Division of Orthodontics, College of Dentistry, Ohio State University, Columbus, Ohio.
Division of Biostatistics, The Ohio State University College of Public health, Columbus, Ohio.
Am J Orthod Dentofacial Orthop. 2023 Jul;164(1):24-33. doi: 10.1016/j.ajodo.2022.10.024. Epub 2023 Feb 15.
This research project aimed to compare the number of maxillary incisors and canine movement between Invisalign and fixed orthodontic appliances using artificial intelligence and identify any limitations of Invisalign.
Sixty patients (Invisalign, n = 30; braces, n = 30) were randomly selected from the Ohio State University Graduate Orthodontic Clinic archive. Peer Assessment Rating (PAR) analysis was used to indicate the severity of the patients in both groups. To analyze the incisors and canine movement, specific landmarks were identified on incisors and canines using an artificial intelligence framework, two-stage mesh deep learning. Total average tooth movement in the maxilla and individual (incisors and canine) tooth movement in 6 directions (buccolingual, mesiodistal, vertical, tipping, torque, rotation) were then analyzed at a significance level of α = 0.05.
Based on the posttreatment Peer Assessment Rating scores, the quality of finished patients in both groups was similar. In maxillary incisors and canines, there was a significant difference in movement between Invisalign and conventional appliances for all 6 movement directions (P <0.05). The greatest differences were with rotation and tipping of the maxillary canine, along with incisor and canine torque. The smallest statistical differences observed for incisors and canines were crown translational tooth movement in the mesiodistal and buccolingual directions.
When comparing fixed orthodontic appliances to Invisalign, patients treated with fixed appliances were found to have significantly more maxillary tooth movement in all directions, especially with rotation and tipping of the maxillary canine.
本研究项目旨在比较 Invisalign 和固定正畸矫治器在使用人工智能时上颌切牙和犬牙的移动数量，并确定 Invisalign 的任何局限性。
从俄亥俄州立大学研究生正畸诊所档案中随机选择 60 名患者（Invisalign，n=30；牙套，n=30）。使用同行评估评分（PAR）分析来表示两组患者的严重程度。为了分析切牙和犬牙的移动，使用人工智能框架（两阶段网格深度学习）在上颌切牙和犬牙上识别特定的标志点。然后，在 6 个方向（颊舌向、近远中向、垂直向、倾斜、转矩、旋转）分析上颌牙齿和个别（切牙和犬牙）牙齿的总平均牙齿移动，显著水平设为α=0.05。
根据治疗后的同行评估评分，两组患者的治疗效果质量相似。在上颌切牙和犬牙中，Invisalign 和传统矫治器在所有 6 个运动方向上的运动差异均具有统计学意义（P<0.05）。最大的差异是上颌犬牙的旋转和倾斜，以及切牙和犬牙的转矩。在上颌切牙和犬牙中，观察到的最小统计学差异是近远中向和颊舌向的冠向牙齿移动。
与固定正畸矫治器相比，使用固定矫治器的患者在所有方向上的上颌牙齿移动明显更多，尤其是上颌犬牙的旋转和倾斜。
