Evaluation of AI-generated responses by different artificial intelligence chatbots to the clinical decision-making case-based questions in oral and maxillofacial surgery.

OBJECTIVES: This study aims to evaluate the correctness of the generated answers by Google Bard, GPT-3.5, GPT-4, Claude-Instant, and Bing chatbots to decision-making clinical questions in the oral and maxillofacial surgery (OMFS) area. STUDY DESIGN: A group of 3 board-certified oral and maxillofacial surgeons designed a questionnaire with 50 case-based questions in multiple-choice and open-ended formats. Answers of chatbots to multiple-choice questions were examined against the chosen option by 3 referees. The chatbots' answers to the open-ended questions were evaluated based on the modified global quality scale. A P-value under .05 was considered significant. RESULTS: Bard, GPT-3.5, GPT-4, Claude-Instant, and Bing answered 34%, 36%, 38%, 38%, and 26% of the questions correctly, respectively. In open-ended questions, GPT-4 scored the most answers evaluated as grades "4" or "5," and Bing scored the most answers evaluated as grades "1" or "2." There were no statistically significant differences between the 5 chatbots in responding to the open-ended (P = .275) and multiple-choice (P = .699) questions. CONCLUSION: Considering the major inaccuracies in the responses of chatbots, despite their relatively good performance in answering open-ended questions, this technology yet cannot be trusted as a consultant for clinicians in decision-making situations.

Accuracy and consistency of chatbots versus clinicians for answering pediatric dentistry questions: A pilot study.

OBJECTIVES: Artificial Intelligence has applications such as Large Language Models (LLMs), which simulate human-like conversations. The potential of LLMs in healthcare is not fully evaluated. This pilot study assessed the accuracy and consistency of chatbots and clinicians in answering common questions in pediatric dentistry. METHODS: Two expert pediatric dentists developed thirty true or false questions involving different aspects of pediatric dentistry. Publicly accessible chatbots (Google Bard, ChatGPT4, ChatGPT 3.5, Llama, Sage, Claude 2 100k, Claude-instant, Claude-instant-100k, and Google Palm) were employed to answer the questions (3 independent new conversations). Three groups of clinicians (general dentists, pediatric specialists, and students; n = 20/group) also answered. Responses were graded by two pediatric dentistry faculty members, along with a third independent pediatric dentist. Resulting accuracies (percentage of correct responses) were compared using analysis of variance (ANOVA), and post-hoc pairwise group comparisons were corrected using Tukey's HSD method. ACronbach's alpha was calculated to determine consistency. RESULTS: Pediatric dentists were significantly more accurate (mean±SD 96.67 %± 4.3 %) than other clinicians and chatbots (p < 0.001). General dentists (88.0 % ± 6.1 %) also demonstrated significantly higher accuracy than chatbots (p < 0.001), followed by students (80.8 %±6.9 %). ChatGPT showed the highest accuracy (78 %±3 %) among chatbots. All chatbots except ChatGPT3.5 showed acceptable consistency (Cronbach alpha>0.7). CLINICAL SIGNIFICANCE: Based on this pilot study, chatbots may be valuable adjuncts for educational purposes and for distributing information to patients. However, they are not yet ready to serve as substitutes for human clinicians in diagnostic decision-making. CONCLUSION: In this pilot study, chatbots showed lower accuracy than dentists. Chatbots may not yet be recommended for clinical pediatric dentistry.

Performance of AI chatbots on controversial topics in oral medicine, pathology, and radiology.

OBJECTIVES: In this study, we assessed 6 different artificial intelligence (AI) chatbots (Bing, GPT-3.5, GPT-4, Google Bard, Claude, Sage) responses to controversial and difficult questions in oral pathology, oral medicine, and oral radiology. STUDY DESIGN: The chatbots' answers were evaluated by board-certified specialists using a modified version of the global quality score on a 5-point Likert scale. The quality and validity of chatbot citations were evaluated. RESULTS: Claude had the highest mean score of 4.341 ± 0.582 for oral pathology and medicine. Bing had the lowest scores of 3.447 ± 0.566. In oral radiology, GPT-4 had the highest mean score of 3.621 ± 1.009 and Bing the lowest score of 2.379 ± 0.978. GPT-4 achieved the highest mean score of 4.066 ± 0.825 for performance across all disciplines. 82 out of 349 (23.50%) of generated citations from chatbots were fake. CONCLUSIONS: The most superior chatbot in providing high-quality information for controversial topics in various dental disciplines was GPT-4. Although the majority of chatbots performed well, it is suggested that developers of AI medical chatbots incorporate scientific citation authenticators to validate the outputted citations given the relatively high number of fabricated citations.

Current Applications of Artificial Intelligence for Pediatric Dentistry: A Systematic Review and Meta-Analysis.

Purpose: To systematically evaluate artificial intelligence applications for diagnostic and treatment planning possibilities in pediatric dentistry. Methods: PubMed®, EMBASE®, Scopus, Web of Science™, IEEE, medRxiv, arXiv, and Google Scholar were searched using specific search queries. The Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) checklist was used to assess the risk of bias assessment of the included studies. Results: Based on the initial screening, 33 eligible studies were included (among 3,542). Eleven studies appeared to have low bias risk across all QUADAS-2 domains. Most applications focused on early childhood caries diagnosis and prediction, tooth identification, oral health evaluation, and supernumerary tooth identification. Six studies evaluated AI tools for mesiodens or supernumerary tooth identification on radigraphs, four for primary tooth identification and/or numbering, seven studies to detect caries on radiographs, and 12 to predict early childhood caries. For these four tasks, the reported accuracy of AI varied from 60 percent to 99 percent, sensitivity was from 20 percent to 100 percent, specificity was from 49 percent to 100 percent, F1-score was from 60 percent to 97 percent, and the area-under-the-curve varied from 87 percent to 100 percent. Conclusions: The overall body of evidence regarding artificial intelligence applications in pediatric dentistry does not allow for firm conclusions. For a wide range of applications, AI shows promising accuracy. Future studies should focus on a comparison of AI against the standard of care and employ a set of standardized outcomes and metrics to allow comparison across studies.

Artificial Intelligence in Endodontic Education.

AIMS: The future dental and endodontic education must adapt to the current digitalized healthcare system in a hyper-connected world. The purpose of this scoping review was to investigate the ways an endodontic education curriculum could benefit from the implementation of artificial intelligence (AI) and overcome the limitations of this technology in the delivery of healthcare to patients. METHODS: An electronic search was carried out up to December 2023 using MEDLINE, Web of Science, Cochrane Library, and a manual search of reference literature. Grey literature, ongoing clinical trials were also searched using ClinicalTrials.gov. RESULTS: The search identified 251 records, of which 35 were deemed relevant to artificial intelligence (AI) and Endodontic education. Areas in which AI might aid students with their didactic and clinical endodontic education were identified as follows: 1) radiographic interpretation; 2) differential diagnosis; 3) treatment planning and decision-making; 4) case difficulty assessment; 5) preclinical training; 6) advanced clinical simulation and case-based training, 7) real-time clinical guidance; 8) autonomous systems and robotics; 9) progress evaluation and personalized education; 10) calibration and standardization. CONCLUSIONS: AI in endodontic education will support clinical and didactic teaching through individualized feedback; enhanced, augmented, and virtually generated training aids; automated detection and diagnosis; treatment planning and decision support; and AI-based student progress evaluation, and personalized education. Its implementation will inarguably change the current concept of teaching Endodontics. Dental educators would benefit from introducing AI in clinical and didactic pedagogy; however, they must be aware of AI's limitations and challenges to overcome.

Deep learning for tooth identification and numbering on dental radiography: a systematic review and meta-analysis.

OBJECTIVES: Improved tools based on deep learning can be used to accurately number and identify teeth. This study aims to review the use of deep learning in tooth numbering and identification. METHODS: An electronic search was performed through October 2023 on PubMed, Scopus, Cochrane, Google Scholar, IEEE, arXiv, and medRxiv. Studies that used deep learning models with segmentation, object detection, or classification tasks for teeth identification and numbering of human dental radiographs were included. For risk of bias assessment, included studies were critically analysed using quality assessment of diagnostic accuracy studies (QUADAS-2). To generate plots for meta-analysis, MetaDiSc and STATA 17 (StataCorp LP, College Station, TX, USA) were used. Pooled outcome diagnostic odds ratios (DORs) were determined through calculation. RESULTS: The initial search yielded 1618 studies, of which 29 were eligible based on the inclusion criteria. Five studies were found to have low bias across all domains of the QUADAS-2 tool. Deep learning has been reported to have an accuracy range of 81.8%-99% in tooth identification and numbering and a precision range of 84.5%-99.94%. Furthermore, sensitivity was reported as 82.7%-98% and F1-scores ranged from 87% to 98%. Sensitivity was 75.5%-98% and specificity was 79.9%-99%. Only 6 studies found the deep learning model to be less than 90% accurate. The average DOR of the pooled data set was 1612, the sensitivity was 89%, the specificity was 99%, and the area under the curve was 96%. CONCLUSION: Deep learning models successfully can detect, identify, and number teeth on dental radiographs. Deep learning-powered tooth numbering systems can enhance complex automated processes, such as accurately reporting which teeth have caries, thus aiding clinicians in making informed decisions during clinical practice.

Artificial intelligence for classification and detection of oral mucosa lesions on photographs: a systematic review and meta-analysis.

OBJECTIVE: This study aimed to review and synthesize studies using artificial intelligence (AI) for classifying, detecting, or segmenting oral mucosal lesions on photographs. MATERIALS AND METHOD: Inclusion criteria were (1) studies employing AI to (2) classify, detect, or segment oral mucosa lesions, (3) on oral photographs of human subjects. Included studies were assessed for risk of bias using Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2). A PubMed, Scopus, Embase, Web of Science, IEEE, arXiv, medRxiv, and grey literature (Google Scholar) search was conducted until June 2023, without language limitation. RESULTS: After initial searching, 36 eligible studies (from 8734 identified records) were included. Based on QUADAS-2, only 7% of studies were at low risk of bias for all domains. Studies employed different AI models and reported a wide range of outcomes and metrics. The accuracy of AI for detecting oral mucosal lesions ranged from 74 to 100%, while that for clinicians un-aided by AI ranged from 61 to 98%. Pooled diagnostic odds ratio for studies which evaluated AI for diagnosing or discriminating potentially malignant lesions was 155 (95% confidence interval 23-1019), while that for cancerous lesions was 114 (59-221). CONCLUSIONS: AI may assist in oral mucosa lesion screening while the expected accuracy gains or further health benefits remain unclear so far. CLINICAL RELEVANCE: Artificial intelligence assists oral mucosa lesion screening and may foster more targeted testing and referral in the hands of non-specialist providers, for example. So far, it remains unclear if accuracy gains compared with specialized can be realized.

Artificial Intelligence for Detection of External Cervical Resorption Using Label-Efficient Self-Supervised Learning Method.

INTRODUCTION: The aim of this study was to leverage label-efficient self-supervised learning (SSL) to train a model that can detect ECR and differentiate it from caries. METHODS: Periapical (PA) radiographs of teeth with ECR defects were collected. Two board-certified endodontists reviewed PA radiographs and cone beam computed tomographic (CBCT) images independently to determine presence of ECR (ground truth). Radiographic data were divided into 3 regions of interest (ROIs): healthy teeth, teeth with ECR, and teeth with caries. Nine contrastive SSL models (SimCLR v2, MoCo v2, BYOL, DINO, NNCLR, SwAV, MSN, Barlow Twins, and SimSiam) were implemented in the assessment alongside 7 baseline deep learning models (ResNet-18, ResNet-50, VGG16, DenseNet, MobileNetV2, ResNeXt-50, and InceptionV3). A 10-fold cross-validation strategy and a hold-out test set were employed for model evaluation. Model performance was assessed via various metrics including classification accuracy, precision, recall, and F1-score. RESULTS: Included were 190 PA radiographs, composed of 470 ROIs. Results from 10-fold cross-validation demonstrated that most SSL models outperformed the transfer learning baseline models, with DINO achieving the highest mean accuracy (85.64 ± 4.56), significantly outperforming 13 other models (P < .05). DINO reached the highest test set (ie, 3 ROIs) accuracy (84.09%) while MoCo v2 exhibited the highest recall and F1-score (77.37% and 82.93%, respectively). CONCLUSIONS: This study showed that AI can assist clinicians in detecting ECR and differentiating it from caries. Additionally, it introduced the application of SSL in detecting ECR, emphasizing that SSL-based models can outperform transfer learning baselines and reduce reliance on large, labeled datasets.

Deep learning for diagnosis of head and neck cancers through radiographic data: a systematic review and meta-analysis.

PURPOSE: This study aims to review deep learning applications for detecting head and neck cancer (HNC) using magnetic resonance imaging (MRI) and radiographic data. METHODS: Through January 2023, a PubMed, Scopus, Embase, Google Scholar, IEEE, and arXiv search were carried out. The inclusion criteria were implementing head and neck medical images (computed tomography (CT), positron emission tomography (PET), MRI, Planar scans, and panoramic X-ray) of human subjects with segmentation, object detection, and classification deep learning models for head and neck cancers. The risk of bias was rated with the quality assessment of diagnostic accuracy studies (QUADAS-2) tool. For the meta-analysis diagnostic odds ratio (DOR) was calculated. Deeks' funnel plot was used to assess publication bias. MIDAS and Metandi packages were used to analyze diagnostic test accuracy in STATA. RESULTS: From 1967 studies, 32 were found eligible after the search and screening procedures. According to the QUADAS-2 tool, 7 included studies had a low risk of bias for all domains. According to the results of all included studies, the accuracy varied from 82.6 to 100%. Additionally, specificity ranged from 66.6 to 90.1%, sensitivity from 74 to 99.68%. Fourteen studies that provided sufficient data were included for meta-analysis. The pooled sensitivity was 90% (95% CI 0.820.94), and the pooled specificity was 92% (CI 95% 0.87-0.96). The DORs were 103 (27-251). Publication bias was not detected based on the p-value of 0.75 in the meta-analysis. CONCLUSION: With a head and neck screening deep learning model, detectable screening processes can be enhanced with high specificity and sensitivity.

Validity and reliability of artificial intelligence chatbots as public sources of information on endodontics.

AIM: This study aimed to evaluate and compare the validity and reliability of responses provided by GPT-3.5, Google Bard, and Bing to frequently asked questions (FAQs) in the field of endodontics. METHODOLOGY: FAQs were formulated by expert endodontists (n = 10) and collected through GPT-3.5 queries (n = 10), with every question posed to each chatbot three times. Responses (N = 180) were independently evaluated by two board-certified endodontists using a modified Global Quality Score (GQS) on a 5-point Likert scale (5: strongly agree; 4: agree; 3: neutral; 2: disagree; 1: strongly disagree). Disagreements on scoring were resolved through evidence-based discussions. The validity of responses was analysed by categorizing scores into valid or invalid at two thresholds: The low threshold was set at score ≥4 for all three responses whilst the high threshold was set at score 5 for all three responses. Fisher's exact test was conducted to compare the validity of responses between chatbots. Cronbach's alpha was calculated to assess the reliability by assessing the consistency of repeated responses for each chatbot. RESULTS: All three chatbots provided answers to all questions. Using the low-threshold validity test (GPT-3.5: 95%; Google Bard: 85%; Bing: 75%), there was no significant difference between the platforms (p > .05). When using the high-threshold validity test, the chatbot scores were substantially lower (GPT-3.5: 60%; Google Bard: 15%; Bing: 15%). The validity of GPT-3.5 responses was significantly higher than Google Bard and Bing (p = .008). All three chatbots achieved an acceptable level of reliability (Cronbach's alpha >0.7). CONCLUSIONS: GPT-3.5 provided more credible information on topics related to endodontics compared to Google Bard and Bing.

A deep learning framework to scale linear facial measurements to actual size using horizontal visible iris diameter: a study on an Iranian population.

Digital images allow for the objective evaluation of facial appearance and abnormalities as well as treatment outcomes and stability. With the advancement of technology, manual clinical measurements can be replaced with fully automatic photographic assessments. However, obtaining millimetric measurements on photographs does not provide clinicians with their actual value due to different image magnification ratios. A deep learning tool was developed to estimate linear measurements on images with unknown magnification using the iris diameter. A framework was designed to segment the eyes' iris and calculate the horizontal visible iris diameter (HVID) in pixels. A constant value of 12.2 mm was assigned as the HVID value in all the photographs. A vertical and a horizontal distance were measured in pixels on photographs of 94 subjects and were estimated in millimeters by calculating the magnification ratio using HVID. Manual measurement of the distances was conducted on the subjects and the actual and estimated amounts were compared using Bland-Altman analysis. The obtained error was calculated as mean absolute percentage error (MAPE) of 2.9% and 4.3% in horizontal and vertical measurements. Our study shows that due to the consistent size and narrow range of HVID values, the iris diameter can be used as a reliable scale to calibrate the magnification of the images to obtain precise measurements in further research.

Ethical considerations on artificial intelligence in dentistry: A framework and checklist.

OBJECTIVE: Artificial Intelligence (AI) refers to the ability of machines to perform cognitive and intellectual human tasks. In dentistry, AI offers the potential to enhance diagnostic accuracy, improve patient outcomes and streamline workflows. The present study provides a framework and a checklist to evaluate AI applications in dentistry from this perspective. METHODS: Lending from existing guidance documents, an initial draft of the checklist and an explanatory paper were derived and discussed among the groups members. RESULTS: The checklist was consented to in an anonymous voting process by 29 Topic Group Dental Diagnostics and Digital Dentistry, ITU/WHO Focus Group AI on Health's members. Overall, 11 principles were identified (diversity, transparency, wellness, privacy protection, solidarity, equity, prudence, law and governance, sustainable development, accountability, and responsibility, respect of autonomy, decision-making). CONCLUSIONS: Providers, patients, researchers, industry, and other stakeholders should consider these principles when developing, implementing, or receiving AI applications in dentistry. CLINICAL SIGNIFICANCE: While AI has become increasingly commonplace in dentistry, there are ethical concerns around its usage, and users (providers, patients, and other stakeholders), as well as the industry should consider these when developing, implementing, or receiving AI applications based on comprehensive framework to address the associated ethical challenges.

Federated vs Local vs Central Deep Learning of Tooth Segmentation on Panoramic Radiographs.

OBJECTIVE: Federated Learning (FL) enables collaborative training of artificial intelligence (AI) models from multiple data sources without directly sharing data. Due to the large amount of sensitive data in dentistry, FL may be particularly relevant for oral and dental research and applications. This study, for the first time, employed FL for a dental task, automated tooth segmentation on panoramic radiographs. METHODS: We employed a dataset of 4,177 panoramic radiographs collected from nine different centers (n = 143 to n = 1881 per center) across the globe and used FL to train a machine learning model for tooth segmentation. FL performance was compared against Local Learning (LL), i.e., training models on isolated data from each center (assuming data sharing not to be an option). Further, the performance gap to Central Learning (CL), i.e., training on centrally pooled data (based on data sharing agreements) was quantified. Generalizability of models was evaluated on a pooled test dataset from all centers. RESULTS: For 8 out of 9 centers, FL outperformed LL with statistical significance (p<0.05); only the center providing the largest amount of data FL did not have such an advantage. For generalizability, FL outperformed LL across all centers. CL surpassed both FL and LL for performance and generalizability. CONCLUSION: If data pooling (for CL) is not feasible, FL is shown to be a useful alternative to train performant and, more importantly, generalizable deep learning models in dentistry, where data protection barriers are high. CLINICAL SIGNIFICANCE: This study proves the validity and utility of FL in the field of dentistry, which encourages researchers to adopt this method to improve the generalizability of dental AI models and ease their transition to the clinical environment.

Super-Resolution of Dental Panoramic Radiographs Using Deep Learning: A Pilot Study.

Using super-resolution (SR) algorithms, an image with a low resolution can be converted into a high-quality image. Our objective was to compare deep learning-based SR models to a conventional approach for improving the resolution of dental panoramic radiographs. A total of 888 dental panoramic radiographs were obtained. Our study involved five state-of-the-art deep learning-based SR approaches, including SR convolutional neural networks (SRCNN), SR generative adversarial network (SRGAN), U-Net, Swin for image restoration (SwinIr), and local texture estimator (LTE). Their results were compared with one another and with conventional bicubic interpolation. The performance of each model was evaluated using the metrics of mean squared error (MSE), peak signal-to-noise ratio (PNSR), structural similarity index (SSIM), and mean opinion score by four experts (MOS). Among all the models evaluated, the LTE model presented the highest performance, with MSE, SSIM, PSNR, and MOS results of 7.42 ± 0.44, 39.74 ± 0.17, 0.919 ± 0.003, and 3.59 ± 0.54, respectively. Additionally, compared with low-resolution images, the output of all the used approaches showed significant improvements in MOS evaluation. A significant enhancement in the quality of panoramic radiographs can be achieved by SR. The LTE model outperformed the other models.

Deep learning: A primer for dentists and dental researchers.

OBJECTIVES: Despite deep learning's wide adoption in dental artificial intelligence (AI) research, researchers from other dental fields and, more so, dental professionals may find it challenging to understand and interpret deep learning studies, their employed methods, and outcomes. The objective of this primer is to explain the basic concept of deep learning. It will lay out the commonly used terms, and describe different deep learning approaches, their methods, and outcomes. METHODS: Our research is based on the latest review studies, medical primers, as well as the state-of-the-art research on AI and deep learning, which have been gathered in the current study. RESULTS: In this study, a basic understanding of deep learning models and various approaches to deep learning is presented. An overview of data management strategies for deep learning projects is presented, including data collection, data curation, data annotation, and data preprocessing. Additionally, we provided a step-by-step guide for completing a real-world project. CONCLUSION: Researchers and clinicians can benefit from this study by gaining insight into deep learning. It can be used to critically appraise existing work or plan new deep learning projects. CLINICAL SIGNIFICANCE: This study may be useful to dental researchers and professionals who are assessing and appraising deep learning studies within the field of dentistry.

Deep Learning for Detection of Periapical Radiolucent Lesions: A Systematic Review and Meta-analysis of Diagnostic Test Accuracy.

INTRODUCTION: The aim of this systematic review and meta-analysis was to investigate the overall accuracy of deep learning models in detecting periapical (PA) radiolucent lesions in dental radiographs, when compared to expert clinicians. METHODS: Electronic databases of Medline (via PubMed), Embase (via Ovid), Scopus, Google Scholar, and arXiv were searched. Quality of eligible studies was assessed by using Quality Assessment and Diagnostic Accuracy Tool-2. Quantitative analyses were conducted using hierarchical logistic regression for meta-analyses on diagnostic accuracy. Subgroup analyses on different image modalities (PA radiographs, panoramic radiographs, and cone beam computed tomographic images) and on different deep learning tasks (classification, segmentation, object detection) were conducted. Certainty of evidence was assessed by using Grading of Recommendations Assessment, Development, and Evaluation system. RESULTS: A total of 932 studies were screened. Eighteen studies were included in the systematic review, out of which 6 studies were selected for quantitative analyses. Six studies had low risk of bias. Twelve studies had risk of bias. Pooled sensitivity, specificity, positive likelihood ratio, negative likelihood ratio, and diagnostic odds ratio of included studies (all image modalities; all tasks) were 0.925 (95% confidence interval [CI], 0.862-0.960), 0.852 (95% CI, 0.810-0.885), 6.261 (95% CI, 4.717-8.311), 0.087 (95% CI, 0.045-0.168), and 71.692 (95% CI, 29.957-171.565), respectively. No publication bias was detected (Egger's test, P = .82). Grading of Recommendations Assessment, Development and Evaluationshowed a "high" certainty of evidence for the studies included in the meta-analyses. CONCLUSION: Compared to expert clinicians, deep learning showed highly accurate results in detecting PA radiolucent lesions in dental radiographs. Most studies had risk of bias. There was a lack of prospective studies.

Deep learning for tooth identification and enumeration in panoramic radiographs.

Background: Dentists begin the diagnosis by identifying and enumerating teeth. Panoramic radiographs are widely used for tooth identification due to their large field of view and low exposure dose. The automatic numbering of teeth in panoramic radiographs can assist clinicians in avoiding errors. Deep learning has emerged as a promising tool for automating tasks. Our goal is to evaluate the accuracy of a two-step deep learning method for tooth identification and enumeration in panoramic radiographs. Materials and Methods: In this retrospective observational study, 1007 panoramic radiographs were labeled by three experienced dentists. It involved drawing bounding boxes in two distinct ways: one for teeth and one for quadrants. All images were preprocessed using the contrast-limited adaptive histogram equalization method. First, panoramic images were allocated to a quadrant detection model, and the outputs of this model were provided to the tooth numbering models. A faster region-based convolutional neural network model was used in each step. Results: Average precision (AP) was calculated in different intersection-over-union thresholds. The AP50 of quadrant detection and tooth enumeration was 100% and 95%, respectively. Conclusion: We have obtained promising results with a high level of AP using our two-step deep learning framework for automatic tooth enumeration on panoramic radiographs. Further research should be conducted on diverse datasets and real-life situations.

Emulating Clinical Diagnostic Reasoning for Jaw Cysts with Machine Learning.

The detection and classification of cystic lesions of the jaw is of high clinical relevance and represents a topic of interest in medical artificial intelligence research. The human clinical diagnostic reasoning process uses contextual information, including the spatial relation of the detected lesion to other anatomical structures, to establish a preliminary classification. Here, we aimed to emulate clinical diagnostic reasoning step by step by using a combined object detection and image segmentation approach on panoramic radiographs (OPGs). We used a multicenter training dataset of 855 OPGs (all positives) and an evaluation set of 384 OPGs (240 negatives). We further compared our models to an international human control group of ten dental professionals from seven countries. The object detection model achieved an average precision of 0.42 (intersection over union (IoU): 0.50, maximal detections: 100) and an average recall of 0.394 (IoU: 0.50-0.95, maximal detections: 100). The classification model achieved a sensitivity of 0.84 for odontogenic cysts and 0.56 for non-odontogenic cysts as well as a specificity of 0.59 for odontogenic cysts and 0.84 for non-odontogenic cysts (IoU: 0.30). The human control group achieved a sensitivity of 0.70 for odontogenic cysts, 0.44 for non-odontogenic cysts, and 0.56 for OPGs without cysts as well as a specificity of 0.62 for odontogenic cysts, 0.95 for non-odontogenic cysts, and 0.76 for OPGs without cysts. Taken together, our results show that a combined object detection and image segmentation approach is feasible in emulating the human clinical diagnostic reasoning process in classifying cystic lesions of the jaw.

Deep learning in periodontology and oral implantology: A scoping review.

Deep learning (DL) has been employed for a wide range of tasks in dentistry. We aimed to systematically review studies employing DL for periodontal and implantological purposes. A systematic electronic search was conducted on four databases (Medline via PubMed, Google Scholar, Scopus, and Embase) and a repository (ArXiv) for publications after 2010, without any limitation on language. In the present review, we included studies that reported deep learning models' performance on periodontal or oral implantological tasks. Given the heterogeneities in the included studies, no meta-analysis was performed. The risk of bias was assessed using the QUADAS-2 tool. We included 47 studies: focusing on imaging data (n = 20) and non-imaging data in periodontology (n = 12), or dental implantology (n = 15). The detection of periodontitis and gingivitis or periodontal bone loss, the classification of dental implant systems, or the prediction of treatment outcomes in periodontology and implantology were major use cases. The performance of the models was generally high. However, it varied given the employed methods (which includes various types of convolutional neural networks (CNN) and multi-layered perceptron (MLP)), the variety in specific modeling tasks, as well as the chosen and reported outcomes, outcome measures and outcome level. Only a few studies (n = 7) showed a low risk of bias across all assessed domains. A growing number of studies evaluated DL for periodontal or implantological objectives. Heterogeneity in study design, poor reporting and a high risk of bias severely limit the comparability of studies and the robustness of the overall evidence.