Int J Stomatol

Int J Stomatol ›› 2022, Vol. 49 ›› Issue (1): 100-108.doi: 10.7518/gjkq.2022006

• Orginal Article • Previous Articles     Next Articles

Research progress on two-dimensional and three-dimensional cephalometric automatic landmarking

Liu Lijia1(),Mao Jing1,Long Huan1,Pu Yalong1,Wang Jun2()   

  1. 1. State Key Laboratory of Oral Diseases & National Clinical Research Center for Oral Diseases & West China School of Stomatology, Sichuan University, Chengdu 610041, China;
    2. State Key Laboratory of Oral Diseases & National Clinical Research Center for Oral Diseases & Dept. of Orthodontics, West China Hospital of Stomatology, Sichuan University, Chengdu 610041, China;
  • Received:2021-04-04 Revised:2021-09-01 Online:2022-01-01 Published:2022-01-07
  • Contact: Jun Wang E-mail:llj2186@163.com;wangjunv@scu.edu.cn
  • Supported by:
    This study was supported by National Natural Science Foundation of China(81771114);This study was supported by National Natural Science Foundation of China(81970967);Chengdu Science and Technology Program(2019-YF05-00349-SN)

RichHTML

22

PDF (PC)

1377

Abstract:

Cephalometry is an indispensable measurement of orthodontic and orthognathic diagnosis and treatment. With the development of computer-aided technology, automatic landmarking of cephalometry has been achieved in two-dimensional cephalometry with high precision, which greatly reduces the workload of operators. However, without geome-tric distortion, tissue overlap, and other defects, the cone beam computed tomography images can accurately locate the anatomical landmarks of cephalometric analysis. This analysis is helpful for the diagnosis and analysis of congenital or deve-lopmental craniomaxillofacial asymmetry. Nowadays, the automatic landmarking of three-dimensional cephalometry is an important research direction in the field of cephalometry. Based on the classification of different automatic fixed-point methods, this study reviews the progress of two- and three-dimensional cephalometric automatic landmarking. Furthermore, the study discusses the accuracy of different automatic landmarking methods and proposes future research directions.

Key words: cephalometry, X-ray, three-dimensional imaging, automatic landmarking, cone beam computed tomography


TrendMD: 

Fig 1

Schematic diagram of the basic principle of CNN in 2D cephalometric measurement"

Tab 1

Two-dimensional cephalometric measurement automatically fixed-point collection article abstract"

定点方法 参考文献 样本信息 样本量 标志点
数量		 自动定点误差
图像过滤与基于知识的标志点
搜索		 [11] 2名12岁和10岁少女的头颅X线片 测量样本：2幅X线片 36 成功定位23个标志点（具体误差不详）
[12] — 测量样本：5幅X线片 27 9个标志点平均误差2.06 mm（18%<1 mm,58%<2 mm）
[13] — 训练样本：3幅X线片
测量样本：10幅X线片		 19 平均误差：3.0 mm
[14] — 测量样本：10幅X线片 19 63%<1 mm,74%<2 mm
基于模型的方法 [17] 不同年龄、种族群体和错??畸形的不同质量的X线片 训练样本：63幅X线片
测量样本：63幅X线片（与训练样本相同）		 16 平均误差：（4.08±3.74）mm（35%<2 mm,13%<1 mm）
[18] 不同年龄、种族群体和错??畸形的不同质量的X线片 训练样本：96幅X线片
测量样本：96幅X线片（与训练样本相同）		 43 平均误差：（2.48±1.66）mm
[19] 年龄7.2~25.6岁进行正畸评估和治疗的患者 训练样本：60幅X线片
测量样本：60幅X线片（与训练样本相同）		 17 平均误差：1.68 mm
[21] 不分年龄、性别、种族 训练样本：134幅X线片
测量样本：134幅X线片（与训练样本相同）		 18 平均误差为（1.84±1.24）mm
基于学习的方法 [10] 年龄7~76岁（平均27.0岁） 测量样本：400幅X线片 19 平均误差：1.2 mm
[22] — 训练样本：150幅X线片
测量样本：250幅X线片		 19 75.58%标志点平均误差<2 mm
[23] — 训练样本：1 731幅X线片
测量样本：61幅X线片		 8个角度
3个距离
1个比值		 平均误差：0.185°
平均误差：0.147 mm
平均误差：0.25%
[24] 包括具有固定矫治器、义齿等的患者图像;排除质量极差的图像 训练样本：1 028幅X线片
测量样本：283幅X线片		 80 38个标志点：YOLOv3准确性>SSD
40个标志点无明显差异
[25] 包括具有固定矫治器、义齿等的患者图像;排除质量极差的图像 训练样本：1 028幅X线片
测量样本：283幅X线片		 80 YOLOv3平均误差：（1.46±2.97）mm
专家人工测量平均差值：（1.50±1.48）mm
混合方法 [27] 年龄6~60岁 训练样本：100幅X线片
测量样本：100幅X线片		 19 平均误差：（1.82±1.61）mm
（平均径向误差19.19个像素）
[28] 年龄6~60岁 训练样本：150幅X线片
测量样本：100幅X线片		 19 Lindner组的平均径向误差16.56个像素;Ibragimov组的平均径向误差18.51个像素;Vandaele组的平均径向误差17.79个像素

Tab 2

Three-dimensional cephalometric measurement automatic fixed-point collection of article abstracts"

定点方法 参考文献 样本信息 样本量 标志点数量 自动定点误差
基于知识的
方法		 [30] 不分年龄、性别和种族 测量样本：30幅CBCT 20 平均误差（1.88±1.10）mm
[31] 不分年龄、性别和种族 测量样本：30幅CBCT 20 平均误差（2.01±1.23）mm
[32] 不分年龄、性别和种族 测量样本：30幅CBCT 标志点21个
长度28个
角度18个
比值7个		 未显示
最大平均误差2.63 mm
最大平均误差2.12°
最大平均误差0.03
基于模型的
方法		 [33] 女性,年龄37~74岁,健康白种人 测量样本：18幅CBCT 21 误差中位数1.99 mm
[34] 年龄10~45岁,排除有严重骨折或严重骨骼异常的图像 训练样本：8幅CBCT
测量样本：20幅CBCT		 14 平均误差3.40 mm
[35-36] 不分年龄、性别和种族 训练样本：24幅CBCT
测量样本：24幅CBCT
（与训练样本相同）		 18 平均误差（3.64±1.43）mm
改进后平均误差（2.51±1.60）mm
基于学习的
方法		 [38] 1）41名非综合征性牙颌面畸形患者（CBCT）;2）30名正常人（多层螺旋CT） 训练样本：41幅CBCT、30幅多层螺旋CT
测量样本：41幅CBCT
（与训练样本相同）		 15 平均误差1.44 mm
[39] 年龄16~54岁,不分性别和种族 训练样本：39幅CT
测量样本：39幅CT
（与训练样本相同）		 33 9个标志点平均误差<2 mm
[40] 许多样本有如出血、肿瘤等病理变化和与年龄有关的变化 训练样本：201幅CT
测量样本：20幅CT		 22 相对于专家A：平均误差（2.45±2.53）mm
相对于专家B：平均误差（3.49±2.88）mm
[41] 矫治前后颅颌面部畸形患者 训练样本：20幅CT
测量样本：20幅CT
（与训练样本相同）		 4 热图和Softargmax回归方法：97%标志点平均误差<4 mm
[42] 正常韩国成年人,平均年龄约24岁,不分性别 训练样本：20幅CT
测量样本：7幅CT		 7 x轴：0.49 mm、y轴：1.02 mm、z轴：1.40 mm,平均误差1.80 mm
[4] 1）26名正常韩国成年人;2）229名牙颌面畸形和错??畸形患者 CNN：1）中22幅CT用于训练,4幅CT用于测试;VAE和非线性回归：1）和2）中230幅CT用于训练,25幅CT用于测试 93 平均误差（3.63±1.41）mm
[37,43] 1）77名非综合征性牙颌面畸形患者（CBCT）
2）30名正常人（多层螺旋CT）		 训练样本：77幅CBCT,30幅多层螺旋CT
测量样本：77幅CBCT
（与训练样本相同）		 15 平均误差（1.10±0.71）mm
[44] 有颅面部先天性出生缺陷、发育异常、颅颌面创伤、手术治疗的儿童和成人患者,不分男女 训练样本：50幅CBCT
测量样本：50幅CBCT
（与训练样本相同）		 9 8个标志点平均误差<0.5 mm,颏前点平均误差为1.54 mm
[1] Cardillo J, Sid-Ahmed MA. An image processing system for locating craniofacial landmarks[J]. IEEE Trans Med Imaging, 1994,13(2):275-289.
[2] Medellín-Castillo HI, Govea-Valladares EH, Pérez-Guerrero CN, et al. The evaluation of a novel haptic-enabled virtual reality approach for computer-aided cephalometry[J]. Comput Methods Programs Bio-med, 2016,130:46-53.
[3] Cohen AM, Ip HH, Linney AD. A preliminary study of computer recognition and identification of skeletal landmarks as a new method of cephalometric a-nalysis[J]. Br J Orthod, 1984,11(3):143-154.
[4] Yun HS, Jang TJ, Lee SM, et al. Learning-based local-to-global landmark annotation for automatic 3D cephalometry[J]. Phys Med Biol, 2020,65(8):08-5018.
[5] Douglas TS. Image processing for craniofacial landmark identification and measurement: a review of photogrammetry and cephalometry[J]. Comput Med Imaging Graph, 2004,28(7):401-409.
[6] Tam WK, Lee HJ. Improving point correspondence in cephalograms by using a two-stage rectified point transform[J]. Comput Biol Med, 2015,65:114-123.
[7] Kafieh R, Mehri A, Sadri S. Automatic landmark detection in cephalometry using a modified active sha-pe model with sub image matching[C]// 2007 Int Conf Mach Vis. Isalambad, Pakistan: IEEE, 2007: 73-78.
[8] Leonardi R, Giordano D, Maiorana F, et al. Automatic cephalometric analysis[J]. Angle Orthod, 2008,78(1):145-151.
[9] Vandaele R, Aceto J, Muller M, et al. Landmark detection in 2D bioimages for geometric morphome-trics: a multi-resolution tree-based approach[J]. Sci Rep, 2018,8(1):538.
[10] Lindner C, Wang CW, Huang CT, et al. Fully automatic system for accurate localisation and analysis of cephalometric landmarks in lateral cephalograms[J]. Sci Rep, 2016,6:33581.
[11] Lévy-Mandel AD, Venetsanopoulos AN, Tsotsos JK. Knowledge-based landmarking of cephalograms[J]. Comput Biomed Res, 1986,19(3):282-309.
[12] Parthasarathy S, Nugent ST, Gregson PG, et al. Automatic landmarking of cephalograms[J]. Comput Biomed Res, 1989,22(3):248-269.
[13] Davis DN, Forsyth D. Knowledge-based cephalometric analysis: a comparison with clinicians using interactive computer methods[J]. Comput Biomed Res, 1994,27(3):210-228.
[14] Forsyth DB, Davis DN. Assessment of an automa-ted cephalometric analysis system[J]. Eur J Orthod, 1996,18(5):471-478.
[15] Cootes TF, Taylor CJ, Cooper DH, et al. Active shape models-their training and application[J]. Comput Vis Image Underst, 1995,61(1):38-59.
[16] Cootes TF, Taylor CJ. Statistical models of appea-rance for medical image analysis and computer vision[C]// Medical imaging 2001. San Diego, CA, USA: Image Processing, 2001,4322:236-248.
[17] Hutton TJ, Cunningham S, Hammond P. An evaluation of active shape models for the automatic identification of cephalometric landmarks[J]. Eur J Orthod, 2000,22(5):499-508.
[18] Rueda S, Alcañiz M. An approach for the automatic cephalometric landmark detection using mathematical morphology and active appearance models[J]. Med Image Comput Comput Assist Interv, 2006,9(Pt 1):159-166.
[19] Vucinić P, Trpovski Z, Sćepan I. Automatic landmarking of cephalograms using active appearance models[J]. Eur J Orthod, 2010,32(3):233-241.
[20] Kafieh R, Mehri A, Sadri S. Automatic landmark detection in cephalometry using a modified active sha-pe model with sub image matching[C]// 2007 International Conference on Machine Vision. December 28-29, 2007, Isalambad, Pakistan. IEEE, 2007: 73-78.
[21] Kaur A, Singh C. Automatic cephalometric landmark detection using Zernike moments and templa-te matching[J]. SIViP, 2015,9(1):117-132.
[22] Arık SÖ, Ibragimov B, Xing L. Fully automated quantitative cephalometry using convolutional neural networks[J]. J Med Imaging (Bellingham), 2017,4(1):014501.
[23] Kunz F, Stellzig-Eisenhauer A, Zeman F, et al. Artificial intelligence in orthodontics: evaluation of a fully automated cephalometric analysis using a customized convolutional neural network[J]. J Orofac Orthop, 2020,81(1):52-68.
[24] Park JH, Hwang HW, Moon JH, et al. Automated identification of cephalometric landmarks: part 1—comparisons between the latest deep-learning metho-ds YOLOV3 and SSD[J]. Angle Orthod, 2019,89(6):903-909.
[25] Hwang HW, Park JH, Moon JH, et al. Automated identification of cephalometric landmarks: part 2—might it be better than human[J]. Angle Orthod, 2020,90(1):69-76.
[26] Dai XB, Zhao H, Liu TL, et al. Locating anatomical landmarks on 2D lateral cephalograms through adversarial encoder-decoder networks[J]. IEEE Acce-ss, 2019,7:132738-132747.
[27] Wang CW, Huang CT, Hsieh MC, et al. Evaluation and comparison of anatomical landmark detection methods for cephalometric X-ray images: a grand challenge[J]. IEEE Trans Med Imaging, 2015,34(9):1890-1900.
[28] Wang CW, Huang CT, Lee JH, et al. A benchmark for comparison of dental radiography analysis algorithms[J]. Med Image Anal, 2016,31:63-76.
[29] Dot G, Rafflenbeul F, Arbotto M, et al. Accuracy and reliability of automatic three-dimensional ce-phalometric landmarking[J]. Int J Oral Maxillofac Surg, 2020,49(10):1367-1378.
[30] Neelapu BC, Kharbanda OP, Sardana V, et al. Automatic localization of three-dimensional cephalome-tric landmarks on CBCT images by extracting symmetry features of the skull[J]. Dentomaxillofac Radiol, 2018,47(2):20170054.
[31] Gupta A, Kharbanda OP, Sardana V, et al. A know-ledge-based algorithm for automatic detection of ce-phalometric landmarks on CBCT images[J]. Int J Comput Assist Radiol Surg, 2015,10(11):1737-1752.
[32] Gupta A, Kharbanda OP, Sardana V, et al. Accuracy of 3D cephalometric measurements based on an automatic knowledge-based landmark detection algorithm[J]. Int J Comput Assist Radiol Surg, 2016,11(7):1297-1309.
[33] Codari M, Caffini M, Tartaglia GM, et al. Computer-aided cephalometric landmark annotation for CBCT data[J]. Int J Comput Assist Radiol Surg, 2017,12(1):113-121.
[34] Shahidi S, Bahrampour E, Soltanimehr E, et al. The accuracy of a designed software for automated loca-lization of craniofacial landmarks on CBCT images[J]. BMC Med Imaging, 2014,14:32.
[35] Montúfar J, Romero M, Scougall-Vilchis RJ. Automatic 3-dimensional cephalometric landmarking ba-sed on active shape models in related projections[J]. Am J Orthod Dentofacial Orthop, 2018,153(3):449-458.
[36] Montúfar J, Romero M, Scougall-Vilchis RJ. Hybrid approach for automatic cephalometric landmark annotation on cone-beam computed tomography volumes[J]. Am J Orthod Dentofacial Orthop, 2018,154(1):140-150.
[37] Zhang J, Liu MX, Wang L, et al. Context-guided fully convolutional networks for joint craniomaxillofacial bone segmentation and landmark digitization[J]. Med Image Anal, 2020,60:101621.
[38] Zhang J, Gao YZ, Wang L, et al. Automatic craniomaxillofacial landmark digitization via segmentation-guided partially-joint regression forest model and multiscale statistical features[J]. IEEE Trans Biomed Eng, 2016,63(9):1820-1829.
[39] de Jong MA, Gül A, de Gijt JP, et al. Automated human skull landmarking with 2D Gabor wavelets[J]. Phys Med Biol, 2018,63(10):105011.
[40] O’Neil AQ, Kascenas A, Henry J, et al. Attaining human-level performance with atlas location autocontext for anatomical landmark detection in 3D CT data[J]. Lect Notes Comput Sci, 2019: 470-484.
[41] Lachinov D, Getmanskaya A, Turlapov V. Cephalometric landmark regression with convolutional neural networks on 3D computed tomography data[J]. Pattern Recognit Image Anal, 2020,30(3):512-522.
[42] Lee SM, Kim HP, Jeon K, et al. Automatic 3D cephalometric annotation system using shadowed 2D image-based machine learning[J]. Phys Med Biol, 2019,64(5):055002.
[43] Zhang J, Liu MX, Wang L, et al. Joint craniomaxillofacial bone segmentation and landmark digitization by context-guided fully convolutional networks[J]. Med Image Comput Comput Assist Interv, 2017,10434:720-728.
[44] Torosdagli N, Liberton DK, Verma P, et al. Deep geodesic learning for segmentation and anatomical landmarking[J]. IEEE Trans Med Imaging, 2019,38(4):919-931.
[1] Liu Panming,Li Zhengze,Li Junhe,Cui Shuxia.. Cone beam computed tomography study of maxillary sinus volume and oropharyngeal airway volume with diffe-rent vertical skeletal faces in adult skeletal Class Ⅱ patients [J]. Int J Stomatol, 2023, 50(5): 528-537.
[2] Yang Yunan,Liu Peng,Wang Hu,You Meng. Evaluation of maxillary sinusitis with cone beam computed tomography [J]. Int J Stomatol, 2023, 50(3): 302-307.
[3] Ye Zelin,Liu Lu,Long Hu,You Meng. Research progress on imaging evaluation and treatment of anterior dilacerated teeth [J]. Int J Stomatol, 2022, 49(2): 173-181.
[4] Zhou Mengqi,Chen Xuepeng,Fu Baiping. Strategies for preventing alveolar-bone dehiscence and fenestration during orthodontic treatment [J]. Int J Stomatol, 2021, 48(5): 600-608.
[5] Shi Danni,Yang Xin,Wu Jianyong. Research progress of reference coordinate system for three-dimensional cephalometric based on cone beam computed tomography [J]. Int J Stomatol, 2021, 48(4): 398-404.
[6] Wu Chunlan,Tang Hua,Chen Jun. Alveolar bone morphology of anterior teeth areas in patients with high-angle skeletal Class Ⅱ open bite [J]. Int J Stomatol, 2021, 48(4): 426-432.
[7] Wang Lidong,Ma Wen,Fu Shuai,Zhang Changbin,Cui Qingying,Liang Yan,Li Ming. Research and accuracy of different methods of digital occlusal splint fabrication for orthognathic surgery [J]. Int J Stomatol, 2021, 48(2): 156-164.
[8] Wang Ben,Xu Zhezhen,Wei Xi. Application and progress of a digitalized minimally invasive technique in endodontics [J]. Int J Stomatol, 2021, 48(1): 110-118.
[9] Zhang Zihan,Xiong Xin,Wang Jun. Research status and application development of three-dimensional cephalometry [J]. Int J Stomatol, 2020, 47(6): 739-744.
[10] Tang Bei,Zhao Wenjun,Wang Hu,Zheng Guangning,You Meng. Inferior alveolar nerve injury due to apical overfilling: two cases reports [J]. Int J Stomatol, 2020, 47(3): 293-296.
[11] Zhang Tingting,Hu Changhong,Peng Yan,Zhou Wenqiao,Zhang Huicong,Liu Die. Analysis of upper lip profile features in 300 dentate subjects of different ages via three-dimensional measurement of cone beam computed tomography [J]. Int J Stomatol, 2020, 47(2): 182-188.
[12] Li Qi, Huang Shaohong. Cone beam computed tomography study on root and root canal morphology of mandibular second permanent molars in Guangfu population in Lingnan area [J]. Int J Stomatol, 2019, 46(6): 640-649.
[13] Liu Xiaohua, Wang Enbo. Application progress on cone beam computed tomography imaging localization in extraction of embedded supernumerary teeth in children [J]. Inter J Stomatol, 2018, 45(3): 295-300.
[14] Yin Yijia, Feng Jie, Luo Mengqi, Han Xianglong. Research progress on the correction of natural head position with three-dimensional stereo-photography [J]. Inter J Stomatol, 2018, 45(3): 301-306.
[15] Xie Chengjia, Ge Shaohua.. Progress review of diagnosis and treatment of cemental tear [J]. Inter J Stomatol, 2017, 44(3): 315-319.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
[1] . [J]. Foreign Med Sci: Stomatol, 1999, 26(06): .
[2] . [J]. Foreign Med Sci: Stomatol, 1999, 26(06): .
[3] . [J]. Foreign Med Sci: Stomatol, 1999, 26(06): .
[4] . [J]. Foreign Med Sci: Stomatol, 1999, 26(06): .
[5] . [J]. Foreign Med Sci: Stomatol, 1999, 26(06): .
[6] . [J]. Foreign Med Sci: Stomatol, 1999, 26(05): .
[7] . [J]. Foreign Med Sci: Stomatol, 1999, 26(05): .
[8] . [J]. Foreign Med Sci: Stomatol, 1999, 26(05): .
[9] . [J]. Foreign Med Sci: Stomatol, 1999, 26(04): .
[10] . [J]. Foreign Med Sci: Stomatol, 1999, 26(04): .