ABSTRACT
Objective@#The purpose of this study was to investigate the accuracy of one-step automated orthodontic diagnosis of skeletodental discrepancies using a convolutional neural network (CNN) and lateral cephalogram images with different qualities from nationwide multi-hospitals. @*Methods@#Among 2,174 lateral cephalograms, 1,993 cephalograms from two hospitals were used for training and internal test sets and 181 cephalograms from eight other hospitals were used for an external test set. They were divided into three classification groups according to anteroposterior skeletal discrepancies (Class I, II, and III), vertical skeletal discrepancies (normodivergent, hypodivergent, and hyperdivergent patterns), and vertical dental discrepancies (normal overbite, deep bite, and open bite) as a gold standard. Pre-trained DenseNet-169 was used as a CNN classifier model. Diagnostic performance was evaluated by receiver operating characteristic (ROC) analysis, t-stochastic neighbor embedding (t-SNE), and gradientweighted class activation mapping (Grad-CAM). @*Results@#In the ROC analysis, the mean area under the curve and the mean accuracy of all classifications were high with both internal and external test sets (all, > 0.89 and > 0.80). In the t-SNE analysis, our model succeeded in creating good separation between three classification groups. Grad-CAM figures showed differences in the location and size of the focus areas between three classification groups in each diagnosis. @*Conclusions@#Since the accuracy of our model was validated with both internal and external test sets, it shows the possible usefulness of a one-step automated orthodontic diagnosis tool using a CNN model. However, it still needs technical improvement in terms of classifying vertical dental discrepancies.
ABSTRACT
OBJECTIVE@#To investigate the distribution and phenotypes of hemifacial microsomia (HFM) and its association with other anomalies.@*METHODS@#This study included 249 Korean patients with HFM, whose charts, photographs, radiographs, and/or computed tomography scans acquired during 1998–2018 were available from Seoul National University Hospital and Dental Hospital. Prevalence according to sex, side involvement, degree of mandibular deformity, compensatory growth of the mandibular body, and Angle's classification, and its association with other anomalies were statistically analyzed.@*RESULTS@#Prevalence was not different between male and female patients (55.0% vs. 45.0%, p > 0.05). Unilateral HFM (UHFM) was more prevalent than bilateral HFM (BHFM) (86.3% vs. 13.7%, p 0.05). Among patients with BHFM, prevalence of different Pruzansky–Kaban types on the right and left sides was greater than that of the same type on both sides (67.6% vs. 32.4%, p < 0.05). Despite hypoplasia of the condyle/ramus complex, compensatory growth of the mandibular body on the ipsilateral side occurred in 35 patients (14.1%). Class I and II molar relationships were more prevalent than Class III molar relationships (93.2% vs. 6.8%, p < 0.001). Forty-eight patients (19.3%) had other anomalies, with 50.0% and 14.4% in the BHFM and UHFM groups (p < 0.001).@*CONCLUSIONS@#Patients with HFM require individualized diagnosis and treatment planning because of diverse phenotypes and associations with other anomalies.
ABSTRACT
Objective@#To investigate the effects of the long-term use of a maxillary protraction facemask with miniplate (FM-MP) on pharyngeal airway dimensions in growing patients with cleft lip and palate (CLP). @*Methods@#The study included 24 boys with CLP (mean age, 12.2 years; mean duration of FM-MP therapy, 4.9 years), divided into two groups according to the amount of A point advancement to the vertical reference plane (VRP): Group 1, > 4 mm; Group 2, < 2 mm; n = 12/group. After evaluating the skeletodental and airway variables using lateral cephalograms acquired before and after FM-MP therapy, statistical analyses were performed. @*Results@#Group 1 showed greater forward and downward displacements of the posterior maxilla (posterior nasal spine [PNS]-horizontal reference plane [HRP]; PNSVRP), greater increase in ANB, more forward tongue position (tongue tip-Pt vertical line to Frankfort horizontal plane), and greater increase in the oropharynx (superior posterior airway space [SPAS]; middle airway space [MAS]) and upper nasopharynx (PNS-adenoid2) than did Group 2. While maxillary advancement (A-VRP and PNS-VRP) correlated with increases in SPAS, MAS, and PNS-adenoid2, downward displacement of the PNS (PNS-HRP) correlated with increases in SPAS, MAS, PNSadenoid1, and PNS-adenoid2, and with a decrease in vertical airway length (VAL). Mandibular forward displacement and decrease in mandibular plane correlated with increases in MAS. @*Conclusions@#FM-MP therapy had positive effects on the oropharyngeal and nasopharyngeal airway spaces without increases in VAL in Group 1 rather than in Group 2. However, further validation using an untreated control group is necessary.
ABSTRACT
Objective@#To investigate the treatment modalities (Tx-Mods) for patients with unilateral hemifacial microsomia (UHFM) according to Pruzansky–Kaban types and growth stages. @*Methods@#The samples consisted of 82 Korean UHFM patients. Tx-Mods were defined as follows: Tx-Mod-1, growth observation due to mild facial asymmetry; Tx-Mod-2, unilateral functional appliance; TxMod-3, fixed orthodontic treatment; Tx-Mod-4, growth observation due to a definite need for surgical intervention; Tx-Mod-5, unilateral mandibular or bimaxillary distraction osteogenesis (DO); Tx-Mod-6, maxillary fixation using LeFort I osteotomy and mandibular DO/sagittal split ramus osteotomy; TxMod-7, orthognathic surgery; and Tx-Mod-8, costochondral grafting. The type and frequency of Tx-Mod, the number of patients who underwent surgical procedures, and the number of surgeries that each patient underwent, were investigated. Results: The degree of invasiveness and complexity of Tx-Mod increased, with an increase in treatment stage and Pruzansky–Kaban type (initial 0.05). @*Conclusions@#These findings might be used as basic guidelines for successful treatment planning and prognosis prediction in UHFM patients.