[Deep learning-based assessment of periodontal ligament area changes in maxillary central incisors under different...

Objective: To investigate the differences in the changes of periodontal ligament area (PDLA) and related clinical indicators before and after maxillary central incisor movement under different orthodontic treatment regimens. Methods: This study was designed as a retrospective case-control trial. A total of 49 orthodontic patients (98 maxillary central incisors) who received the treatment in the Department of Stomatology, Beijing Friendship Hospital, Capital Medical University, from May 2021 to December 2022 were enrolled. The cone beam CT (CBCT) images and medical records before and after treatment were collected. Patients were divided into two groups based on extraction or anchorage type: extraction group (52 incisors) and non-extraction group (46 incisors), implant anchorage group (41 incisors) and non-implant anchorage group (57 incisors). The PDLA of the bilateral maxillary central incisors before and after treatment in each group were measured using the CBCT periodontal ligament segmentation network. Orthodontic measurement and analysis software were adopted to determine the root length, palatal displacement of the incisal edge and apex, and the inclination angle of maxillary central incisors (U1-NA angle). The reduction in root length (ΔRL) and its percentage (ΔRL%), the reduction in PDLA (ΔPDLA) and its percentage (ΔPDLA%), as well as the reduction in U1-NA angle (ΔU1-NA angle) after orthodontic treatment were calculated. The Mann-Whitney U test was used to compare the differences in each measurement index among the groups. Spearman's correlation analysis was performed to explore the correlations between the ΔPDLA and the palatal displacement of the incisal edge, ΔU1-NA angle, and ΔRL of the maxillary central incisors. Results: After orthodontic treatment, the ΔRL, ΔRL% and ΔPDLA in the extraction group were 0.82 (0.85) mm, 6.34% (7.29%) and 9.89 (15.98) mm², respectively, showing significant differences compared with those in the non-extraction group [0.45 (0.71) mm, 3.89% (7.99%), 5.46 (8.65) mm²] (all P<0.05). Whereas there was no statistically significant difference in ΔPDLA% between the extraction group and the non-extraction group (P>0.05). No statistically significant differences were observed in ΔRL, ΔRL%, ΔPDLA, or ΔPDLA% between the implant anchorage group and the non-implant anchorage group (P>0.05). The palatal displacement of the incisal edge and ΔU1-NA angle of the maxillary central incisors in the extraction group were 2.90 (4.05) mm and 6.69°±7.90°, respectively, which were significantly greater than those in the non-extraction group [-0.25 (3.28) mm, -1.60°±9.17°] (both P<0.05). The palatal displacement of the incisal edge [2.70 (4.95) mm] in the implant anchorage group were also significantly larger than those in the non-implant anchorage group [1.00 (3.25) mm] (P<0.05). The ΔPDLA was significantly positively correlated with the palatal displacement of the incisal edge (r=0.44, P<0.001), ΔRL (r=0.36, P<0.001), and ΔU1-NA angle (r=0.42, P<0.001), with the strongest correlation observed between ΔPDLA and the palatal displacement of incisal edge. Conclusions: Although extraction treatment increased root resorption during the palatal movement of maxillary central incisors, it did not significantly affect the PDLA of these teeth. Implant anchorage did not increase the risk of the root resorption or PDLA reduction during the palatal movement of maxillary central incisors. Compared with the measurement of RL, the application of the CBCT periodontal ligament segmentation network for PDLA measurement enabled a more comprehensive evaluation of changes in the periodontal supporting tissues of the maxillary central incisors.