El Bsat Abdul Rehman, Shammas Elie, Asmar Daniel, Zeno Kinan G, Macari Anthony T, Ghafari Joseph G
Department of Mechanical Engineering, Maroun Semaan Faculty of Engineering and Architecture, American University of Beirut, Beirut P.O. Box 11-0236, Lebanon.
Department of Dentofacial Medicine, Faculty of Medicine, American University of Beirut, Beirut P.O. Box 11-0236, Lebanon.
Diagnostics (Basel). 2025 Jun 2;15(11):1415. doi: 10.3390/diagnostics15111415.
The segmentation of individual teeth in three-dimensional (3D) dental models is a key step in orthodontic computer-aided design systems. Traditional methods lack robustness when handling challenging cases such as missing or misaligned teeth. to semantically segment maxillary teeth and palatal rugae in 3D textured scans using Convolutional Neural Networks (CNNs) and assess tooth movement after orthodontic treatment using stable rugae references. Building on the robustness of two-dimensional image semantic segmentation, we developed a method to convert 3D textured palate scans into two-dimensional images for segmentation, then back projected them onto the original 3D meshes. A dataset of 100 textured scans from 100 patients seeking orthodontic treatment was manually segmented by orthodontic experts. The proposed 3D segmentation method was applied to these scans. Finally, each pair of segmented 3D scans from the same patient, before and after treatment, was aligned by superimposing them on the stable rugae region. The 3D segmentation method achieved an accuracy of 98.69% and an average Intersection over Union (IoU) of 84.5%. The common stable coordinate frame for both scans using the rugae area as a stable reference enabled the computation of the 3D translational and rotational motions of each maxillary tooth. Neither pre- nor post-processing of the data was required to enhance segmentation. The proposed method enabled successful motion measurement of teeth using the rugal area as a stable reference and providing rotation and translational measurements of the maxillary teeth.
在三维（3D）牙科模型中对单个牙齿进行分割是正畸计算机辅助设计系统中的关键步骤。传统方法在处理诸如牙齿缺失或排列不齐等具有挑战性的情况时缺乏鲁棒性。本研究旨在使用卷积神经网络（CNN）对3D纹理扫描中的上颌牙齿和腭皱襞进行语义分割，并使用稳定的皱襞参考来评估正畸治疗后的牙齿移动。基于二维图像语义分割的鲁棒性，我们开发了一种方法，将3D纹理腭扫描转换为二维图像进行分割，然后将其反向投影到原始3D网格上。由正畸专家对来自100名寻求正畸治疗患者的100次纹理扫描数据集进行手动分割。将所提出的3D分割方法应用于这些扫描。最后，通过将同一患者治疗前后的每对分割后的3D扫描叠加在稳定的皱襞区域上进行对齐。该3D分割方法的准确率达到98.69%，平均交并比（IoU）为84.5%。以皱襞区域作为稳定参考的两次扫描的公共稳定坐标系能够计算每个上颌牙齿的3D平移和旋转运动。无需对数据进行预处理或后处理来增强分割效果。所提出的方法能够以皱襞区域作为稳定参考成功测量牙齿运动，并提供上颌牙齿的旋转和平移测量结果。
