in 2009. Currently is a researcher at Lobachevsky State University, Institute of Information Technologies, Mathematics and Mechanics. Area of research: medical imaging and image processing.
Vadim Turlapov. Born in 1949. In 1994, he received a PhD degree, and in 2002, Doctor of Science, in the field of engineering geometry and computer graphics. Currently is a professor of Lobachevsky State University, head of Computer Graphics Lab in the Institute of Information Technologies, Mathematics, and Mechanics. Area of research: scientific visualization, medical imaging, image processing, hyperspectral imaging, global illumination, photorealistic image synthesis.
Rights and permissions
About this article
Cite this article
Lachinov, D., Getmanskaya, A. & Turlapov, V. Cephalometric Landmark Regression with Convolutional Neural Networks on 3D Computed Tomography Data. Pattern Recognit. Image Anal. 30, 512–522 (2020). https://doi.org/10.1134/S1054661820030165
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1134/S1054661820030165