Influence of shift work on periodontitis according to the occupation group.

This study aimed to investigate the effects of shift work on periodontal disease in blue-and white-collar workers and to examine the interaction effects between occupation and work patterns. Data were collected from the Korea National Health and Nutrition Examination Survey conducted by the Korean Ministry of Health and Welfare for a total of nine years from 2007 to 2015. Participants with missing outcome variables were excluded from the analysis and a total of 32,336 participants were included in the final analysis. Univariable odds ratios (OR) were calculated using a logistic regression model with 95% confidence interval (CI). A multivariable logistic regression analysis was performed using the backward elimination method. The CONTRAST statement was used to analyze the interaction effect between occupation and work patterns. Multivariable logistic regression analysis revealed that interaction effects are present between the terms, occupational type and work pattern. Crude OR of shift work for periodontitis was 1.269 [CI 1.213-1.327, P < 0.05]. However, following adjustment for multiple confounding factors and the interaction effect term considered, this OR (1.269) increased to 1.381 [CI 1.253-1.523] in white-collar group while it decreased to 1.198 [1.119-1.283] in blue-collar. Crude OR of blue-collar (OR = 3.123, CI 2.972-3.281, P < 0.05) decreased to 1.151 [CI 1.049-1.262] when interaction effect to the shift work was considered. Shift work pattern increases the risk for periodontitis and this adverse effect is greater when white-collar workers are engaged comparing to blue-collar. The result of this study suggests that 24/7 lifestyle of the modern society poses health risks to the relevant people and the potential harm can be greater to white-collar workers.

Evaluation of artificial intelligence model for crowding categorization and extraction diagnosis using intraoral photographs.

Determining the severity of dental crowding and the necessity of tooth extraction for orthodontic treatment planning are time-consuming processes and there are no firm criteria. Thus, automated assistance would be useful to clinicians. This study aimed to construct and evaluate artificial intelligence (AI) systems to assist with such treatment planning. A total of 3,136 orthodontic occlusal photographs with annotations by two orthodontists were obtained. Four convolutional neural network (CNN) models, namely ResNet50, ResNet101, VGG16, and VGG19, were adopted for the AI process. Using the intraoral photographs as input, the crowding group and the necessity of tooth extraction were obtained. Arch length discrepancy analysis with AI-detected landmarks was used for crowding categorization. Various statistical and visual analyses were conducted to evaluate the performance. The maxillary and mandibular VGG19 models showed minimum mean errors of 0.84 mm and 1.06 mm for teeth landmark detection, respectively. Analysis of Cohen's weighted kappa coefficient indicated that crowding categorization performance was best in VGG19 (0.73), decreasing in the order of VGG16, ResNet101, and ResNet50. For tooth extraction, the maxillary VGG19 model showed the highest accuracy (0.922) and AUC (0.961). By utilizing deep learning with orthodontic photographs, dental crowding categorization and diagnosis of orthodontic extraction were successfully determined. This suggests that AI can assist clinicians in the diagnosis and decision making of treatment plans.

Perceptual and quantitative analysis of discoloration of orthodontic elastomeric chains by food.

BACKGROUND: The objectives of this study were to use a digital camera to measure the discoloration of orthodontic elastomeric chains in various immersion solutions over different time periods and to determine the valid acceptability tolerances for color changes in orthodontic elastomeric chains by surveying digital photographs. METHODS: Orthodontic elastomeric chains were applied to the maxillary anterior teeth of nine typodont models. The models were divided into three groups and immersed in the curry, coffee, and wine solutions. The digital images of the elastomeric modules were captured and processed using commercial software after 30 min of immersion, thrice a day, for 28 days. The differences in color changes ([Formula: see text]), depending on the type of immersion solution and period, were analyzed using a repeated-measures analysis of variance (ANOVA) test. A web-based survey questionnaire was randomly distributed to 50 respondents for a visual analysis of the elastomeric chain discoloration. The relationship between the surveying score and [Formula: see text] value was analyzed using Spearman's correlation coefficient. The perceptibility and acceptability of elastomeric chain discoloration ([Formula: see text]) based on the type of immersion solutions and periods were analyzed using a regression model. RESULTS: Significant differences were observed in the discoloration of the elastomeric power chain depending on the immersion solution and period. The amount of discoloration was highest in curry, followed by coffee and wine (P < 0.05). The mean discoloration ([Formula: see text]) continued to increase over the entire immersion period. There was a significant correlation between visual scoring and discoloration ([Formula: see text]) over the entire period, especially in the early stages compared to the later stages (r = 0.918, P < 0.05). In 50% of the respondents, the predicted clinically unacceptable discoloration was between 4.46 and 9.98 and in 90% of the respondents, it was between 16.52 and 19.85. CONCLUSIONS: The amount of discoloration was the highest for curry, followed by coffee and wine, and continued to gradually increase during the observation period. Significant differences were found between the color measurements obtained and the visual assessment by observers. The observers varied in their tolerance for perceptibility and acceptability for elastomeric chain discoloration based on the type of dietary media.

Application of deep learning artificial intelligence technique to the classification of clinical orthodontic photos.

BACKGROUND: Taking facial and intraoral clinical photos is one of the essential parts of orthodontic diagnosis and treatment planning. Among the diagnostic procedures, classification of the shuffled clinical photos with their orientations will be the initial step while it was not easy for a machine to classify photos with a variety of facial and dental situations. This article presents a convolutional neural networks (CNNs) deep learning technique to classify orthodontic clinical photos according to their orientations. METHODS: To build an automated classification system, CNNs models of facial and intraoral categories were constructed, and the clinical photos that are routinely taken for orthodontic diagnosis were used to train the models with data augmentation. Prediction procedures were evaluated with separate photos whose purpose was only for prediction. RESULTS: Overall, a 98.0% valid prediction rate resulted for both facial and intraoral photo classification. The highest prediction rate was 100% for facial lateral profile, intraoral upper, and lower photos. CONCLUSION: An artificial intelligence system that utilizes deep learning with proper training models can successfully classify orthodontic facial and intraoral photos automatically. This technique can be used for the first step of a fully automated orthodontic diagnostic system in the future.

Assessment of Morphologic Change of Mandibular Condyle in Temporomandibular Joint Osteoarthritis Patients with Stabilization Splint Therapy: A Pilot Study.

(1) Background: The purpose of this study was to evaluate the 3-dimensional bony changes of the mandibular condyle in temporomandibular joints-osteoarthritis (TMJ-OA) patients treated with stabilization splint (SS) therapy using shape correspondence analysis. (2) Methods: A total of 27 adult patients (2 men and 25 women) with a mean age of 24.6 ± 3.9 years were included in this study. All patients were diagnosed with TMJ-OA and were treated with an SS. Cone-beam computed tomography data of the condylar head before and after SS therapy from 42 condyles (15 bilateral and 12 unilateral TMJ-OA) were used for the analysis. For the performance shape correspondence analysis (SPHARM-PDM), statistical differences were performed using the one-way analysis of variance and Scheffe post hoc tests. (3) Results: After SS treatment in TMJ-OA patients, bone resorption of the condyle head surface was predominant in the anterosuperior, superolateral, and superior areas, and bone formation was superior in the lateral, medial, posterosuperior, and posteromedial areas. The change in the condylar volume between the two groups was not statistically significant. (4) Conclusions: After SS treatment in TMJ-OA patients, there was both bone resorption and bone formation on the mandibular condyle head surface, which induced morphological changes in the condyle head.

Deep semi-supervised learning for automatic segmentation of inferior alveolar nerve using a convolutional neural network.

BACKGROUND: The inferior alveolar nerve (IAN) innervates and regulates the sensation of the mandibular teeth and lower lip. The position of the IAN should be monitored prior to surgery. Therefore, a study using artificial intelligence (AI) was planned to image and track the position of the IAN automatically for a quicker and safer surgery. METHODS: A total of 138 cone-beam computed tomography datasets (Internal: 98, External: 40) collected from multiple centers (three hospitals) were used in the study. A customized 3D nnU-Net was used for image segmentation. Active learning, which consists of three steps, was carried out in iterations for 83 datasets with cumulative additions after each step. Subsequently, the accuracy of the model for IAN segmentation was evaluated using the 50 datasets. The accuracy by deriving the dice similarity coefficient (DSC) value and the segmentation time for each learning step were compared. In addition, visual scoring was considered to comparatively evaluate the manual and automatic segmentation. RESULTS: After learning, the DSC gradually increased to 0.48 ± 0.11 to 0.50 ± 0.11, and 0.58 ± 0.08. The DSC for the external dataset was 0.49 ± 0.12. The times required for segmentation were 124.8, 143.4, and 86.4 s, showing a large decrease at the final stage. In visual scoring, the accuracy of manual segmentation was found to be higher than that of automatic segmentation. CONCLUSIONS: The deep active learning framework can serve as a fast, accurate, and robust clinical tool for demarcating IAN location.

Shift work sleep disorder is closely associated with an increased risk for periodontal disease.

AIM: This study aimed to investigate the association between work patterns and periodontal disease. MATERIALS AND METHODS: Data were collected from the Korea National Health and Nutrition Examination Survey between 2007 and 2012, and data from 22,508 subjects aged ≥19 years were included. An individual's work pattern was classified as either daytime or shift work. Sleep duration was categorized into three ranges: ≤5, 6-8, and ≥9 h/day. A multivariate logistic regression model was used to determine the adjusted odds ratio (OR) for CPI (Community Periodontal Index) ≥3. The CONTRAST statement was used to show the interaction effect of work patterns and sleep duration. RESULTS: The adjusted OR of shift work was 2.168 (CI: 1.929-2.438, p < .0001). Participants who sleep ≤5 or ≥9 h/day showed ORs 0.735 and 0.663, respectively (p = .0181). Interaction effect analysis revealed that the work pattern had a strong influence on periodontal condition when combined with the sleep amount. Shift workers with ≤5 or ≥ 9 h of sleep showed significantly increased ORs for CPI ≥3 (2.1406 and 2.3251, respectively, p < .0001). The ORs for daytime workers were comparable to the original values (≤5: 0.7348, p = .0292; ≥9: 0.6633, p = .0428). CONCLUSION: Altered sleep patterns caused by shift work have more influence on periodontal disease than sleep duration.

Influence of the Depth of the Convolutional Neural Networks on an Artificial Intelligence Model for Diagnosis of Orthognathic Surgery.

The aim of this study was to investigate the relationship between image patterns in cephalometric radiographs and the diagnosis of orthognathic surgery and propose a method to improve the accuracy of predictive models according to the depth of the neural networks. The study included 640 and 320 patients requiring non-surgical and surgical orthodontic treatments, respectively. The data of 150 patients were exclusively classified as a test set. The data of the remaining 810 patients were split into five groups and a five-fold cross-validation was performed. The convolutional neural network models used were ResNet-18, 34, 50, and 101. The number in the model name represents the difference in the depth of the blocks that constitute the model. The accuracy, sensitivity, and specificity of each model were estimated and compared. The average success rate in the test set for the ResNet-18, 34, 50, and 101 was 93.80%, 93.60%, 91.13%, and 91.33%, respectively. In screening, ResNet-18 had the best performance with an area under the curve of 0.979, followed by ResNets-34, 50, and 101 at 0.974, 0.945, and 0.944, respectively. This study suggests the required characteristics of the structure of an artificial intelligence model for decision-making based on medical images.

A long-term longitudinal study of the osteoarthritic changes to the temporomandibular joint evaluated using a novel three-dimensional superimposition method.

The aim of this study was to assess the changes in individual condyles from 5 to 8 years in patients with temporomandibular joint (TMJ) osteoarthritis using 3-dimensional cone beam computed tomography (3D CBCT) reconstruction and superimposition. To assess the longitudinal TMJ changes, CBCT was performed at initial (T0) and final (T2) timepoints that were at least 5 years apart and at a middle (T1) timepoint. To improve the accuracy, we used a novel superimposition method that designated areas of coronoid process and mandibular body. The differences in the resorption and apposition amounts were calculated between each model via maximum surface distances. The greatest resorption and apposition observed were - 7.48 and 2.66 mm, respectively. Evaluation of the changes in each condyle showed that osteoarthritis leads to both resorption and apposition. Resorption was mainly observed in the superior region, while high apposition rates were observed (in decreasing order) in the posterior, lateral, and anterior regions. The medial parts showed greater apposition than the lateral parts in all regions. Our superimposition method reveals that both resorption and apposition were observed in condyles with TMJ osteoarthritis, and resorption/apposition patterns depend on the individual condyle and its sites.

Comparative Analysis of the Differences in Dentofacial Morphology According to the Tongue and Lip Pressure.

The aim of this study was to evaluate the effects of the tongue and lip pressure on dentofacial morphology. The subjects comprised 194 patients with malocclusion. Anterior and posterior tongue elevation and lip pressures were evaluated using the Iowa Oral Performance Instrument (IOPI) device. The lateral cephalograms of each subject were traced and digitized to perform the analysis. Statistical analysis was used to investigate the relationship between perioral muscle force and the cephalometric variables. Anterior and posterior tongue pressure was both higher in males than in females. No sex difference in lip pressure was observed. The group with a low posterior tongue pressure showed a short ramus height, short posterior facial height, and clockwise-rotated mandible. On the other hand, lip pressure had a significant influence on maxillary incisor angulation. Skeletal pattern was not found to be significantly related with lip pressure. The anterior tongue pressure appeared as a mixed pattern of the two results. Tongue pressure was related to skeletal measurements, such as short posterior facial height, and lip pressure was related to the angulation of the anterior teeth. This study suggests that there may be differences in dentofacial morphology according to the differences in perioral muscle force.

Modulation of Lithium Disilicate Translucency through Heat Treatment.

The aim of this study was to present a control method for modulating the translucency of lithium disilicate ceramics through thermal refinement. Identical lithium disilicate blocks were thermally refined using four different heat treatment schedules, and the microstructure, translucency, and flexural strength of the ceramics were investigated in detail by SEM, spectroscopy, and a piston-on-three-ball test. The results showed that ceramics treated under higher heat had larger grains, with an average size between 240 and 1080 nm. In addition, a higher transmittance of all wavelengths was observed in ceramics treated under lower heat, and the transmittance in the 550 nm wavelength ranged from 27 to 34%. The results suggest that the translucency of ceramics can be modified through thermal refinement under two conditions: (1) the particle size of the ceramic is small enough to achieve minimal grain-boundary light scattering, and (2) the percentage of particles allowing visible light transmission is altered by the heat treatment.

Three-Dimensional Reproducibility of the Soft Tissue Landmarks Taken by Structured-Light Facial Scanner in Accordance with the Head Position Change.

The aim of this study was to evaluate the three-dimensional reproducibility of the structured-light facial scanner according to the head position change. A mannequin head was used and angle of the mannequin's axis-orbital plane to the true horizontal plane was adjusted to +10, +5, 0, -5, and -10°. Facial scanning was conducted 30 times, respectively, and 150 3D images were obtained. Reoriented landmarks of each group were compared and analyzed. Reproducibility decreased as the distance from the facial center increased. Additionally, the landmarks below showed lower reproducibility and higher dispersion than landmarks above. These differences occurred mainly in the anteroposterior direction as opposed to other directions. Positive inclination of the head position showed superior reproducibility compared to a negative inclination. This study showed that reproducibility of a structured-light scanner could be varied depending on the head position. Inaccuracies of landmarks in the anteroposterior direction are greater than in other directions. This means that evaluations of the profile using a structured-light scanner should be made carefully. Therefore, the proper head position should be set to ensure the accuracy of the image.

Deep Active Learning for Automatic Segmentation of Maxillary Sinus Lesions Using a Convolutional Neural Network.

The aim of this study was to segment the maxillary sinus into the maxillary bone, air, and lesion, and to evaluate its accuracy by comparing and analyzing the results performed by the experts. We randomly selected 83 cases of deep active learning. Our active learning framework consists of three steps. This framework adds new volumes per step to improve the performance of the model with limited training datasets, while inferring automatically using the model trained in the previous step. We determined the effect of active learning on cone-beam computed tomography (CBCT) volumes of dental with our customized 3D nnU-Net in all three steps. The dice similarity coefficients (DSCs) at each stage of air were 0.920 ± 0.17, 0.925 ± 0.16, and 0.930 ± 0.16, respectively. The DSCs at each stage of the lesion were 0.770 ± 0.18, 0.750 ± 0.19, and 0.760 ± 0.18, respectively. The time consumed by the convolutional neural network (CNN) assisted and manually modified segmentation decreased by approximately 493.2 s for 30 scans in the second step, and by approximately 362.7 s for 76 scans in the last step. In conclusion, this study demonstrates that a deep active learning framework can alleviate annotation efforts and costs by efficiently training on limited CBCT datasets.

Evaluation of Transfer Learning with Deep Convolutional Neural Networks for Screening Osteoporosis in Dental Panoramic Radiographs.

: Dental panoramic radiographs (DPRs) provide information required to potentially evaluate bone density changes through a textural and morphological feature analysis on a mandible. This study aims to evaluate the discriminating performance of deep convolutional neural networks (CNNs), employed with various transfer learning strategies, on the classification of specific features of osteoporosis in DPRs. For objective labeling, we collected a dataset containing 680 images from different patients who underwent both skeletal bone mineral density and digital panoramic radiographic examinations at the Korea University Ansan Hospital between 2009 and 2018. Four study groups were used to evaluate the impact of various transfer learning strategies on deep CNN models as follows: a basic CNN model with three convolutional layers (CNN3), visual geometry group deep CNN model (VGG-16), transfer learning model from VGG-16 (VGG-16_TF), and fine-tuning with the transfer learning model (VGG-16_TF_FT). The best performing model achieved an overall area under the receiver operating characteristic of 0.858. In this study, transfer learning and fine-tuning improved the performance of a deep CNN for screening osteoporosis in DPR images. In addition, using the gradient-weighted class activation mapping technique, a visual interpretation of the best performing deep CNN model indicated that the model relied on image features in the lower left and right border of the mandibular. This result suggests that deep learning-based assessment of DPR images could be useful and reliable in the automated screening of osteoporosis patients.

Artificial Intelligent Model With Neural Network Machine Learning for the Diagnosis of Orthognathic Surgery.

Diagnosis and treatment planning are the most important steps in the orthognathic surgery for the successful treatment. The purpose of this study was to develop a new artificial intelligent model for surgery/non-surgery decision and extraction determination, and to evaluate the performance of this model. The sample used in this study consisted of 316 patients in total. Of the total sample, 160 were planned with surgical treatment and 156 were planned with non-surgical treatment. The input values of artificial neural network were obtained from 12 measurement values of the lateral cephalogram and 6 additional indexes. The artificial intelligent model of machine learning consisted of 2-layer neural network with one hidden layer. The learning was carried out in 3 stages, and 4 best performing models were adopted. Using these models, decision-making success rates of surgery/non-surgery, surgery type, and extraction/non-extraction were calculated. The final diagnosis success rate was calculated by comparing the actual diagnosis with the diagnosis obtained by the artificial intelligent model. The success rate of the model showed 96% for the diagnosis of surgery/non-surgery decision, and showed 91% for the detailed diagnosis of surgery type and extraction decision. This study suggests the artificial intelligent model using neural network machine learning could be applied for the diagnosis of orthognathic surgery cases.

Influence of changing various parameters in miniscrew-assisted rapid palatal expansion: A three-dimensional finite element analysis.

OBJECTIVE: This study aimed to analyze the effect of changing various parameters of the bone-borne rapid palatal expander (RPE) using the finite element method (FEM). METHODS: In eight experimental groups, we investigated the effect of the number, position, and length of miniscrews; positional changes of the expander; and changes in the hook length on maxillary expansion. In finite element analysis, we compared the magnitude and distribution of stress, and the displacement changes following expansion of the bone-borne RPE. RESULTS: When we compared the number and position of miniscrews, placing miniscrews in the anterior and posterior sides was advantageous for maxillary expansion in terms of stress distribution and displacement changes. Miniscrew length did not significantly affect stress distribution and displacement changes. Furthermore, anteroposterior displacement of the expander did not significantly affect transverse maxillary expansion but had various effects on vertical changes of the maxilla. The maxilla rotated clockwise when the miniscrews were placed in the anterior region. The hook length of the expander did not show consistent results in terms of changes in stress distribution and magnitude or in displacement changes. CONCLUSIONS: The findings of this study suggest that changes in the location and length of the miniscrews and displacement of the bone-borne RPE could affect the pattern of the maxillary expansion, depending on the combination of these factors.

New approach for the diagnosis of extractions with neural network machine learning.

INTRODUCTION: The decision to extract teeth for orthodontic treatment is important and difficult because it tends to be based on the practitioner's experiences. The purposes of this study were to construct an artificial intelligence expert system for the diagnosis of extractions using neural network machine learning and to evaluate the performance of this model. METHODS: The subjects included 156 patients. Input data consisted of 12 cephalometric variables and an additional 6 indexes. Output data consisted of 3 bits to divide the extraction patterns. Four neural network machine learning models for the diagnosis of extractions were constructed using a back-propagation algorithm and were evaluated. RESULTS: The success rates of the models were 93% for the diagnosis of extraction vs nonextraction and 84% for the detailed diagnosis of the extraction patterns. CONCLUSIONS: This study suggests that artificial intelligence expert systems with neural network machine learning could be useful in orthodontics. Improved performance was achieved by components such as proper selection of the input data, appropriate organization of the modeling, and preferable generalization.

The relevance analysis of hyoid bone position to skeletal or dental openbite and dentofacial characteristics.

OBJECTIVE: The aim of this study was to investigate the relationship between the hyoid bone position and skeletal or dental openbite and to investigate dentofacial characteristics as indicated by the hyoid bone position. STUDY DESIGN: In our study, 182 patients were included on the basis of skeletal and dental openbite. The hyoid bone position of the subjects was compared and evaluated. In addition, by dividing the samples according to the hyoid bone position, dentofacial characteristics of the subjects were compared and analyzed. RESULTS: There were significant differences in the hyoid bone position according to the skeletal pattern, not dental pattern. The skeletal openbite group showed low hyoid bone position. In addition, the low hyoid bone group showed short ramus height, short posterior facial height, retrusive chin, and clockwise-rotated mandible. CONCLUSIONS: Patients with low hyoid bone had a tendency toward skeletal openbite, even though there was no dental openbite. Moreover, low hyoid bone position had relevance to retrognathic dentofacial characteristics.

Treatment of unilateral posterior crossbite with facial asymmetry in a female patient with transverse discrepancy.

A unilateral posterior crossbite with facial asymmetry is difficult to correct with orthodontic treatment alone. This case report describes the orthodontic treatment and additional plasty without orthognathic surgery for a 19-year-old woman with a transverse discrepancy. The posterior crossbite was resolved by expansion of the narrow maxillary arch and space closure in the mandibular arch. This accelerated the correction of the functional shift of the mandible. After resolution of the unilateral posterior crossbite, the problems of the anteroposterior molar relationship were treated using orthodontic mini-implants. Mandibular angle reduction plasty was performed for the asymmetric mandibular border to improve the facial appearance. After treatment, the patient had a more symmetrical facial appearance, normal overjet and overbite, and midline coincidence. The treatment results remained stable 1 year after treatment. This case report demonstrates that a minimally invasive treatment can successfully correct a unilateral posterior crossbite with a transverse discrepancy.