External validation of an artificial intelligence-based method for the detection and classification of molar incisor hypomineralisation in dental photographs.

OBJECTIVES: This ex vivo diagnostic study aimed to externally validate an open-access artificial intelligence (AI)-based model for the detection, classification, localisation and segmentation of enamel/molar incisor hypomineralisation (EH/MIH). METHODS: An independent sample of web images showing teeth with (n = 277) and without (n = 178) EH/MIH was evaluated by a workgroup of dentists whose consensus served as the reference standard. Then, an AI-based model was used for the detection of EH/MIH, followed by automated classification and segmentation of the findings (test method). The accuracy (ACC), sensitivity (SE), specificity (SP) and area under the curve (AUC) were determined. Furthermore, the correctness of EH/MIH lesion localisation and segmentation was evaluated. RESULTS: An overall ACC of 94.3 % was achieved for image-based detection of EH/MIH. Cross-classification of the AI-based class prediction and the reference standard resulted in an agreement of 89.2 % for all diagnostic decisions (n = 594), with an ACC between 91.4 % and 97.8 %. The corresponding SE and SP values ranged from 81.7 % to 92.8 % and 91.9 % to 98.7 %, respectively. The AUC varied between 0.894 and 0.945. Image size had only a limited impact on diagnostic performance. The AI-based model correctly predicted EH/MIH localisation in 97.3 % of cases. For the detected lesions, segmentation was fully correct in 63.4 % of all cases and partially correct in 33.9 %. CONCLUSIONS: This study documented the promising diagnostic performance of an open-access AI tool in the detection and classification of EH/MIH in external images. CLINICAL SIGNIFICANCE: Externally validated AI-based diagnostic methods could facilitate the detection of EH/MIH lesions in dental photographs.

European Organisation for Caries Research and the European Federation of Conservative Dentistry Consensus Report on Clinical Recommendations for Caries Diagnosis: Paper III - Caries Diagnosis at the Individual Level.

OBJECTIVES: The aim of the present consensus paper was to provide recommendations for clinical practice on the individual etiological and modifying factors to be assessed in the individual diagnosis of caries, and the methods for their assessment, supporting personalized treatment decisions. MATERIAL AND METHODS: The executive councils of the European Organisation for Caries Research (ORCA) and the European Federation of Conservative Dentistry (EFCD) nominated ten experts each to join the expert panel. The steering committee formed three work groups which were asked to provide recommendations on (1) caries detection and diagnostic methods, (2) caries activity assessment, and (3) forming individualized caries diagnoses. The experts responsible for "individualised caries diagnosis" searched and evaluated the relevant literature, drafted this manuscript and made provisional consensus recommendations. These recommendations were discussed and refined during the structured process in the whole work group. Finally, the agreement for each recommendation was determined using an anonymous eDelphi survey. The threshold for approval of recommendations was determined at 70% agreement. RESULTS: Ten recommendations were approved and agreed by the whole expert panel, covering medical history, caries experience, plaque, diet, fluoride, and saliva. While the level of evidence was low, the level of agreement was typically very high, except for one recommendation on salivary flow measurement, where 70% agreed. CONCLUSION: It is recommended that all aspects of caries lesion progression and activity, recent caries experience, medical conditions and medications, plaque, diet, fluoride and saliva should be synthesized to arrive at an individual diagnosis. CLINICAL RELEVANCE: The expert panel merged evidence from existing guidelines and scientific literature with practical considerations and provided recommendations for their use in daily dental practice.

ORCA-EFCD consensus report on clinical recommendations for caries diagnosis. Paper II: caries lesion activity and progression assessment.

INTRODUCTION: This consensus paper provides recommendations for oral health professionals on why and how to assess caries activity and progression with special respect to the site of a lesion. METHODS: An expert panel was nominated by the executive councils of the European Organization for Caries Research (ORCA) and the European Federation of Conservative Dentistry (EFCD). The steering committee built three working groups that were asked to provide recommendations on 1) caries detection and diagnostic methods, 2) caries activity and progression assessment and 3) obtain individualized caries diagnoses. The experts of work group 2 phrased and agreed on provisional general and specific recommendations on caries lesion activity and progression, based on a review of the current literature. These recommendations were then discussed and refined in a consensus workshop followed by an anonymous Delphi survey to determine the agreement on each recommendation. RESULTS: The expert panel agreed on general (n=7) and specific recommendations (n=6). The specific recommendations cover coronal caries on pits and fissures, smooth surfaces, proximal surfaces, as well as root caries and secondary caries/ caries adjacent to restorations and sealants (CARS). 3/13 recommendations yielded perfect agreement. CONCLUSION: The most suitable method for lesion activity assessment is the visual-tactile method. No single clinical characteristic is indicative of lesion activity; instead, lesion activity assessment is based on assessing and weighing several clinical signs. The recall intervals for visual and radiographic examination need to be adjusted to the presence of active caries lesions and recent caries progression rates. Modifications should be based on individual patient characteristics.

Diagnostic performance of artificial intelligence-aided caries detection on bitewing radiographs: a systematic review and meta-analysis.

The accuracy of artificial intelligence-aided (AI) caries diagnosis can vary considerably depending on numerous factors. This review aimed to assess the diagnostic accuracy of AI models for caries detection and classification on bitewing radiographs. Publications after 2010 were screened in five databases. A customized risk of bias (RoB) assessment tool was developed and applied to the 14 articles that met the inclusion criteria out of 935 references. Dataset sizes ranged from 112 to 3686 radiographs. While 86 % of the studies reported a model with an accuracy of ≥80 %, most exhibited unclear or high risk of bias. Three studies compared the model's diagnostic performance to dentists, in which the models consistently showed higher average sensitivity. Five studies were included in a bivariate diagnostic random-effects meta-analysis for overall caries detection. The diagnostic odds ratio was 55.8 (95 % CI= 28.8 - 108.3), and the summary sensitivity and specificity were 0.87 (0.76 - 0.94) and 0.89 (0.75 - 0.960), respectively. Independent meta-analyses for dentin and enamel caries detection were conducted and showed sensitivities of 0.84 (0.80 - 0.87) and 0.71 (0.66 - 0.75), respectively. Despite the promising diagnostic performance of AI models, the lack of high-quality, adequately reported, and externally validated studies highlight current challenges and future research needs.

ORCA-EFCD consensus report on clinical recommendation for caries diagnosis. Paper I: caries lesion detection and depth assessment.

OBJECTIVES: The aim of the present consensus paper was to provide recommendations for clinical practice considering the use of visual examination, dental radiography and adjunct methods for primary caries detection. MATERIALS AND METHODS: The executive councils of the European Organisation for Caries Research (ORCA) and the European Federation of Conservative Dentistry (EFCD) nominated ten experts each to join the expert panel. The steering committee formed three work groups that were asked to provide recommendations on (1) caries detection and diagnostic methods, (2) caries activity assessment and (3) forming individualised caries diagnoses. The experts responsible for "caries detection and diagnostic methods" searched and evaluated the relevant literature, drafted this manuscript and made provisional consensus recommendations. These recommendations were discussed and refined during the structured process in the whole work group. Finally, the agreement for each recommendation was determined using an anonymous Delphi survey. RESULTS: Recommendations (N = 8) were approved and agreed upon by the whole expert panel: visual examination (N = 3), dental radiography (N = 3) and additional diagnostic methods (N = 2). While the quality of evidence was found to be heterogeneous, all recommendations were agreed upon by the expert panel. CONCLUSION: Visual examination is recommended as the first-choice method for the detection and assessment of caries lesions on accessible surfaces. Intraoral radiography, preferably bitewing, is recommended as an additional method. Adjunct, non-ionising radiation methods might also be useful in certain clinical situations. CLINICAL RELEVANCE: The expert panel merged evidence from the scientific literature with practical considerations and provided recommendations for their use in daily dental practice.

Automatized Detection of Periodontal Bone Loss on Periapical Radiographs by Vision Transformer Networks.

Several artificial intelligence-based models have been presented for the detection of periodontal bone loss (PBL), mostly using convolutional neural networks, which are the state of the art in deep learning. Given the emerging breakthrough of transformer networks in computer vision, we aimed to evaluate various models for automatized PBL detection. An image data set of 21,819 anonymized periapical radiographs from the upper/lower and anterior/posterior regions was assessed by calibrated dentists according to PBL. Five vision transformer networks (ViT-base/ViT-large from Google, BEiT-base/BEiT-large from Microsoft, DeiT-base from Facebook/Meta) were utilized and evaluated. Accuracy (ACC), sensitivity (SE), specificity (SP), positive/negative predictive value (PPV/NPV) and area under the ROC curve (AUC) were statistically determined. The overall diagnostic ACC and AUC values ranged from 83.4 to 85.2% and 0.899 to 0.918 for all evaluated transformer networks, respectively. Differences in diagnostic performance were evident for lower (ACC 94.1-96.7%; AUC 0.944-0.970) and upper anterior (86.7-90.2%; 0.948-0.958) and lower (85.6-87.2%; 0.913-0.937) and upper posterior teeth (78.1-81.0%; 0.851-0.875). In this study, only minor differences among the tested networks were detected for PBL detection. To increase the diagnostic performance and to support the clinical use of such networks, further optimisations with larger and manually annotated image data sets are needed.

Detection of Periodontal Bone Loss on Periapical Radiographs-A Diagnostic Study Using Different Convolutional Neural Networks.

Interest in machine learning models and convolutional neural networks (CNNs) for diagnostic purposes is steadily increasing in dentistry. Here, CNNs can potentially help in the classification of periodontal bone loss (PBL). In this study, the diagnostic performance of five CNNs in detecting PBL on periapical radiographs was analyzed. A set of anonymized periapical radiographs (N = 21,819) was evaluated by a group of trained and calibrated dentists and classified into radiographs without PBL or with mild, moderate, or severe PBL. Five CNNs were trained over five epochs. Statistically, diagnostic performance was analyzed using accuracy (ACC), sensitivity (SE), specificity (SP), and area under the receiver operating curve (AUC). Here, overall ACC ranged from 82.0% to 84.8%, SE 88.8-90.7%, SP 66.2-71.2%, and AUC 0.884-0.913, indicating similar diagnostic performance of the five CNNs. Furthermore, performance differences were evident in the individual sextant groups. Here, the highest values were found for the mandibular anterior teeth (ACC 94.9-96.0%) and the lowest values for the maxillary posterior teeth (78.0-80.7%). It can be concluded that automatic assessment of PBL seems to be possible, but that diagnostic accuracy varies depending on the location in the dentition. Future research is needed to improve performance for all tooth groups.

Detection and localization of caries and hypomineralization on dental photographs with a vision transformer model.

Caries and molar-incisor hypomineralization (MIH) are among the most prevalent diseases worldwide and need to be reliably diagnosed. The use of dental photographs and artificial intelligence (AI) methods may potentially contribute to realizing accurate and automated diagnostic visual examinations in the future. Therefore, the present study aimed to develop an AI-based algorithm that can detect, classify and localize caries and MIH. This study included an image set of 18,179 anonymous photographs. Pixelwise image labeling was achieved by trained and calibrated annotators using the Computer Vision Annotation Tool (CVAT). All annotations were made according to standard methods and were independently checked by an experienced dentist. The entire image set was divided into training (N = 16,679), validation (N = 500) and test sets (N = 1000). The AI-based algorithm was trained and finetuned over 250 epochs by using image augmentation and adapting a vision transformer network (SegFormer-B5). Statistics included the determination of the intersection over union (IoU), average precision (AP) and accuracy (ACC). The overall diagnostic performance in terms of IoU, AP and ACC were 0.959, 0.977 and 0.978 for the finetuned model, respectively. The corresponding data for the most relevant caries classes of non-cavitations (0.630, 0.813 and 0.990) and dentin cavities (0.692, 0.830, and 0.997) were found to be high. MIH-related demarcated opacity (0.672, 0.827, and 0.993) and atypical restoration (0.829, 0.902, and 0.999) showed similar results. Here, we report that the model achieves excellent precision for pixelwise detection and localization of caries and MIH. Nevertheless, the model needs to be further improved and externally validated.

Asthma medication and risk of dental diseases in children - A prospective cohort study.

BACKGROUND: Dental caries and enamel defects are the main causes of poor dental health in children, with a substantial impact on their well-being. Use of inhaled asthma medication is a suspected risk factor, but there is a lack of prospective studies investigating this and other prenatal and early life risk factors. METHODS: Copenhagen Prospective Studies on Asthma in Childhood 2010 mother-child cohort (COPSAC2010 ) consists of 700 women who were recruited at 24 weeks of pregnancy. 588 of their children participated in a dental examination at 6 years of age (84%) at the COPSAC2010 research unit. Caries was defined as decayed, missing, or filled surfaces. Enamel defect was defined as demarcated opacity, post-eruptive enamel breakdown, and/or atypical restoration on at least one molar. Caries and enamel defects were assessed in both deciduous and permanent dentitions. RESULTS: We found no associations between inhaled corticosteroids or ß2 -agonists or asthma symptoms in early childhood and the risk of caries or enamel defects by 6 years of age. Furthermore, we found no strong pre-, peri-, or postnatal risk factors for dental diseases at 6 years, except from nominally significant associations between antibiotic use in pregnancy (OR = 1.25, [1.01-1.54]), maternal education level (OR = 1.57, [1.01-2.45]), having a dog at home (OR = 0.50, [0.27-0.93]), and risk of enamel defects. CONCLUSIONS: Use of inhaled corticosteroids, ß2 -agonists, or asthma symptoms in the first 6 years of life were not associated with the development of caries or enamel defects. This finding is reassuring for parents and physicians prescribing asthma medication for young children.

Chapter 7: Technological Aids and Coronal Caries.

In recent decades, dentistry has developed significantly in all areas. While in the past, caries was mainly treated operatively, the today's management has shifted toward noninvasive, minimal invasive, and, only if needed, invasive treatment options. Aiming at enabling the most noninvasive or conservative treatment option requires early caries detection, which, however, remains challenging. The progression of early or noncavitated caries lesions can nowadays be successfully controlled, as well as lesions arrested by oral hygiene procedures combined with the use of fluorides, sealants, or resin infiltration. Methods such as near-infrared light transillumination, fibre-optic transillumination, digital fibre-optic transillumination, laser fluorescence, or quantitative light fluorescence measurements were introduced in the dental market to provide X-ray-free caries detection, assessment, and monitoring. For approximal surfaces that are not directly visible, bitewing radiography is still the standard in detecting caries lesions. The use of artificial intelligence has become the most recent technological aid for the detection of caries lesions on bitewing radiographs and clinical images and has to be understood as an emerging technology, which requires extensive research in the future. The aim of this chapter is to give an overview of different possibilities to detect coronal caries lesions and suggestions of how to improve this process.

The Detection of Dental Pathologies on Periapical Radiographs-Results from a Reliability Study.

(1) Background: Caries, periapical lesions, periodontal bone loss (PBL), and endo-perio lesions are common dental findings that require an accurate diagnostic assessment to allow appropriate disease management. The purpose of this reliability study was to compare the inter- and intra-rater reliability for the detection of the above-mentioned pathologies on periapical radiographs. (2) Methods: Fourteen dentists (three with more than two years and eleven with less than two years of work experience) participated in a training workshop prior to data acquisition. A total of 150 radiographs were assessed by all raters in two rounds. Cohen's Kappa (CK) values and a binary logistic regression were calculated. (3) Results: The reliability was found in a moderate and substantial range of agreement: caries (mean inter-rater CK value/first round 0.704/mean inter-rater CK value/second round 0.659/mean intra-rater CK value 0.778), periapical lesions (0.643/0.611/0.768), PBL (0.454/0.482/0.739) and endo-perio lesion (0.702/0.689/0.840). The regression model revealed a significant influence of the clinical experience, and furthermore, periapical pathologies and PBL were identified less reliably in comparison to caries and endo-perio lesions. (4) Conclusions: The dentist's ability to detect the chosen pathologies was linked with significant differences. Periapical lesions and PBL were identified less reliably than caries and endo-perio lesions.

Oral inflammation and exhaled nitric oxide fraction: a cross-sectional study.

Oral inflammation is not associated with increased F ENO in nonasthmatic children and adolescents. The observed inverse association implies that gingival bleeding might decrease F ENO but this needs more study to be confirmed. https://bit.ly/3BhMP6f.

The impact of gingivitis reduction on lung function: a randomized trial under intensified oral hygiene.

BACKGROUND: Periodontal disease and lung function impairment were found to be associated with low-grade systemic or local inflammation, and it might be that gingival/periodontal inflammation triggers lung function due to systemic inflammation or the transfer of oral bacteria or its components to the lung. A recent observational study in non-smoking subjects showed that lung volumes and flow rates were significantly reduced by 71-185 ml for severe gingivitis regardless of the adjustment for potential confounders. The result did not show any confounding by smoking, and the association between gingivitis and lower lung function was not modified by systemic inflammation. The designed interventional trial primarily aims to test the hypothesis that gingivitis reduction by optimized daily oral hygiene, professional tooth cleaning and antibacterial chlorhexidine (CHX)-containing mouth rinse improves lung function in terms of forced vital capacity (FVC) by at least 2%. The secondary objective will test the hypothesis that gingivitis reduction improves forced expiratory volume in 1 s (FEV1) and forced expiratory flow at 25-75% of the pulmonary volume (FEF25-75) by at least 2%. Furthermore, the influence of the oral microbiome will be taken into account. METHODS: The study has to include 120 non-smoking subjects aged between 18 and 30 years with biofilm-induced gingivitis. The chosen "waiting control group design" will compare the immediate intervention group with the delayed intervention group, which serves as a control group. Dental and gingival status, lung function and oral microbiome will be recorded. The intensified preventive intervention-professional tooth cleaning, one-stage full-mouth disinfection with CHX and safeguarding an optimal daily oral hygiene by each subject-cannot be blinded, but the outcome measurement in terms of lung function tests is blind. DISCUSSION: This proposed multidisciplinary study has several strengths. Only one previous intervention study with patients with severe periodontitis (mostly smokers) has been performed. It is novel to include non-smoking subjects with mild and potentially reversible oral inflammation. Furthermore, this research is innovative, because it includes evidence-based interventions for gingivitis reduction, standardized measures of the outcome on lung function and oral microbiome and combines expertise from dentistry, lung physiology, oral microbiology and epidemiology/statistical modelling. TRIAL REGISTRATION: German Clinical Trial Register DRKS00028176. Registered on February 2022.

The Dentist's Ability to Detect Different Restorative Procedures on Periapical Radiographs-Results from a Reliability Study.

(1) Background: This in vitro reliability study aimed to determine the inter- and intra-examiner reliability for the detection of direct fillings, indirect crown restorations, root canal fillings and implants on periapical radiographs. (2) Methods: Fourteen dentists (<2 years of clinical experience = 11; >2 years of clinical experience = 3) participated in this diagnostic reliability study in which included a theoretical and practical educational training prior to data collection. The image set of periapical radiographs (N = 150) was examined in two evaluation rounds by all the dentists. Cohen's Kappa (CK) and a binary logistic regression model were computed. (3) Results: The inter- and intra-examiner reliability were found to be in almost perfect agreement: direct fillings (1st round 0.859/2nd round 0.844/intra 0.910), indirect crown restorations (0.932/0.926/0.955), root canal fillings (0.920/0.886/0.941) and dental implants (0.994/0.988/0.987). The binary logistic regression model revealed that the "evaluation round" and "dentist's clinical experience" had no significant influence, but for the "diagnostic category"; small, but statistically significant differences were documented. (4) Conclusions: The reliability for detecting direct and indirect restorations, root canal fillings or implants on periapical radiographs was found to be high in the present reliability study on periapical radiographs.

Revised FDI criteria for evaluating direct and indirect dental restorations-recommendations for its clinical use, interpretation, and reporting.

OBJECTIVES: The FDI criteria for the evaluation of direct and indirect dental restorations were first published in 2007 and updated in 2010. Meanwhile, their scientific use increased steadily, but several questions from users justified some clarification and improvement of the living document. MATERIALS AND METHODS: An expert panel (N = 10) initiated the revision and consensus process that included a kick-off workshop and multiple online meetings by using the Delphi method. During and after each round of discussion, all opinions were collected, and the aggregated summary was presented to the experts aiming to adjust the wording of the criteria as precisely as possible. Finally, the expert panel agreed on the revision. RESULTS: Some categories were redefined, ambiguities were cleared, and the descriptions of all scores were harmonized to cross-link different clinical situations with possible management strategies: reviewing/monitoring (score 1-4), refurbishment/reseal (score 3), repair (score 4), and replacement (score 5). Functional properties (domain F: fracture of material and retention, marginal adaptation, proximal contact, form and contour, occlusion and wear) were now placed at the beginning followed by biological (domain B: caries at restoration margin, hard tissue defects, postoperative hypersensitivity) and aesthetic characteristics (domain A: surface luster and texture, marginal staining, color match). CONCLUSION: The most frequently used eleven categories of the FDI criteria set were revised for better understanding and handling. CLINICAL RELEVANCE: The improved description and structuring of the criteria may help to standardize the evaluation of direct and indirect restorations and may enhance their acceptance by researchers, teachers, and dental practitioners.

Evaluation of direct restorations using the revised FDI criteria: results from a reliability study.

OBJECTIVES: The purpose of this in vitro reliability study was to determine the intra- and inter-examiner agreement of the revised FDI criteria including the categories "fracture of material and retention" (F1) and "caries at restoration margin" (B1). MATERIALS AND METHODS: Forty-nine photographs of direct tooth-coloured posterior (n = 25) and anterior (n = 24) restorations with common deficiencies were included. Ten dental experts repeated the assessment in three blinded rounds. Later, the experts re-evaluated together all photographs and agreed on a reference standard. Statistical analysis included the calculation of Cohen's (Cκ), Fleiss' (Fκ), and weighted Kappa (wκ), the development of a logistic regression with a backward elimination model and Bland/Altman plots. RESULTS: Intra- and inter-examiner reliability exhibited mostly moderate to substantial Cκ, Fκ, and wκ values for posterior restorations (e.g. Intra: F1 Cκ = 0.57, wκ = 0.74; B1 Cκ = 0.57, wκ = 0.73/Inter F1 Fκ = 0.32, wκ = 0.53; B1 Fκ = 0.41, wκ = 0.64) and anterior restorations (e.g. Intra F1 Cκ = 0.63, wκ = 0.76; B1 Cκ = 0.48, wκ = 0.68/Inter F1 Fκ = 0.42, wκ = 0.57; B1 Fκ = 0.40, wκ = 0.51). Logistic regression analyses revealed significant differences between the evaluation rounds, examiners, categories, and tooth type. Both the intra- and inter-examiner reliability increased along with the evaluation rounds. The overall agreement was higher for anterior restorations compared to posterior restorations. CONCLUSIONS: The overall reliability of the revised FDI criteria set was found to be moderate to substantial. CLINICAL RELEVANCE: If properly trained, the revised FDI criteria set are a valid tool to evaluate direct and indirect restorations in a standardized way. However, training and calibration are needed to ensure reliable application.

3-Year Clinical Performance of a New Pit and Fissure Sealant.

The aim of this 3-year, randomized clinical trial (RCT) in split-mouth design was to explore the clinical survival of a Bis-GMA-free pit and fissure sealant (Helioseal F Plus) in comparison to a control material (Helioseal F). The initial population consisted of 92 adolescents. Follow-ups took place after one year (N = 85), two years (N = 82) and three years (N = 76) after application. At each examination, sealant retention and the presence of caries were recorded. The statistical analysis included the calculation of Kaplan-Meier survival curves, log-rank tests and a Cox proportional hazard regression model. No adverse events were documented. The proportion of completely intact sealants and those with minimal loss was almost identical in both groups, at 84.3% (Helioseal F; 113/134) and 81.7% (Helioseal F Plus; 107/131) after three years of observation. The regression analysis revealed an operator dependency, but no significant differences were found between the materials, the study centers, the chosen isolation technique, patient age or sex. After 3 years, 91.7% and 100.0% of all molars were free of non-cavitated carious lesions or carious cavities, respectively. It can be concluded that the new fissure sealing material can be considered as at least equivalent in terms of survival and retention behavior compared to the predecessor material.

Clinical performance of a new fissure sealant-results from a 2-year randomized clinical trial.

OBJECTIVES: The aim of this randomized clinical trial (RCT) was to explore the clinical survival of a new, Bis-GMA-free pit and fissure sealant (Helioseal F Plus) in comparison to an established control material (Helioseal F). MATERIAL AND METHODS: This in vivo study was designed as a prospective, 2-year, two-centre RCT with a split-mouth design. The initial study population consisted of 92 adolescents who were followed up 1 month (N = 89), 6 months (N = 88), 1 year (N = 85) and 2 years (N = 82) after sealant application. The attrition rate was 10.9% after 2 years. At each examination, the sealant retention and presence of caries were recorded. The statistical analysis included the calculation of Kaplan-Meier survival curves, log-rank tests and a Cox proportional hazard regression model. RESULTS: No adverse events during the application or any of the follow-up visits were documented. The proportion of completely intact sealants and those with minimal loss was almost identical in both groups at 85.9% (Helioseal F Plus) and 86.5% Helioseal F) after 2 years of observation. The regression analysis revealed operator dependency; no significant differences were found between the materials, the study centres, the chosen isolation technique and patient age or sex. CONCLUSION: The newly developed sealant can be evaluated as at least equivalent in terms of survival and retention behaviour compared to the established control material. CLINICAL RELEVANCE: The new sealant can be recommended for clinical use. With respect to the material properties (Bis-GMA-free, less light polymerisation time and better thixotropic behaviour), it offers additional advantages with clinical relevance.

Artificial intelligence-based diagnostics of molar-incisor-hypomineralization (MIH) on intraoral photographs.

OBJECTIVE: The aim of this study was to develop and validate a deep learning-based convolutional neural network (CNN) for the automated detection and categorization of teeth affected by molar-incisor-hypomineralization (MIH) on intraoral photographs. MATERIALS AND METHODS: The data set consisted of 3241 intraoral images (767 teeth with no MIH/no intervention, 76 with no MIH/atypical restoration, 742 with no MIH/sealant, 815 with demarcated opacity/no intervention, 158 with demarcated opacity/atypical restoration, 181 with demarcated opacity/sealant, 290 with enamel breakdown/no intervention, 169 with enamel breakdown/atypical restoration, and 43 with enamel breakdown/sealant). These images were divided into a training (N = 2596) and a test sample (N = 649). All images were evaluated by an expert group, and each diagnosis served as a reference standard for cyclic training and evaluation of the CNN (ResNeXt-101-32 × 8d). Statistical analysis included the calculation of contingency tables, areas under the receiver operating characteristic curve (AUCs) and saliency maps. RESULTS: The developed CNN was able to categorize teeth with MIH correctly with an overall diagnostic accuracy of 95.2%. The overall SE and SP amounted to 78.6% and 97.3%, respectively, which indicate that the CNN performed better in healthy teeth compared to those with MIH. The AUC values ranging from 0.873 (enamel breakdown/sealant) to 0.994 (atypical restoration/no MIH). CONCLUSION: It was possible to categorize the majority of clinical photographs automatically by using a trained deep learning-based CNN with an acceptably high diagnostic accuracy. CLINICAL RELEVANCE: Artificial intelligence-based dental diagnostics may support dental diagnostics in the future regardless of the need to improve accuracy.