Department of Orthodontics, Faculty of Dentistry, Alexandria University, Champollion Street, El Azarita, Alexandria, Egypt.
Department of Orthodontics, Saveetha Dental College, Saveetha Insitute of Medical and Technical Sciences, Chennai, India.
Prog Orthod. 2022 Mar 14;23(1):8. doi: 10.1186/s40510-022-00402-x.
To investigate the accuracy of three different 3D digital model registration software for tip, torque and rotation measurements, with reference to a 3D digital virtual setup. Twenty maxillary and mandibular pre-treatment scans of patients undergoing clear aligner therapy were used. Digital setups were generated from pre-treatment scans using a tooth movement software. Both the pretreatment digital scans (T1) and digital setups (T2) were converted to STL files to be exported to the 3 studied software that employed: (1) Semiautomatic best fit registration (S-BF), (2) Interactive surface-based registration (I-SB), and (3) Automatic best fit registration (A-BF) respectively. Changes in tip, torque and rotation were calculated for all the registered pairs.
The change in tooth position was compared between the calculated tooth movement using each of the registration software packages versus the actual generated tooth movement from the digital setups. Continuous data was expressed as mean and standard deviation. Intra Class Correlation Coefficient for agreement between digital simulation and each software was used. Intra and Inter-examiner reliabilities were also assessed using Intra Class Correlation Coefficient. Significance of the obtained results was expressed at p ≤ 0.01. Semiautomatic best fit registration software showed excellent agreement (> 0.90) for all tooth movements, except for good agreement for torque (0.808). Interactive surface-based registration software showed moderate agreement for all measurements (0.50 and < 0.75), except for good agreement for rotation (0.783). Automatic best fit registration software demonstrated excellent agreement (> 0.90) for rotation, good agreement for tip (0.890) and moderate agreement for torque (0.740).
Overall, semiautomatic best fit registration software consistently showed excellent agreement in superimpositions compared to other software types. Automatic best fit registration software consistently demonstrated better agreement for mandibular superimpositions, compared to others. Accuracy of digital model superimpositions for tooth movements studied in superimposition studies, can be attributed to the algorithm employed for quantification.
为了研究三种不同的 3D 数字模型配准软件在尖端、扭矩和旋转测量方面的准确性，本研究以 3D 数字虚拟设置为参考。使用 20 名接受透明矫正治疗的上颌和下颌患者的预处理扫描。使用牙齿移动软件从预处理扫描生成数字设置。将预处理数字扫描（T1）和数字设置（T2）都转换为 STL 文件，以便输出到研究中使用的 3 种软件，分别为：（1）半自动最佳拟合配准（S-BF），（2）交互式基于表面的配准（I-SB），以及（3）自动最佳拟合配准（A-BF）。为所有配准对计算了尖端、扭矩和旋转的变化。
比较了使用每种配准软件包计算的牙齿位置变化与数字设置中实际生成的牙齿运动之间的差异。连续数据以平均值和标准差表示。使用组内相关系数评估数字模拟与每种软件之间的一致性。还使用组内相关系数评估了内部和外部检查者的可靠性。表示结果的显著性为 p ≤ 0.01。半自动最佳拟合配准软件显示出对所有牙齿运动的极好一致性（> 0.90），除了扭矩的良好一致性（0.808）。交互式基于表面的配准软件对所有测量值显示出中度一致性（0.50 和 < 0.75），除了旋转的良好一致性（0.783）。自动最佳拟合配准软件显示出对旋转的极好一致性（> 0.90），对尖端的良好一致性（0.890）和对扭矩的中度一致性（0.740）。
总体而言，与其他软件类型相比，半自动最佳拟合配准软件在叠加中始终显示出极好的一致性。与其他软件相比，自动最佳拟合配准软件在对下颌叠加的一致性方面表现更好。叠加研究中用于牙齿运动的数字模型叠加的准确性可归因于用于量化的算法。
