基于深度学习的个性化牙齿形态重建智能系统初步探索
该研究融合隐式模板与深度学习构建可变形牙齿深度隐式网络(T-DDIN),用550例口内扫描模型(训练500例、测试50例)评估其对Ⅱ类洞缺损和咬合面磨耗的形态重建精度、稳定性与效率;结果显示T-DDIN组三维偏差、调𬌗体积、牙尖角差值均小于传统CAD组(均P<0.001),重建时间更短。
要点
- T-DDIN组Ⅱ类洞与磨耗重建三维偏差分别为(0.14±0.05)、(0.16±0.09)mm,均小于传统CAD组(均P<0.001)。
- T-DDIN组调𬌗体积分别为(0.44±0.03)、(0.49±0.03)mm3,牙尖角差值分别为5.69°±1.90°、6.04°±0.53°,均小于传统CAD组(均P<0.001)。
- T-DDIN组重建时间分别为(37.2±7.7)、(39.4±6.2)s,均少于传统CAD组(均P<0.001)。
核心结论
- 针对Ⅱ类洞缺损和咬合面磨耗的形态重建,T-DDIN均表现出优异的形态重建稳定性、精度和效率。
研究设计与样本
- 收集2022年3月至2024年3月四川大学华西口腔医院正畸科和修复科收治的550例牙体缺损或错𬌗畸形患者的口内扫描模型,训练集500例、测试集50例。
- 研发T-DDIN,通过隐式模板与潜在编码预测网络实现缺损牙体的形态重建。
- 对测试集数据模拟Ⅱ类洞缺损和咬合面磨耗,分别用传统计算机辅助设计(CAD)和T-DDIN进行形态重建(分别为传统CAD组和T-DDIN组),以原始牙为真值计算下颌右侧第一磨牙形态重建后的三维偏差、调𬌗体积、牙尖角差值和形态重建时间。
主要发现
- T-DDIN组Ⅱ类洞缺损与咬合面磨耗重建的三维偏差分别为(0.14±0.05)、(0.16±0.09)mm,均显著小于传统CAD组相应缺损重建(均P<0.001)。
- T-DDIN组Ⅱ类洞缺损与咬合面磨耗重建的调𬌗体积分别为(0.44±0.03)、(0.49±0.03)mm3,均显著小于传统CAD组相应缺损重建(均P<0.001)。
- T-DDIN组Ⅱ类洞缺损与咬合面磨耗重建的牙尖角差值分别为5.69°±1.90°、6.04°±0.53°,均显著小于传统CAD组相应缺损重建(均P<0.001)。
- T-DDIN组两类缺损重建的三维偏差(P=0.098)、调𬌗体积(P=0.154)及牙尖角差值(P=0.196)的组内差异均无统计学意义。
- 传统CAD组咬合面磨耗重建的三维偏差、调𬌗体积与牙尖角差值均显著大于Ⅱ类洞缺损重建(均P<0.001)。
临床怎么用
- 原文未据此给出推荐。
局限性
- 原文未报告局限性。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
Objective: To integrate implicit templates with deep learning techniques, a novel neural network, the tooth-deformable deep implicit network (T-DDIN), was constructed to achieve high-precision shape completion of tooth defects in a personalized manner. Methods: A total of 550 intraoral scan models were collected from patients treated at the Department of Orthodontics and Department of Prosthodontics, West China Hospital of Stomatology, Sichuan University (500 for training and 50 for testing), between March 2022 and March 2024. T-DDIN reconstructed defective tooth morphology using an implicit template and a latent encoding prediction network. During model evaluation, Class Ⅱ cavity defects and occlusal wear defects were simulated in the test set. Morphological restoration was performed using both traditional computer aided design (CAD) methods and the T-DDIN deep learning approach. The two methods were compared based on three-dimensional deviation, occlusal adjustment volumes, cusp angle deviation, and restoration time. Results: The T-DDIN group demonstrated significantly lower three-dimensional deviation for Class Ⅱ cavity defects and occlusal wear restoration [(0.14±0.05) and (0.16±0.09) mm], occlusal adjustment volumes [(0.44±0.03) and (0.49±0.03) mm3], and difference value of the tooth cusp angles (5.69°±1.90° and 6.04°±0.53°) compared to the traditional CAD group (both P<0.001). No significant differences were observed within the T-DDIN group between the two defect types in terms of three-dimensional deviation (P=0.098) or occlusal adjustment volume (P=0.154) or difference value of the tooth cusp angles (P=0.196). However, in the traditional CAD group, three-dimensional deviation, occlusal adjustment volume and difference value of the tooth cusp angles was significantly higher in occlusal wear restorations than in Class Ⅱ cavity defects restorations (P<0.001). The T-DDIN group, which involved Class Ⅱ cavity defects and occlusal wear, demonstrated significantly less recovery time of morphology (37.2±7.7) and (39.4±6.2) s compared to the traditional CAD group (P<0.001). Conclusions: T-DDIN demonstrated superior stability and accuracy in morphological reconstruction for various types of dental defects while significantly reducing restoration time. 目的: 通过融合隐式模板与深度学习技术,构建全新可变形牙齿深度隐式网络(tooth-deformable deep implicit network,T-DDIN),以评估不同缺损条件下T-DDIN形态重建的精度、稳定性及效率。 方法: 收集2022年3月至2024年3月四川大学华西口腔医院正畸科和修复科收治的550例牙体缺损或错(牙合)畸形患者的口内扫描模型(训练集500例,测试集50例)。研发T-DDIN,通过隐式模板与潜在编码预测网络实现缺损牙体的形态重建。模型性能评估阶段,对测试集数据模拟Ⅱ类洞缺损和(牙合)面磨耗。采用传统计算机辅助设计(CAD)和T-DDIN深度学习模型分别进行形态重建(分别为传统CAD组和T-DDIN组),以原始牙为真值,计算两组下颌右侧第一磨牙形态重建后与真值的三维偏差以及调(牙合)体积、牙尖角差值和形态重建时间,综合评价两组差异。 结果: T-DDIN组Ⅱ类洞缺损与(牙合)面磨耗重建的三维偏差[分别为(0.14±0.05)、(0.16±0.09)mm]、调(牙合)体积[分别为(0.44±0.03)、(0.49±0.03)mm3]及牙尖角差值(分别为5.69°±1.90°、6.04°±0.53°)均显著小于传统CAD组相应缺损重建(均P<0.001)。T-DDIN组两类缺损重建的三维偏差(P=0.098)、调(牙合)体积(P=0.154)及牙尖角差值(P=0.196)的组内差异均无统计学意义,而传统CAD组(牙合)面磨耗重建的三维偏差、调(牙合)体积与牙尖角差值均显著大于Ⅱ类洞缺损重建(均P<0.001)。T-DDIN组两类缺损的形态重建时间分别为(37.2±7.7)和(39.4±6.2)s,均显著少于传统CAD组相应缺损重建(均P<0.001)。 结论: 针对Ⅱ类洞缺损和(牙合)面磨耗形态重建,T-DDIN均表现出优异的形态重建稳定性、精度和效率。.
引用
- Yu MQ, Chen D, Wang ZY, et al. [A preliminary exploration of an intelligent system for personalized tooth morphology reconstruction based on deep learning]. Zhonghua Kou Qiang Yi Xue Za Zhi. 2025. PMID: 40443316. doi:10.3760/cma.j.cn112144-20250331-00110
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。