Big Data in Epidemiology: Brave New World?

Epidemiology is experiencing a significant shift toward the utilization of big data for health monitoring and decision-making. This article discusses the recent example of the World Health Organization (WHO) global oral health status report and regional summaries, which faced criticisms due to its reliance on big data from the Global Burden of Disease (GBD) study. We address the arguments for and against the use of big data in epidemiology and provide an assessment of the value and limitations of big data epidemiology. Moreover, we provide recommendations as to how the oral health community should reconcile traditional epidemiologic approaches with big data and advanced data analytics. This Perspective article highlights the challenges of the current epidemiologic landscape, the potential of big data, and the need for a balanced approach to data utilization in epidemiology.

Trustworthy Artificial Intelligence in Dentistry: Learnings from the EU AI Act.

Artificial intelligence systems (AISs) gain relevance in dentistry, encompassing diagnostics, treatment planning, patient management, and therapy. However, questions about the generalizability, fairness, and transparency of these systems remain. Regulatory and governance bodies worldwide are aiming to address these questions using various frameworks. On March 13, 2024, members of the European Parliament approved the Artificial Intelligence Act (AIA), which emphasizes trustworthiness and human-centeredness as relevant aspects to regulate AISs beyond safety and efficacy. This review presents the AIA and similar regulatory and governance efforts in other jurisdictions and lays out that regulations such as the AIA are part of a complex ecosystem of interdependent and interwoven legal requirements and standards. Current efforts to regulate dental AISs require active input from the dental community, with participation of dental research, education, providers, and patients being relevant to shape the future of dental AISs.

Visualization of Pulpal Structures by SWIR in Endodontic Access Preparation.

Endodontic access preparation is one of the initial steps in root canal treatments and can be hindered by the obliteration of pulp canals and formation of tertiary dentin. Until now, methods for direct intraoperative visualization of the 3-dimensional anatomy of teeth have been missing. Here, we evaluate the use of shortwave infrared radiation (SWIR) for navigation during stepwise access preparation. Nine teeth (3 anteriors, 3 premolars, and 3 molars) were explanted en bloc with intact periodontium including alveolar bone and mucosa from the upper or lower jaw of human body donors. Analysis was performed at baseline as well as at preparation depths of 5 mm, 7 mm, and 9 mm, respectively. For reflection, SWIR was used at a wavelength of 1,550 nm from the occlusal direction, whereas for transillumination, SWIR was passed through each sample at the marginal gingiva from the buccal as well as oral side at a wavelength of 1,300 nm. Pulpal structures could be identified as darker areas approximately 2 mm before reaching the pulp chamber using SWIR transillumination, although they were indistinguishable under normal circumstances. Furcation areas in molars appeared with higher intensity than areas with canals. The location of pulpal structures was confirmed by superimposition of segmented micro-computed tomography (µCT) images. By radiomic analysis, significant differences between pulpal and parapulpal areas could be detected in image features. With hierarchical cluster analysis, both segments could be confirmed and associated with specific clusters. The local thickness of µCTs was calculated and correlated with SWIR transillumination images, by which a linear dependency of thickness and intensity could be demonstrated. Lastly, by in silico simulations of light propagation, dentin tubules were shown to be a crucial factor for understanding the visibility of the pulp. In conclusion, SWIR transillumination may allow direct clinical live navigation during endodontic access preparation.

Diagnostic Strategies for Restorations Management: A 70-Month RCT.

We aimed to evaluate the impact of 2 visual diagnostic strategies for assessing secondary caries and managing permanent posterior restorations on long-term survival. We conducted a diagnostic cluster-randomized clinical trial with 2 parallel groups using different diagnostic strategies: (C+AS) based on caries assessment, marginal adaptation, and marginal staining aspects of the FDI (World Dental Federation) criteria and (C) based on caries assessment using the Caries Associated with Restorations or Sealants (CARS) criteria described by the International Caries Detection and Assessment System (ICDAS). The treatment for the restoration was conducted based on the decision made following the allocated diagnostic strategy. The restorations were then clinically reevaluated for up to 71 mo. The primary outcome was restoration failure (including tooth-level failure: pain, endodontic treatment, and extraction). Cox regression analyses with shared frailty were conducted in the intention-to-treat population, and hazard ratios (HRs) and 95% confidence intervals (95% CIs) were derived. We included 727 restorations from 185 participants and reassessed 502 (69.1%) restorations during follow-up. The evaluations occurred between 6 and 71 mo. At baseline, C led to almost 4 times fewer interventions compared with the C+AS strategy. A total of 371 restorations were assessed in the C group, from which 31 (8.4%) were repaired or replaced. In contrast, the C+AS group had 356 restorations assessed, from which 113 (31.7%) were repaired or replaced. During follow-up, 34 (9.2%) failures were detected in the restorations allocated to the C group and 30 (8.4%) allocated to the C+AS group in the intention-to-treat population, with no significant difference between the groups (HR = 0.83; 95% CI = 0.51 to 1.38; P = 0.435, C+AS as reference). In conclusion, a diagnostic strategy focusing on marginal defects results in more initial interventions but does not improve longevity over the caries-focused strategy, suggesting the need for more conservative approaches.

Artificial Intelligence in Orthodontics: Critical Review.

With increasing digitalization in orthodontics, certain orthodontic manufacturing processes such as the fabrication of indirect bonding trays, aligner production, or wire bending can be automated. However, orthodontic treatment planning and evaluation remains a specialist's task and responsibility. As the prediction of growth in orthodontic patients and response to orthodontic treatment is inherently complex and individual, orthodontists make use of features gathered from longitudinal, multimodal, and standardized orthodontic data sets. Currently, these data sets are used by the orthodontist to make informed, rule-based treatment decisions. In research, artificial intelligence (AI) has been successfully applied to assist orthodontists with the extraction of relevant data from such data sets. Here, AI has been applied for the analysis of clinical imagery, such as automated landmark detection in lateral cephalograms but also for evaluation of intraoral scans or photographic data. Furthermore, AI is applied to help orthodontists with decision support for treatment decisions such as the need for orthognathic surgery or for orthodontic tooth extractions. One major challenge in current AI research in orthodontics is the limited generalizability, as most studies use unicentric data with high risks of bias. Moreover, comparing AI across different studies and tasks is virtually impossible as both outcomes and outcome metrics vary widely, and underlying data sets are not standardized. Notably, only few AI applications in orthodontics have reached full clinical maturity and regulatory approval, and researchers in the field are tasked with tackling real-world evaluation and implementation of AI into the orthodontic workflow.

Does information about MIH on dental homepages in Germany offer high quality? A systematic search and analysis.

PURPOSE: The internet is increasingly used to seek health information. A dental condition of increasing concern and public interest is molar incisor hypomineralisation (MIH), why we evaluated the information quality of German dentists 'websites on the topic of MIH. METHODS: A systematic search was performed by two independent investigators using three search engines. The information content of websites on MIH and technical, functional aspects, overall quality, and risk of bias were assessed using validated instruments (LIDA, DISCERN). Practice-related characteristics (practice type, specialization, setting, number and mean age of dentists) were recorded, and associations of these characteristics with websites' overall quality were explored using multivariable linear regression modelling. RESULTS: 70 sites were included. 52% were multipractices in urban areas (49%). The most common age group was middle-aged individuals (41-50 years). The average number of dentists/practice was 2.5. The majority met more than 50% of the DISCERN and LIDA criteria (90%, 91%). The MIH definition was frequently used (67%), MIH symptoms were described (64%), and 58% mentioned therapies. The prevalence of MIH was mentioned less frequently (48%). MIH example photographs were rarely shown (14%). In multivariable analysis, most practice-related factors were not significant for overall site quality. Only chain practices had slightly higher quality in this regard (2.2; 95% CI of 0.3-4.1). CONCLUSIONS: MIH is mentioned on a large proportion of dentists' websites. Overall technical, functional, and generic quality was high. Risk of bias is limited. While most websites provided a basic definition of MIH and its symptoms, important information for patients was missing.

Multi-modal deep learning for automated assembly of periapical radiographs.

OBJECTIVES: Periapical radiographs are oftentimes taken in series to display all teeth present in the oral cavity. Our aim was to automatically assemble such a series of periapical radiographs into an anatomically correct status using a multi-modal deep learning model. METHODS: 4,707 periapical images from 387 patients (on average, 12 images per patient) were used. Radiographs were labeled according to their field of view and the dataset split into a training, validation, and test set, stratified by patient. In addition to the radiograph the timestamp of image generation was extracted and abstracted as follows: A matrix, containing the normalized timestamps of all images of a patient was constructed, representing the order in which images were taken, providing temporal context information to the deep learning model. Using the image data together with the time sequence data a multi-modal deep learning model consisting of two residual convolutional neural networks (ResNet-152 for image data, ResNet-50 for time data) was trained. Additionally, two uni-modal models were trained on image data and time data, respectively. A custom scoring technique was used to measure model performance. RESULTS: Multi-modal deep learning outperformed both uni-modal image-based learning (p<0.001) and time-based learning (p<0.05). The multi-modal deep learning model predicted tooth labels with an F1-score, sensitivity and precision of 0.79, respectively, and an accuracy of 0.99. 37 out of 77 patient datasets were fully correctly assembled by multi-modal learning; in the remaining ones, usually only one image was incorrectly labeled. CONCLUSIONS: Multi-modal modeling allowed automated assembly of periapical radiographs and outperformed both uni-modal models. Dental machine learning models can benefit from additional data modalities. CLINICAL SIGNIFICANCE: Like humans, deep learning models may profit from multiple data sources for decision-making. We demonstrate how multi-modal learning can assist assembling periapical radiographs into an anatomically correct status. Multi-modal learning should be considered for more complex tasks, as clinically a wealth of data is usually available and could be leveraged.

Identification of Dental Implant Systems Using a Large-Scale Multicenter Data Set.

This study aimed to evaluate the efficacy of deep learning (DL) for the identification and classification of various types of dental implant systems (DISs) using a large-scale multicenter data set. We also compared the classification accuracy of DL and dental professionals. The data set, which was collected from 5 college dental hospitals and 10 private dental clinics, contained 37,442 (24.8%) periapical and 113,291 (75.2%) panoramic radiographic images and consisted of a total of 10 manufacturers and 25 different types of DISs. The classification accuracy of DL was evaluated using a pretrained and modified ResNet-50 architecture, and comparison of accuracy performance and reading time between DL and dental professionals was conducted using a self-reported questionnaire. When comparing the accuracy performance for classification of DISs, DL (accuracy: 82.0%; 95% confidence interval [CI], 75.9%-87.0%) outperformed most of the participants (mean accuracy: 23.5% ± 18.5%; 95% CI, 18.5%-32.3%), including dentists specialized (mean accuracy: 43.3% ± 20.4%; 95% CI, 12.7%-56.2%) and not specialized (mean accuracy: 16.8% ± 9.0%; 95% CI, 12.8%-20.9%) in implantology. In addition, DL tends to require lesser reading and classification time (4.5 min) than dentists who specialized (75.6 ± 31.0 min; 95% CI, 13.1-78.4) and did not specialize (91.3 ± 38.3 min; 95% CI, 74.1-108.6) in implantology. DL achieved reliable outcomes in the identification and classification of various types of DISs, and the classification accuracy performance of DL was significantly superior to that of specialized or nonspecialized dental professionals. DL as a decision support aid can be successfully used for the identification and classification of DISs encountered in clinical practice.

Preoperative risk assessment does not allow to predict root filling length using machine learning: A longitudinal study.

OBJECTIVES: First we aimed to identify significant associations between preoperative risk factors and achieving optimal root filling length (RFL) during orthograde root canal treatments (RCT) and second to predict successful RFL using machine learning. METHODS: Teeth receiving RCT at one university clinic from 2016-2020 with complete documentation were included. Successful RFL was defined to be 0-2mm of the apex, suboptimal RFL >2mm or beyond the apex. Logistic regression (logR) was used for association analyses; logR and more advanced machine learning (random forest (RF), support vector machine (SVM), decision tree (DT), gradient boosting machine (GBM) and extreme gradient boosting (XGB)) were employed for predictive modeling. RESULTS: 555 completed RCT (343 patients, female/male 32.1/67.9%) were included. In our association analysis (involving the full dataset), unsuccessful RFL was more likely in undergraduate students (US): OR 2.74, 95% CI [1.61, 4.75], p < 0.001), teeth with indistinct canal paths (OR 11.04, [2.87, 44.88], p < 0.001), root canals reduced in size (OR 2.56, [1.49, 4.46], p < 0.01), retreatments (OR 3.13, [1.6, 6.41], p < 0.001). Subgroup analyses revealed that dentists were more successful in mitigating risks than undergraduate students. Prediction of RFL on a separate testset was limitedly possible regardless of the machine learning approach. CONCLUSIONS: Achieving RFL is depending on the operator and several risk factors. The predictive performance on the technical outcome of a root canal treatment utilizing ML algorithms was insufficient. CLINICAL SIGNIFICANCE: Preoperative risk assessment is a relevant step in endodontic treatment planning. Single radiographic risk factors were significantly associated with achieving (or not achieving) optimal RFL and showed higher predictive value than a more complex risk assessment form.

Artificial Intelligence for Caries Detection: Value of Data and Information.

If increasing practitioners' diagnostic accuracy, medical artificial intelligence (AI) may lead to better treatment decisions at lower costs, while uncertainty remains around the resulting cost-effectiveness. In the present study, we assessed how enlarging the data set used for training an AI for caries detection on bitewings affects cost-effectiveness and also determined the value of information by reducing the uncertainty around other input parameters (namely, the costs of AI and the population's caries risk profile). We employed a convolutional neural network and trained it on 10%, 25%, 50%, or 100% of a labeled data set containing 29,011 teeth without and 19,760 teeth with caries lesions stemming from bitewing radiographs. We employed an established health economic modeling and analytical framework to quantify cost-effectiveness and value of information. We adopted a mixed public-private payer perspective in German health care; the health outcome was tooth retention years. A Markov model, allowing to follow posterior teeth over the lifetime of an initially 12-y-old individual, and Monte Carlo microsimulations were employed. With an increasing amount of data used to train the AI sensitivity and specificity increased nonlinearly, increasing the data set from 10% to 25% had the largest impact on accuracy and, consequently, cost-effectiveness. In the base-case scenario, AI was more effective (tooth retention for a mean [2.5%-97.5%] 62.8 [59.2-65.5] y) and less costly (378 [284-499] euros) than dentists without AI (60.4 [55.8-64.4] y; 419 [270-593] euros), with considerable uncertainty. The economic value of reducing the uncertainty around AI's accuracy or costs was limited, while information on the population's risk profile was more relevant. When developing dental AI, informed choices about the data set size may be recommended, and research toward individualized application of AI for caries detection seems warranted to optimize cost-effectiveness.

Federated Learning in Dentistry: Chances and Challenges.

Building performant and robust artificial intelligence (AI)-based applications for dentistry requires large and high-quality data sets, which usually reside in distributed data silos from multiple sources (e.g., different clinical institutes). Collaborative efforts are limited as privacy constraints forbid direct sharing across the borders of these data silos. Federated learning is a scalable and privacy-preserving framework for collaborative training of AI models without data sharing, where instead the knowledge is exchanged in form of wisdom learned from the data. This article aims at introducing the established concept of federated learning together with chances and challenges to foster collaboration on AI-based applications within the dental research community.

Towards Trustworthy AI in Dentistry.

Medical and dental artificial intelligence (AI) require the trust of both users and recipients of the AI to enhance implementation, acceptability, reach, and maintenance. Standardization is one strategy to generate such trust, with quality standards pushing for improvements in AI and reliable quality in a number of attributes. In the present brief review, we summarize ongoing activities from research and standardization that contribute to the trustworthiness of medical and, specifically, dental AI and discuss the role of standardization and some of its key elements. Furthermore, we discuss how explainable AI methods can support the development of trustworthy AI models in dentistry. In particular, we demonstrate the practical benefits of using explainable AI on the use case of caries prediction on near-infrared light transillumination images.

Benchmarking Deep Learning Models for Tooth Structure Segmentation.

A wide range of deep learning (DL) architectures with varying depths are available, with developers usually choosing one or a few of them for their specific task in a nonsystematic way. Benchmarking (i.e., the systematic comparison of state-of-the art architectures on a specific task) may provide guidance in the model development process and may allow developers to make better decisions. However, comprehensive benchmarking has not been performed in dentistry yet. We aimed to benchmark a range of architecture designs for 1 specific, exemplary case: tooth structure segmentation on dental bitewing radiographs. We built 72 models for tooth structure (enamel, dentin, pulp, fillings, crowns) segmentation by combining 6 different DL network architectures (U-Net, U-Net++, Feature Pyramid Networks, LinkNet, Pyramid Scene Parsing Network, Mask Attention Network) with 12 encoders from 3 different encoder families (ResNet, VGG, DenseNet) of varying depth (e.g., VGG13, VGG16, VGG19). On each model design, 3 initialization strategies (ImageNet, CheXpert, random initialization) were applied, resulting overall into 216 trained models, which were trained up to 200 epochs with the Adam optimizer (learning rate = 0.0001) and a batch size of 32. Our data set consisted of 1,625 human-annotated dental bitewing radiographs. We used a 5-fold cross-validation scheme and quantified model performances primarily by the F1-score. Initialization with ImageNet or CheXpert weights significantly outperformed random initialization (P < 0.05). Deeper and more complex models did not necessarily perform better than less complex alternatives. VGG-based models were more robust across model configurations, while more complex models (e.g., from the ResNet family) achieved peak performances. In conclusion, initializing models with pretrained weights may be recommended when training models for dental radiographic analysis. Less complex model architectures may be competitive alternatives if computational resources and training time are restricting factors. Models developed and found superior on nondental data sets may not show this behavior for dental domain-specific tasks.

Normative Approaches for Oral Health: Standards, Specifications, and Guidelines.

Normative approaches have been developed with the aim of providing high-quality methods and strict criteria that, when applied correctly, lead to reliable results. Standards, specifications, and guidelines are needed to facilitate exchange of goods or information and secure comparability of data derived from different laboratories and sources. They are available along the whole flow from study development to test selection, study conduct, and reporting and are widely used for the evaluation of medical devices, market approval, and harmonization of terms and devices. Standards are developed by specific national and international organizations or by dedicated interest groups, mainly scientists in their respective fields. ISO (International Organization for Standardization) standards are developed following stringent regulations, and groups of experts formulate such standards. They should come from different areas (multistakeholder approach) to have as much and as broad input as possible and to avoid single-interest dominance. However, the presence of academia in such groups has been comparatively low. There is a clear need and responsibility of the oral health community to participate in the development of normative documents to provide methodological knowledge and experience, balance the interests of other stakeholders, and finally improve oral health. This will help to ensure that rapidly advancing fields of research, such as the oral health impacts of COVID-19 or the application of artificial intelligence in dentistry, benefit from standardization of approaches and reporting.

Data Dentistry: How Data Are Changing Clinical Care and Research.

Data are a key resource for modern societies and expected to improve quality, accessibility, affordability, safety, and equity of health care. Dental care and research are currently transforming into what we term data dentistry, with 3 main applications: 1) medical data analysis uses deep learning, allowing one to master unprecedented amounts of data (language, speech, imagery) and put them to productive use. 2) Data-enriched clinical care integrates data from individual (e.g., demographic, social, clinical and omics data, consumer data), setting (e.g., geospatial, environmental, provider-related data), and systems level (payer or regulatory data to characterize input, throughput, output, and outcomes of health care) to provide a comprehensive and continuous real-time assessment of biologic perturbations, individual behaviors, and context. Such care may contribute to a deeper understanding of health and disease and a more precise, personalized, predictive, and preventive care. 3) Data for research include open research data and data sharing, allowing one to appraise, benchmark, pool, replicate, and reuse data. Concerns and confidence into data-driven applications, stakeholders' and system's capabilities, and lack of data standardization and harmonization currently limit the development and implementation of data dentistry. Aspects of bias and data-user interaction require attention. Action items for the dental community circle around increasing data availability, refinement, and usage; demonstrating safety, value, and usefulness of applications; educating the dental workforce and consumers; providing performant and standardized infrastructure and processes; and incentivizing and adopting open data and data sharing.