| 摘要: | Objective
The objective of this thesis is to investigate the factors influencing dental, skeletal, and soft tissue morphology changes following orthognathic surgery among individuals with skeletal Class III using a machine learning approach.
Material and methods
This retrospective study was conducted in the Department of Orthodontics at Taipei Medical University Hospital, Taipei, Taiwan, involving lateral cephalometric radiographs of 58 skeletal Class III malocclusion patients who underwent orthognathic surgery. All samples, consisting of pre-treatment and post-treatment records, were obtained from the hospital patient database. All cephalograms were traced manually and verified by two orthodontists prior to anatomical landmark identification and measurement. The measurements were done using a web-based cephalometric measurement software and include dental, skeletal and soft tissue measurements. The measurements calculated the cephalometric values for pre-treatment (T0), post-treatment (T1), and the changes (T1-T0). Following this, the data were processed and analyzed using SPSS and machine learning to investigate the affecting factors, which are described in the feature of importance (FI).
Result
All cephalometric variables analyzed had significant changes from T0 (pre-treatment) to T1 (post-treatment), except for SNA, A to NP, Overbite, and Lower lip to E-plane. It was also found that ANB is significantly influenced by surgery type (p-value=0.045), while IMPA and Lower lip to E-plane are significantly affected by gender (IMPA p-value=0.029; Lower lip to E-plane p-value=0.033). When considering the factors influenced by the pre-treatment condition (T0), Overjet appears to play a dominant role in several dependent variables, including changes in ANB(FI=0.226), B to N-Perp (FH)(FI=0.259), and Pog to N-Perp (FH)(FI=0.257). Overjet(FI=0.908), Overbite(FI=0.852) and L1 to NB (deg)(FI=0.388) are self-referential. In addition, the affecting factors among the cephalometric changes value (T1-T0) seem to be self-reliant in general.
Conclusion
Utilizing machine learning to consider factors influenced by the pre-treatment condition (T0), Overjet appears to play a dominant role in several dependent variables, including changes in ANB, B to N-Perp (FH), and Pog to N-Perp (FH). Overjet, Overbite, and L1 to NB (deg) are self-referential. Additionally, the factors affecting cephalometric changes (T1-T0) generally seem to be self-reliant. To address the limitations of this study, future research should involve a larger dataset and incorporate 3D examinations. |
