近年來牙科診所廣泛設立，目的就是解決每個人愛美與口腔健康的需求，Cone-beam computed tomography (CBCT)則是在患者在進行治療時所需要照射的影像，現今不管是植牙還是矯正牙齒，醫生都必須依照這些影像來做術前的規劃與判斷，因為CBCT提供了三維的口腔結構來做牙科的臨床應用。
在做植牙手術前，牙醫必須透過專業經驗在三維的口腔中框選出牙齒與上下顎骨才有辦法在缺牙的部分上精準地植上假牙。目前業界的植牙軟體都採用亮度值的方式來分割牙齒與骨頭。但是在靠近上顎與下顎的部分，牙齒和牙槽骨的灰度密度值在根部區域相似，即使對於具有相同射線密度的物體，灰度值也可能根據其相對位置而有所不同。因此，通過簡單的閾值處理方法不能輕易地分離牙齒及其周圍結構，也造成無法精準地分割牙齒與骨頭，導致了執行術前規劃的時候，醫生的判斷會有所錯誤。這往往需要有經驗的醫生才有辦法實際來動手術才有辦法解決這個問題，也導致判斷上費時且對病人有較大的手術風險。
因此本論文所提出的影像處裡系統分成三大部分，上下顎骨分割，牙齒分割以及牙齒辨識，用來自動分割病人的每顆牙齒與上下顎骨，並辨識每顆牙齒的類別，來輔助醫生做植牙前的規劃。這裡利用影像處裡與深度學習AI的技術有效解決目前業界軟體所欠缺的功能。主要在亮度相近的上下顎骨中分割牙齒與骨頭，並在有金屬假牙的口腔中清楚分割牙齒，再來自動辨識每顆牙齒的類別，在臨床應用上大大的幫助醫生解決手術規劃問題。
實驗結果分成三大部分，我們將比較骨頭分割結果，牙齒分割的準確度與牙齒辨識的正確性。因此，本論文所提出的方法能夠提供自動且方便的工具，以幫助牙醫生更準確且一致地進行植牙手術前的評估與治療。

In recent years, dental clinics have been widely established to meet popular needs in beauty and oral health. Cone-beam computed tomography (CBCT) is the imaging instrument that patients need to irradiate during treatment. Nowadays, whether it is implants or orthodontics, doctors must make preoperative planning and judgment based on these images, because CBCT provides a three-dimensional oral structure for clinical dental applications.
Before performing dental implantation surgery, the dentist must use professional experience to select the teeth and the upper and lower jaw bones in the three-dimensional mouth to be able to accurately implant the dentures on the location of missing teeth. At present, the dental implant software in the industry uses the brightness value to segment teeth and bones. However, the gray density values of teeth and alveolar bone are similar in the root area near the upper and lower jaws, since the objects are with similar radiation. However, the gray value may also vary according to its relative position. Therefore, the teeth and their surrounding structures sometimes cannot be easily separated by a simple threshold processing method, which results in the inability to accurately segment the teeth and bones. As a result, when performing preoperative planning, the doctor’s judgment will be prone to wrong arrangements. It often takes an experienced doctor to have the method to actually perform the operation to solve this problem, but this is time-consuming in judgment and also converts surgical risks for the patient.
Therefore, the image processing system proposed in this paper is divided into three parts: upper and lower jawbone segmentation, tooth segmentation and tooth identification, which are used to automatically segment each patient’s teeth and upper and lower jaw bones and identify the type of each tooth to assist the doctor the planning and operation for implantation. Here we use the technology of image processing and deep learning AI to effectively solve the current lack of efficient segmentation and identification functions in the industry. It mainly divides teeth and bones in the upper and lower jaw bones with similar brightness, and clearly divides the teeth in the oral cavity with metal dentures, and then automatically recognizes them. The category of each tooth greatly helps doctors to resolve the problem of surgical planning in clinical applications.
The experimental results are divided into three parts. We will compare the results of bone segmentation, the accuracy of tooth segmentation and the correctness of tooth recognition. Therefore, the method proposed in this thesis can provide automatic and convenient tools to help dentists performing more accurate and consistent evaluation and treatment before implant surgery.

[1] H. Gao and O. Chae, “Individual tooth segmentation from ct images using level set method with shape and intensity prior,” Pattern Recognition, vol. 43, no. 7, pp. 2406–2417, 2010.
[2] Y. Gan, Z. Xia, J. Xiong, Q. Zhao, Y. Hu, and J. Zhang, “Toward accurate tooth segmentation from computed tomography images using a hybrid level set model,” Medical physics, vol. 42, no. 1, pp. 14–27, 2015.
[3] H.-T. Yau, T.-J. Yang, and Y.-C. Chen, “Tooth model reconstruction based upon data fusion for orthodontic treatment simulation,” Computers in biology and medicine, vol. 48, pp. 8–16, 2014.
[4] D. X. Ji, S. H. Ong, and K. W. C. Foong, “A level-set based approach for anterior teeth segmentation in cone beam computed tomography images,” Computers in biology and medicine, vol. 50, pp. 116–128, 2014.
[5] Y. Wang, S. Liu, G. Wang, and Y. Liu, “Accurate tooth segmentation with improved hybrid active contour model,” Physics in Medicine & Biology, vol. 64, no. 1, p. 015012, 2018.
[6] L. Hiew, S. Ong, K. W. Foong, and C. Weng, “Tooth segmentation from cone-beam ct using graph cut,” in Proceedings of the Second APSIPA Annual Summit and Conference, 2010, pp. 272–275.
[7] J. Keustermans, D. Vandermeulen, and P. Suetens, “Integrating statistical shape models into a graph cut framework for tooth segmentation,” in International Workshop on Machine Learning in Medical Imaging. Springer, 2012, pp. 242–249.
[8] S. Lee, S. Woo, J. Yu, J. Seo, J. Lee, and C. Lee, “Automated cnn-based tooth segmentation in cone-beam ct for dental implant planning,” IEEE Access, vol. 8, pp. 50 507–50 518, 2020.
[9] Y. Rao, Y. Wang, F. Meng, J. Pu, J. Sun, and Q. Wang, “A symmetric fully convolutional residual network with dcrf for accurate tooth segmentation,” IEEE Access, vol. 8, pp. 92 028–92 038, 2020.
[10] J. Long, E. Shelhamer, and T. Darrell, “Fully convolutional networks for semantic segmentation,” in Proceedings of the IEEE conference on computer vision and pattern recognition, 2015, pp. 3431–3440.
[11] Y. Chen, H. Du, Z. Yun, S. Yang, Z. Dai, L. Zhong, Q. Feng, and W. Yang, “Automatic segmentation of individual tooth in dental cbct images from tooth surface map by a multi-task fcn,” IEEE Access, 2020.
[12] Z. Cui, C. Li, and W. Wang, “Toothnet: automatic tooth instance segmentation and identification from cone beam ct images,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2019, pp. 6368–6377.
[13] K. He, G. Gkioxari, P. Doll´ar, and R. Girshick, “Mask r-cnn,” in Proceedings of the IEEE international conference on computer vision, 2017, pp. 2961–2969.
[14] Y. Miki, C. Muramatsu, T. Hayashi, X. Zhou, T. Hara, A. Katsumata, and H. Fujita, “Classification of teeth in cone-beam ct using deep convolutional neural network,” Computers in biology and medicine, vol. 80, pp. 24–29, 2017.
[15] P.L. Lin, Y.H. Lai, P.W. Huang, An effective classification and numbering system for dental bitewing radiographs using teeth region and contour information, Pattern Recognit. 43 (2010) 1380–1392.
[16] M. Hosntalab, R.A. Zoroofi, A.A. Tehrani-Fard, G. Shirani, Classification and numbering of teeth in multi-slice CT images using wavelet-Fourier descriptor, Int. J. CARS 5 (2010) 237–249.
[17] Y. Gan, Z. Xia, J. Xiong, G. Li and Q. Zhao, "Tooth and Alveolar Bone Segmentation From Dental Computed Tomography Images," in IEEE Journal of Biomedical and Health Informatics, vol. 22, no. 1, pp. 196-204, Jan. 2018, doi: 10.1109/JBHI.2017.2709406.
[18] Wang, H., J. Minnema, K.J. Batenburg, T. Forouzanfar, F.J. Hu, and G. Wu. “Multiclass CBCT Image Segmentation for Orthodontics with Deep Learning.” Journal of Dental Research, (March 2021).
[19] J. Long, E. Shelhamer and T. Darrell, "Fully convolutional networks for semantic segmentation," 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2015, pp. 3431-3440, doi: 10.1109/CVPR.2015.7298965.
[20] Hoel Kervadec, Jihene Bouchtiba, Christian Desrosiers, Eric Granger, Jose Dolz, and Ismail Ben Ayed. Boundary loss for highly unbalanced segmentation. In Proceedings of The 2nd International Conference on Medical Imaging with Deep Learning, volume 102 of Proceedings of Machine Learning Research, pages 285{296, 2019.
[21] Davood Karimi and Septimiu E. Salcudean. Reducing the hausdor distance in medical image segmentation with convolutional neural networks. IEEE Transactions on Medical Imaging, pages 1{1, 2019.
[22] Yuan Xue, Hui Tang, Zhi Qiao, Guanzhong Gong, Yong Yin, Zhen Qian, Chao Huang, Wei Fan, and Xiaolei Huang. Shape-aware organ segmentation by predicting signed distance maps. In Proceedings of the Thirty-Fourth AAAI Conference on Artificial Intelligence, 2020.
[23] Ronneberger O., Fischer P., Brox T.: ‘U-Net: convolutional networks for biomedical image segmentation’. Part III Proc. 18th Int. Conf. on Medical Image Computing and Computer-Assisted Intervention (MICCAI 2015), Munich, Germany, 5–9 October 2015, pp. 234–241
[24] F. I. Diakogiannis, F. Waldner, P. Caccetta, and C. Wu, “Resunet-a: A deep learning framework for semantic segmentation of remotely sensed data,” Isprs Journal of Photogrammetry and Remote Sensing, vol. 162, pp. 94–114, 2020.
[25] Y. Cao, S. Liu, Y. Peng, and J. Li, “DenseUNet: densely connected UNet for
electron microscopy image segmentation,” IET Image Processing, vol. 14, no. 12, pp. 2682–2689, 2020.
[26] Fausto Milletari, Nassir Navab, and Seyed-Ahmad Ahmadi. V-net: fully convolutional neural networks for volumetric medical image segmentation. In 2016 Fourth International Conference on 3D Vision (3DV), pages 565{571, 2016.
[27] O. Çiçek, A. Abdulkadir, S. S. Lienkamp, T. Brox, and O. Ronneberger,“3D U-Net: Learning dense volumetric segmentation from sparse annotation,” in Proc. Int. Conf. Med. Image Comput. Comput.-Assist. Intervent. Cham, Switzerland: Springer, 2016, pp. 424–432.
[28] L.Yu, J.-Z. Cheng, Q. Dou, X.Yang, H. Chen, J.Qin, and P.-A.Heng, “Automatic 3d cardiovascular mr segmentation with densely-connected volumetric convnets,” in Int. Conf. Medical Image Comput. Comput.-Assisted Intervention, 2017, pp. 287–295.
[29] Toan Duc Bui, Jitae Shin, and Taesup Moon, “3d densely convolutional networks for volumetric segmentation,”arXiv preprint arXiv:1709.03199, 2017.
[30] R. Pauwels, R. Jacobs, S. R. Singer, and M. Mupparapu, ``CBCT-based bone quality assessment: Are hounsfield units applicable?'' Dentomaxillo-facial Radiol., vol. 44, no. 1, Jan. 2015, Art. no. 20140238.
[31] D. V. Tuzoff, L. N. Tuzova, M. M. Bornstein, A. S. Krasnov, M. A. Kharchenko, S. I. Nikolenko, M. M. Sveshnikov, and G. B. Bednenko, “Tooth detection and numbering in panoramic radiographs using convolutional neural networks,” Dentomaxillofacial Radiology, vol. 48, no. 4, p. 20180051, 2019.
[32] J. Fu, J. Liu, H. Tian, Y. Li, Y. Bao, Z. Fang, and H. Lu, “Dual attention network for scene segmentation,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2019, pp. 3146–3154. 3, 8, 10
[33] Han Hu, Zheng Zhang, Zhenda Xie, and Stephen Lin. Local relation networks for image recognition. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 3464–3473, 2019.
[34] A. F. Agarap, “Deep learning using rectified linear units (relu),” arXiv:1803.08375, 2018.
[35] He, Lifeng et al. “Two efficient label-equivalence-based connected-component labeling algorithms for 3-D binary images.” IEEE transactions on image processing: a publication of the IEEE Signal Processing Society vol. 20,8 (2011): 2122-34. doi:10.1109/TIP.2011.2114352
[36] X. Li, W. Wang, X. Hu, and J. Yang, “Selective kernel networks,” in Proc. IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recognition., 2019, pp. 510–519.
[37] Szegedy, C., Ioffe, S., Vanhoucke, V., Alemi, A.A.: Inception-v4, inception-resnet and the impact of residual connections on learning. In: Thirty-First AAAI Conference on Artificial Intelligence (2017)
[38] Yuxin Wu and Kaiming He. Group normalization. In Proceedings of the European Conference on Computer Vision (ECCV), pages 3–19, 2018.
[39] J. Dai, H. Qi, Y. Xiong, Y. Li, G. Zhang, H. Hu, and Y.Wei, “Deformable convolutional networks,” in Proceedings of the IEEE international conference on computer vision, pp. 764–773, 2017.
[40] T. Lehmann, C. Göonner, K. Spitzer, Survey: interpolation methods in medical image processing, IEEE Transactions on Medical Imaging 18 (11) (1999)1049–1075
[41] D. P. Kingma and J. Ba, “Adam: A method for stochastic optimization,” arXiv preprint arXiv:1412.6980, 2014.
[42] Ioffe S, Szegedy C (2015) Batch normalization: accelerating deep network training by reducing internal covariate shift. In: International Conference on Machine Learning, pp. 448–456