Establishment of a novel classification system for alveolar morphology in infants with unilateral complete cleft lip and palate.

OBJECTIVES: Unilateral complete cleft lip and palate (UCCLP) is one of the most severe clinical subtypes among cleft lip and palate (CLP), making repair surgery and subsequent orthodontic treatment particularly challenging. Presurgical nasoalveolar molding (PNAM) has shown conflicting and heterogeneous results in the treatment of UCCLP patients, raising questions about whether the diversity in alveolar anatomical morphology among these patients plays a role in the effectiveness of PNAM treatment. MATERIALS AND METHODS: We collected 90 digital maxillary models of infants with UCCLP and performed mathematical clustering analysis, including principal component analysis (PCA), decision tree modeling, and area under the ROC Curve (AUC) analysis, to classify alveolar morphology and identify key measurements. We also conducted clinical evaluations to assess the association between the alveolar morphology and CLP treatment outcomes. RESULTS: Using mathematical clustering analysis, we classified the alveolar morphology into three distinct types: average form, horizontal form, and longitudinal form. The decision tree model, AUC analysis, and comparison analysis revealed that four measurements (Trans ACG-ACL, ML length, MG length and Inc length) were essential for clustering the alveolar morphology of infants with UCCLP. Furthermore, the blinded clinical evaluation indicated that UCCLP patients with alveolar segments of horizontal form had the lowest treatment outcomes. CONCLUSION: Overall, our findings establish a novel quantitative classification system for the morphology of alveolar bone in infants with UCCLP and suggest that this classification may be associated with the outcomes of CLP treatment. CLINICAL RELEVANCE: The multidisciplinary CLP team should thoroughly evaluate and classify the specific alveolar morphology when administering PNAM to infants with UCCLP.

Lip position analysis of young women with different skeletal patterns during posed smiling using 3-dimensional stereophotogrammetry.

INTRODUCTION: The aim of this study was to explore the internal relationship between posed smile characteristics, lip position, and skeletal patterns in young women. METHODS: Fifty women between the ages of 20 and 30 years were enrolled and divided into 3 groups-vertical, average, and horizontal patterns- using the following parameters: FMA, GoGn-SN, and Jarabak ratio. Each subject was scanned in natural head position and with a posed smile. The interlabial gap, intercommissural width, and smile index were calculated. The frontal region was selected as the reference plane for superimpositions. The changes of the lip landmarks in the vertical, sagittal, and coronal directions were investigated. RESULTS: The smile indexes were listed in the following sequence: vertical < average < horizontal. Significant differences were found in the interlabial gap among the 3 groups. Compared with the average and horizontal groups, the upper lip landmarks of the vertical group showed differences and changed more only in the vertical direction. However, the lower lip landmark showed no differences in any direction. CONCLUSIONS: Different skeletal patterns have characteristic smile features. The vertical skeletal pattern affects upper lip movements because there is more space for upper-lip elevation. However, the vertical skeletal pattern has no effect on lower lip movement.

Quantitative changes of upper airway in class III patients undergoing bimaxillary surgery after one-year follow-up: a retrospective study.

BACKGROUND: Bimaxillary surgery is often performed for class III malocclusion, and its complex influence on the upper airway has been well considered. The aim of this research was to provide a scaled formula between upper airway volume changes and bone movements in Class III patients after orthognathic surgery. MATERIALS AND METHODS: Using a retrospective study design, the investigators enrolled a total of 30 class III malocclusion patients who were undergoing bimaxillary surgery as the study subjects. The subjects included 15 males and 15 females, and their average age was 23.3 ± 3.4 years. CBCT (cone beam tomography) was performed both before and one year after the surgery for each patient. The changes in the soft palate, tongue and upper airway were measured by using CBCT data that was collected before and after surgery. 3D superimposition of CBCT was performed to calculate three-dimensional jaw movements. A multiple regression analysis was used to calculate the quantitative relationship between airway volume changes and jaw movements. RESULTS: The nasopharynx airway volume was observed to be increased by 1064.0 ± 1336.2 mm3, whereas the retropalatal and retroglossal airway volumes were observed to be decreased by 1399.0 ± 2881.6 mm3 and 1433.8 ± 3043.4 mm3, respectively, after the surgery. One millimetre forward and downward movements of the PNS resulted in increases of 626.90 mm3 and 392.18 mm3 in nasopharynx airway volume, respectively. Moreover, one millimetre retrogression of the B point caused decreases of 314.6 mm3 and 656.6 mm3 in the retropalatal and retroglossal airway volume, respectively. The changes in the soft palate contributed to the decrease in the retropalatal airway volume, whereas the tongue compensated for the decrease in the retroglossal airway volume. CONCLUSION: The movements of the PNS and B points could be used to predict upper airway volumetric changes in Class III patients after maxillary advancement and mandibular setback.

Automatic Classification for Sagittal Craniofacial Patterns Based on Different Convolutional Neural Networks.

(1) Background: The present study aims to evaluate and compare the model performances of different convolutional neural networks (CNNs) used for classifying sagittal skeletal patterns. (2) Methods: A total of 2432 lateral cephalometric radiographs were collected. They were labeled as Class I, Class II, and Class III patterns, according to their ANB angles and Wits values. The radiographs were randomly divided into the training, validation, and test sets in the ratio of 70%:15%:15%. Four different CNNs, namely VGG16, GoogLeNet, ResNet152, and DenseNet161, were trained, and their model performances were compared. (3) Results: The accuracy of the four CNNs was ranked as follows: DenseNet161 > ResNet152 > VGG16 > GoogLeNet. DenseNet161 had the highest accuracy, while GoogLeNet possessed the smallest model size and fastest inference speed. The CNNs showed better capabilities for identifying Class III patterns, followed by Classes II and I. Most of the samples that were misclassified by the CNNs were boundary cases. The activation area confirmed the CNNs without overfitting and indicated that artificial intelligence could recognize the compensatory dental features in the anterior region of the jaws and lips. (4) Conclusions: CNNs can quickly and effectively assist orthodontists in the diagnosis of sagittal skeletal classification patterns.

Convolutional neural network-based automatic cervical vertebral maturation classification method.

OBJECTIVES: This study aimed to develop a fully automated artificial intelligence-aided cervical vertebral maturation (CVM) classification method based on convolutional neural networks (CNNs) to provide an auxiliary diagnosis for orthodontists. METHODS: This study consisted of cephalometric images from patients aged between 5 and 18 years. After grouping them into six cervical stages (CSs) by orthodontists, a data set was constructed for analyzing CVM using CNNs. The data set was divided into training, validation, and test sets in the ratio of 70, 15, and 15%. Four CNN models namely, VGG16, GoogLeNet, DenseNet161, and ResNet152 were selected as the candidate models. After training and validation, the models were evaluated to determine which of them is most suitable for CVM analysis. Heat maps were analyzed for a deeper understanding of what the CNNs had learned. RESULTS: The final classification accuracy ranking was ResNet152>DenseNet161>GoogLeNet>VGG16, as evaluated on the test set. ResNet152 proved to be the best model among the four models for CVM classification with a weighted κ of 0.826, an average AUC of 0.933 and total accuracy of 67.06%. The F1 score rank for each subgroup was: CS6>CS1>CS4>CS5>CS3>CS2. The area of the third (C3) and fourth (C4) cervical vertebrae were activated when CNNs were assessing the images. CONCLUSION: CNN models proved to be a convenient, fast and reliable method for CVM analysis. CNN models have the potential to provide automatic auxiliary diagnostic tools in the future.