近年來，隨著大眾對美學意識的提升，每年進行牙齒矯正治療的病例數量逐
漸增多，其中以配戴隱形牙套的治療方式受到患者的歡迎。隨著疫情逐漸趨緩，
隱形牙套市場迎來了顯著的成長，數位牙科領域也出現了許多研究，希望以高效
率、高精準度和永續發展的技術來生產隱形牙套。
本研究提出一種用於生成隱形牙套切割路徑的演算法。首先，計算牙齒模型
的特徵向量，對姿態進行正規化處理，確保生成演算法的穩定性。然後，通過曲
率濾波器和分群濾波器，從數位牙齒模型中提取牙齦邊緣特徵，作為切割路徑的
期望答案。最後，輸入預先建立的初始輪廓，透過迭代的方式使輪廓朝向牙齦邊
緣內縮，直到完成數位切割路徑。
為了取得較佳的切割路徑，本研究針對路徑生成演算法中各個步驟進行討論。
資料前處理透過牙齒模型的特徵向量作姿態和位置的調整，成功地將模型以規範
的姿態輸入演算法。在特徵過濾步驟，由於表面曲率特徵的複雜程度，採用圖形
化使用者介面的方式，整合曲率濾波器和分群濾波器完成特徵提取的任務。在輪
廓迭代步驟，使用中間值的加速函數可以產生品質穩定的收縮結果，同時通過平
滑函數對收縮結果進行優化，使輪廓的頂點呈現等間距排列，有效提升輪廓形狀
的細節程度。
在專業牙醫師的建議與指導下，本研究所提出的切割路徑演算法符合目前隱
形牙套在製作上的需求，能夠協助牙醫師進行隱形牙套的設計和規劃。未來將整
合自動加工設備，完整隱形牙套的自動化生產線。

In recent years, there has been a surge in the number of orthodontic cases as
people's aesthetic awareness has grown. Among the various treatment options, clear
aligners have become the most popular option among patients. After the pandemic,
the clear aligner market has witnessed significant growth. Many researches in digital
dentistry focused precise and sustainable techniques for efficiently producing clear
aligners.
This study proposed an algorithm for generating potential cutting paths of clear
aligners. It began by calculating eigenvectors of the dental models and normalizing
their posture to ensure algorithm stability. Gingival margin was extracted from the
digital models using curvature features and the clustering operation. And it was
served as the answer to the final cutting path. Through iteration of predefined contour,
the algorithm gradually shrank the contour towards gingival margin until the cutting
path was completed.
To have a quality cutting path, each step of the path generation was disclosed.
Data preprocessing adjusted the dental model's posture and position according to
eigenvectors of momentum of the shape. Feature filtering employed a graphical user
interface to visually integrate curvatures and clustering operation, and assisted for
extracting features. Contour iteration utilized an accelerated function and a smoothing
function to achieve stable contraction results with evenly spaced contour vertices.
The proposed cutting path algorithm meets the current requirements for
producing clear aligners. It supports dentists to design and plan for invisible braces. In
the future, there are plans to build an automated production line for clear aligners.

[1]M. J. Peluso, S. D. Josell, S. W. Levine et al., “Digital models: An introduction,” Seminars in Orthodontics, vol. 10, no. 3, pp. 226–238, 2004, doi: 10.1053/j.sodo.2004.05.007.
[2]S. Akyalcin, B. E. Cozad, J. D. English et al., “Diagnostic accuracy of impression-free digital models,” American Journal of Orthodontics and Dentofacial Orthopedics, vol. 144, no. 6, pp. 916–922, 2013, doi: 10.1016/j.ajodo.2013.04.024.
[3]B. S. Akdeni̇z, V. Aykaç, M. Turgut, and S. Çeti̇n, “Digital dental models in orthodontics: A review,” J. Exp. Clin. Med., vol. 39, no. 1, Art. no. 1, 2022.
[4]Y. J. Hsu, K. Y. Wang, H. C. Chiu et al., “Application of digital models in orthodontic and maxillofacial surgery – A literature review,” Taiwanese Journal of Orthodontics, vol. 27, no. 3, 2022, doi: 10.30036/TJO.201606_27(3).0001.
[5]T. V. Flügge, S. Schlager, K. Nelson et al., “Precision of intraoral digital dental impressions with iTero and extraoral digitization with the iTero and a model scanner,” American Journal of Orthodontics and Dentofacial Orthopedics, vol. 144, no. 3, pp. 471–478, 2013, doi: 10.1016/j.ajodo.2013.04.017.
[6]J. A. Burzynski, A. R. Firestone, F. M. Beck et al., “Comparison of digital intraoral scanners and alginate impressions: Time and patient satisfaction,” American Journal of Orthodontics and Dentofacial Orthopedics, vol. 153, no. 4, pp. 534–541, 2018, doi: 10.1016/j.ajodo.2017.08.017.
[7]S. Y. Kim, Y. S. Shin, H. D. Jung et al., “Precision and trueness of dental models manufactured with different 3-dimensional printing techniques,” American Journal of Orthodontics and Dentofacial Orthopedics, vol. 153, no. 1, pp. 144–153, 2018, doi: 10.1016/j.ajodo.2017.05.025.
[8]Z. C. Zhang, P. L. Li, F. T. Chu et al., “Influence of the three-dimensional printing technique and printing layer thickness on model accuracy,” Journal of Orofacial Orthopedics, vol. 80, no. 4, pp. 194–204, 2019, doi: 10.1007/s00056-019-00180-y.
[9]T. T. Wheeler, “Orthodontic clear aligner treatment,” Seminars in Orthodontics, vol. 23, no. 1, pp. 83–89, 2017, doi: 10.1053/j.sodo.2016.10.009.
[10]Boyd, L. Robert, R. J. Miller et al., “The Invisalign system in adult orthodontics: mild crowding and space closure cases,” Journal of Clinical Orthodontics, vol. 34, no. 4, pp. 203–212, 2000.
[11]Y. M. Bichu et al., “Advances in orthodontic clear aligner materials,” Bioactive Materials, vol. 22, pp. 384–403, 2023, doi: 10.1016/j.bioactmat.2022.10.006.
[12]J. Hartshorne and M. B. Wertheimer, “Emerging insights and new developments in clear aligner therapy: A review of the literature,” AJO-DO Clinical Companion, vol. 2, no. 4, pp. 311–324, 2022, doi: 10.1016/j.xaor.2022.04.009.
[13]P. Shivapuja, D. Shah, N. Shah et al., “Direct 3D-printed orthodontic aligners with torque, rotation, and full control anchors,” U.S. Patent and Trademark Office, No. 10,179,035, 2019.
[14]P. Jindal, M. Juneja, F. L. Siena et al., “Mechanical and geometric properties of thermoformed and 3D printed clear dental aligners,” American Journal of Orthodontics and Dentofacial Orthopedics, vol. 156, no. 5, pp. 694–701, 2019, doi: 10.1016/j.ajodo.2019.05.012.
[15]B. Peeters, N. Kiratli, and J. Semeijn, “A barrier analysis for distributed recycling of 3D printing waste: Taking the maker movement perspective,” Journal of Cleaner Production, vol. 241, p. 118313, 2019, doi: 10.1016/j.jclepro.2019.118313.
[16]R. Thukral, A. Gupta, “Invisalign: invisible orthodontic treatment-a review,” Journal of Advanced Medical and Dental Sciences Research, vol. 3, 2015
[17]D. Cowley, “Effect of gingival margin design on retention of thermoformed orthodontic aligners,” UNLV Theses, Dissertations, Professional Papers, and Capstones, 2012, doi: 10.34917/4332643.
[18]X. Lyu, X. Cao, J. Yan et al., “Biomechanical effects of clear aligners with different thicknesses and gingival-margin morphology for appliance design optimization,” American Journal of Orthodontics and Dentofacial Orthopedics, 2023, doi: 10.1016/j.ajodo.2022.12.014.
[19]M. L. Laine, V. Moustakis, L. Koumakis et al., “Modeling Susceptibility to Periodontitis,” Journal of Dental Research, vol. 92, no. 1, pp. 45–50, 2012, doi:10.1177/0022034512465435
[20]W. Wu, N. Yang, X. Feng et al., “Effect of vitamin C administration on hydrogen peroxide‑induced cytotoxicity in periodontal ligament cells,” Molecular Medicine Reports, vol. 11, no. 1, pp. 242–248, 2014, doi: 10.3892/mmr.2014.2712
[21]C. Albuquerque et al., “Canine periodontitis: The dog as an important model for periodontal studies,” The Veterinary Journal, vol. 191, no. 3, pp. 299–305, 2012, doi: 10.1016/j.tvjl.2011.08.017.
[22]A. Di Benedetto, I. Gigante, S. Colucci et al., “Periodontal Disease: Linking the Primary Inflammation to Bone Loss,” Journal of Immunology Research, vol. 2013, p. e503754, 2013, doi: 10.1155/2013/503754.
[23]M. Kuralt and A. Fidler, “A novel computer-aided method for direct measurements and visualization of gingival margin changes,” Journal of Clinical Periodontology, vol. 49, no. 2, pp. 153–163, 2022, doi: 10.1111/jcpe.13573.
[24]M. Kuralt, A. Cmok Kučič, R. Gašperšič et al., “Gingival shape analysis using surface curvature estimation of the intraoral scans,” BMC Oral Health, vol. 22, no. 1, p. 283, 2022, doi: 10.1186/s12903-022-02322-y.
[25]B. Hamann, “Curvature approximation for triangulated surfaces,” in Geometric Modelling, G. Farin, H. Noltemeier, H. Hagen, and W. Knödel, Eds., Computing Supplementum, vol. 8. Vienna: Springer Vienna, 1993, pp. 139–153. doi: 10.1007/978-3-7091-6916-2_10.
[26]G. Taubin, “Estimating the tensor of curvature of a surface from a polyhedral approximation,” in Proceedings of IEEE International Conference on Computer Vision, 1995, pp. 902–907. doi: 10.1109/ICCV.1995.466840.
[27]M. Meyer, M. Desbrun, P. Schröder, and A. H. Barr, “Discrete differential-geometry operators for triangulated 2-manifolds,” in Visualization and Mathematics III, H.-C. Hege and K. Polthier, Eds., in Mathematics and Visualization. Berlin, Heidelberg: Springer, 2003, pp. 35–57. doi: 10.1007/978-3-662-05105-4_2.
[28]A. Razdan and M. Bae, “Curvature estimation scheme for triangle meshes using biquadratic Bézier patches,” Computer-Aided Design, vol. 37, no. 14, pp. 1481–1491, 2005, doi: 10.1016/j.cad.2005.03.003.
[29]G. Guennebaud and M. Gross, “Algebraic point set surfaces,” Association for Computing Machinery, pp. 23-es, 2007, doi: 10.1145/1275808.1276406.
[30]E. Kalogerakis, P. Simari, D. Nowrouzezahrai et al., “Robust statistical estimation of curvature on discretized surfaces,” Symposium on Geometry Processing, no. 10, pp. 13–22, 2007, doi: 10.5555/1281991.1281993
[31]S. Kim and S. Choi, “Automatic tooth segmentation of dental mesh using a transverse plane,” Engineering in Medicine and Biology Society (EMBC), pp. 4122–4125, 2018, doi: 10.1109/EMBC.2018.8513318.
[32]X. Bing, W. E. I. Xin, H. U. Wei et al., “Tooth segmentation algorithm based on segmentation of feature line,” Journal of Computer Applications, vol. 37, no. 3, p. 844, 2017, doi: 10.11772/j.issn.1001-9081.2017.03.844.
[33]J. Zhang, C. Li, Q. Song et al., “Automatic 3D tooth segmentation using convolutional neural networks in harmonic parameter space,” Graphical Models, vol. 109, p. 101071, 2020, doi: 10.1016/j.gmod.2020.101071.