Accuracy in predicting soft tissue changes of orthodontic class III cases using Dolphin® software.

OBJECTIVES: The prediction of posttreatment outcomes is conducive to the final determination of ideal therapeutic options. However, the prediction accuracy in orthodontic class III cases is unclear. Therefore, this study conducted exploration on prediction accuracy in orthodontic class III patients using the Dolphin® software. MATERIALS AND METHODS: In this retrospective study, lateral cephalometric radiographs of pre- and posttreatment were collected from 28 angle class III adults who received completed non-orthognathic orthodontic therapy (8 males, 20 females; mean age = 20.89 ± 4.26 years). The values of 7 posttreatment parameters were recorded and inserted into the Dolphin® Imaging software to generate a predicted outcome, and then the prediction radiograph and actual posttreatment radiograph were superimposed and compared in terms of soft tissue parameters and landmarks. RESULTS: The prediction showed significant differences with the actual outcomes in nasal prominence (the difference between the prediction and the actual value was - 0.78 ± 1.82 mm), the distance from the lower lip to the H line (0.55 ± 1.11 mm), and the distance from the lower lip to the E line (0.77 ± 1.62 mm) (p < 0.05). Point subnasale (Sn) (an accuracy of 92.86% in the horizontal direction and 100% in the vertical direction in 2 mm) and point soft tissue A (ST A) (an accuracy of 92.86% in the horizontal direction and 85.71% in the vertical direction in 2 mm) were proven to be the most accurate landmarks, while the predictions in the chin region were relatively inaccurate. Furthermore, the predictions in the vertical direction were of higher accuracy compared to the horizontal direction except for the points around the chin. CONCLUSIONS: The Dolphin® software demonstrated acceptable prediction accuracy in midfacial changes in class III patients. However, there were still limitations for changes in the chin and lower lip prominence. CLINICAL RELEVANCE: Clarifying the accuracy of Dolphin® software in predicting soft tissue changes of orthodontic class III cases will facilitate physician-patient communication and clinical treatment.

Orthodontic emergencies and mental state of Chinese orthodontic patients during the COVID-19 pandemic.

BACKGROUND: The ongoing COVID-19 pandemic postponed routine follow-up visits of many orthodontic patients, which compromised their treatment process and mental states. This study was aimed to assess orthodontic emergency occurrence and psychological states of Chinese orthodontic patients during this pandemic. METHODS: Orthodontic patients in China were invited to answer an anonymous online questionnaire from February 20, 2020 to March 5, 2020, when routine dental care was suspended in China. The questionnaire included self-assessment of oral hygiene and compliance, orthodontic emergencies, perceptions and feelings about COVID-19 and anxiety self-rating scale, etc. Collected data was statistically analyzed with Chi-square, independent t test and univariable generalized estimating equations regression analysis. RESULTS: A total of 1078 respondents (292 male; 786 female) from 30 provinces of China were included in this study. About one-third (33.67%) of patients reported that they encountered orthodontic problems during the pandemic. Patients with clear aligners reported fewer orthodontic problems than those with fixed appliances or removable appliances. Female patients, elder patients and patients who encountered orthodontic emergencies were more anxious than other patients. CONCLUSIONS: The compliance and occurrence of orthodontic emergencies differed in patients with different orthodontic appliances. Patients with orthodontic emergencies exhibited higher anxiety states.

Artificial intelligence system for automated landmark localization and analysis of cephalometry.

OBJECTIVES: Cephalometric analysis is essential for diagnosis, treatment planning and outcome assessment of orthodontics and orthognathic surgery. Utilizing artificial intelligence (AI) to achieve automated landmark localization has proved feasible and convenient. However, current systems remain insufficient for clinical application, as patients exhibit various malocclusions in cephalograms produced by different manufacturers while limited cephalograms were applied to train AI in these systems. METHODS: A robust and clinically applicable AI system was proposed for automatic cephalometric analysis. First, 9870 cephalograms taken by different radiography machines with various malocclusions of patients were collected from 20 medical institutions. Then 30 landmarks of all these cephalogram samples were manually annotated to train an AI system, composed of a two-stage convolutional neural network and a software-as-a-service system. Further, more than 100 orthodontists participated to refine the AI-output landmark localizations and retrain this system. RESULTS: The average landmark prediction error of this system was as low as 0.94 ± 0.74 mm and the system achieved an average classification accuracy of 89.33%. CONCLUSIONS: An automatic cephalometric analysis system based on convolutional neural network was proposed, which can realize automatic landmark location and cephalometric measurements classification. This system showed promise in improving diagnostic efficiency in clinical circumstances.