To support dentists with limited experience, this study trained and compared six convolutional neural networks to detect crossbites and classify non-crossbite, frontal, and lateral crossbites using 2D intraoral photographs. Based on 676 photographs from 311 orthodontic patients, six convolutional neural network models were trained and compared to classify (1) non-crossbite vs. crossbite and (2) non-crossbite vs. lateral crossbite vs. frontal crossbite. The trained models comprised DenseNet, EfficientNet, MobileNet, ResNet18, ResNet50, and Xception. Among the models, Xception showed the highest accuracy (98.57%) in the test dataset for classifying non-crossbite vs. crossbite images. When additionally distinguishing between lateral and frontal crossbites, average accuracy decreased with the DenseNet architecture achieving the highest accuracy among the models with 91.43% in the test dataset. Convolutional neural networks show high potential in processing clinical photographs and detecting crossbites. This study provides initial insights into how deep learning models can be used for orthodontic diagnosis of malocclusions based on intraoral 2D photographs.

Background: Physiotherapy appears as a promising therapy option for patients with Juvenile Idiopathic Arthritis (JIA) [1, 2], but the effects of physiotherapy and jaw exercises on JIA-related orofacial symptoms remain unknown [3]. The aim of this proof-of-concept study was to assess the impact of orofacial physiotherapy and home-exercise programs in patients with JIA and temporomandibular joint (TMJ) involvement. Methods: Twelve patients with JIA and TMJ involvement received a treatment of physiotherapy, complemented by prescribed home exercises spanning over eight weeks. Orofacial symptoms and dysfunction were monitored pre-treatment, during treatment, after treatment, and at a three-months follow-up. Results: Orofacial pain frequency and intensity significantly decreased during the course of the treatment (p = 0.009 and p = 0.006), with further reductions observed at the three-month follow-up (p = 0.007 and p = 0.002). During treatment, the mandibular function improved significantly in terms of maximal mouth opening capacity, laterotrusion, and protrusion. Conclusions: This proof-of-concept study shows favourable effects of physiotherapy and home excercises in the management of JIA-related orofacial symptoms and dysfunctions.

Continuous advancements in technology have made it possible to integrate clear aligner therapy (CAT) with orthognathic surgery. This case report presents a novel, individually-planned workflow, combining CAT with a surgery-first orthognathic approach (SFOA) in collaborating with engineers for an in-house production of surgical guides and customized titanium plates. The patient was evaluated subjectively, using the Oral Health-Related Quality of Life-14 (OHIP-14) questionnaire and Orthognathic Quality of Life questionnaire (OQLQ), and objectively with the Peer Assessment Rating (PAR) index. The patient displayed the planned occlusal relationship with no report of discomfort in the temporomandibular joint (TMJ) or post-surgical complications. The surgical and occlusal outcomes have remained consistent and stable after debonding. A decreased score was reported in both questionnaires and the PAR after treatment, thereby indicating improvements in both subjective and objective evaluations. This case report demonstrates that with proper individual planning, satisfactory subjective and objective outcomes can be achieved when combining SFOA with CAT.

This investigation aimed to develop a radiographic 3D cephalometric index to grade severity of dentofacial deformity in patients with juvenile idiopathic arthritis (JIA), and to perform a validation against expert evaluations. Data were collected from a population-based Nordic JIA cohort of 240 patients that received a cone-beam computed tomography (CBCT) scan approximately 17 years after onset of JIA. The cohort was randomized into two groups: A baseline group for establishing the index (n = 210) and a test group (n = 30). The standardized index generation process led to the inclusion of six radiographic outcome measures, each representing different facets of dentofacial deformities. Specific severity scores were established for three domains; dentofacial asymmetry (0–8), vertical deviation (0–8), and sagittal deviation (0–8). An overall dentofacial deformity index score was computed by aggregating the scores from the six variables, with the total score ranging from 0 to 24. The validity of the index was assessed by comparing the index score with expert evaluations. Based on the generated index, 10 patients exhibited severe deformities. Reliable agreement between data-generated index scores and expert assessments was found. A 3D radiographic dentofacial deformity index was generated and tested. This index can support an objective assessment of the severity of dentofacial deformities related to JIA in a clinical setting and future research.

Introduction: Management of dentofacial deficiencies requires knowledge about sutural morphology and complexity. The present study assesses midpalatal sutural morphology based on human cone-beam computed tomography (CBCT) using geometric morphometrics (GMM) and complexity scores. The study is the first to apply a sutural complexity score to human CBCT datasets and demonstrates the potential such a score has to improve objectiveness and comparability when analysing the midpalatal suture. Materials and methods: CBCTs of various age and sex groups were analysed retrospectively (n = 48). For the geometric morphometric analysis, landmark acquisition and generalised Procrustes superimposition were combined with principal component analysis to detect variability in sutural shape patterns. For complexity analysis, a windowed short-time Fourier transform with a power spectrum density (PSD) calculation was applied to resampled superimposed semi-landmarks. Results: According to the GMM, younger patients exhibited comparable sutural patterns. With increasing age, the shape variation increased among the samples. The principal components did not sufficiently capture complexity patterns, so an additional methodology was applied to assess characteristics such as sutural interdigitation. According to the complexity analysis, the average PSD complexity score was 1.465 (standard deviation = 0.010). Suture complexity increased with patient age (p < 0.0001), but was not influenced by sex (p = 0.588). The intra-class correlation coefficient exceeded 0.9, indicating intra-rater reliability. Conclusion: Our study demonstrated that GMM applied to human CBCTs can reveal shape variations and allow the comparison of sutural morphologies across samples. We demonstrate that complexity scores can be applied to study human sutures captured in CBCTs and complement GMM for a comprehensive sutural analysis.