Advancements in diagnosing oral potentially malignant disorders: leveraging Vision transformers for multi-class detection.

OBJECTIVES: Diagnosing oral potentially malignant disorders (OPMD) is critical to prevent oral cancer. This study aims to automatically detect and classify the most common pre-malignant oral lesions, such as leukoplakia and oral lichen planus (OLP), and distinguish them from oral squamous cell carcinomas (OSCC) and healthy oral mucosa on clinical photographs using vision transformers. METHODS: 4,161 photographs of healthy mucosa, leukoplakia, OLP, and OSCC were included. Findings were annotated pixel-wise and reviewed by three clinicians. The photographs were divided into 3,337 for training and validation and 824 for testing. The training and validation images were further divided into five folds with stratification. A Mask R-CNN with a Swin Transformer was trained five times with cross-validation, and the held-out test split was used to evaluate the model performance. The precision, F1-score, sensitivity, specificity, and accuracy were calculated. The area under the receiver operating characteristics curve (AUC) and the confusion matrix of the most effective model were presented. RESULTS: The detection of OSCC with the employed model yielded an F1 of 0.852 and AUC of 0.974. The detection of OLP had an F1 of 0.825 and AUC of 0.948. For leukoplakia the F1 was 0.796 and the AUC was 0.938. CONCLUSIONS: OSCC were effectively detected with the employed model, whereas the detection of OLP and leukoplakia was moderately effective. CLINICAL RELEVANCE: Oral cancer is often detected in advanced stages. The demonstrated technology may support the detection and observation of OPMD to lower the disease burden and identify malignant oral cavity lesions earlier.

Combining public datasets for automated tooth assessment in panoramic radiographs.

OBJECTIVE: Panoramic radiographs (PRs) provide a comprehensive view of the oral and maxillofacial region and are used routinely to assess dental and osseous pathologies. Artificial intelligence (AI) can be used to improve the diagnostic accuracy of PRs compared to bitewings and periapical radiographs. This study aimed to evaluate the advantages and challenges of using publicly available datasets in dental AI research, focusing on solving the novel task of predicting tooth segmentations, FDI numbers, and tooth diagnoses, simultaneously. MATERIALS AND METHODS: Datasets from the OdontoAI platform (tooth instance segmentations) and the DENTEX challenge (tooth bounding boxes with associated diagnoses) were combined to develop a two-stage AI model. The first stage implemented tooth instance segmentation with FDI numbering and extracted regions of interest around each tooth segmentation, whereafter the second stage implemented multi-label classification to detect dental caries, impacted teeth, and periapical lesions in PRs. The performance of the automated tooth segmentation algorithm was evaluated using a free-response receiver-operating-characteristics (FROC) curve and mean average precision (mAP) metrics. The diagnostic accuracy of detection and classification of dental pathology was evaluated with ROC curves and F1 and AUC metrics. RESULTS: The two-stage AI model achieved high accuracy in tooth segmentations with a FROC score of 0.988 and a mAP of 0.848. High accuracy was also achieved in the diagnostic classification of impacted teeth (F1 = 0.901, AUC = 0.996), whereas moderate accuracy was achieved in the diagnostic classification of deep caries (F1 = 0.683, AUC = 0.960), early caries (F1 = 0.662, AUC = 0.881), and periapical lesions (F1 = 0.603, AUC = 0.974). The model's performance correlated positively with the quality of annotations in the used public datasets. Selected samples from the DENTEX dataset revealed cases of missing (false-negative) and incorrect (false-positive) diagnoses, which negatively influenced the performance of the AI model. CONCLUSIONS: The use and pooling of public datasets in dental AI research can significantly accelerate the development of new AI models and enable fast exploration of novel tasks. However, standardized quality assurance is essential before using the datasets to ensure reliable outcomes and limit potential biases.

Intra-oral scan segmentation using deep learning.

OBJECTIVE: Intra-oral scans and gypsum cast scans (OS) are widely used in orthodontics, prosthetics, implantology, and orthognathic surgery to plan patient-specific treatments, which require teeth segmentations with high accuracy and resolution. Manual teeth segmentation, the gold standard up until now, is time-consuming, tedious, and observer-dependent. This study aims to develop an automated teeth segmentation and labeling system using deep learning. MATERIAL AND METHODS: As a reference, 1750 OS were manually segmented and labeled. A deep-learning approach based on PointCNN and 3D U-net in combination with a rule-based heuristic algorithm and a combinatorial search algorithm was trained and validated on 1400 OS. Subsequently, the trained algorithm was applied to a test set consisting of 350 OS. The intersection over union (IoU), as a measure of accuracy, was calculated to quantify the degree of similarity between the annotated ground truth and the model predictions. RESULTS: The model achieved accurate teeth segmentations with a mean IoU score of 0.915. The FDI labels of the teeth were predicted with a mean accuracy of 0.894. The optical inspection showed excellent position agreements between the automatically and manually segmented teeth components. Minor flaws were mostly seen at the edges. CONCLUSION: The proposed method forms a promising foundation for time-effective and observer-independent teeth segmentation and labeling on intra-oral scans. CLINICAL SIGNIFICANCE: Deep learning may assist clinicians in virtual treatment planning in orthodontics, prosthetics, implantology, and orthognathic surgery. The impact of using such models in clinical practice should be explored.

The role of muscular traction in the occurrence of skeletal relapse after advancement bilateral sagittal split osteotomy (BSSO): A systematic review.

The aim of this systematic review was (i) to determine the role of muscular traction in the occurrence of skeletal relapse after advancement BSSO and (ii) to investigate the effect of advancement BSSO on the perimandibular muscles. This systematic review reports in accordance with the recommendations proposed by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement. Electronic database searches were performed in the databases MEDLINE, Embase and Cochrane Library. Inclusion criteria were as follows: assessment of relapse after advancement BSSO; assessment of morphological and functional change of the muscles after advancement BSSO; and clinical studies on human subjects. Exclusion criteria were as follows: surgery other than advancement BSSO; studies in which muscle activity/traction was not investigated; and case reports with a sample of five cases or fewer, review articles, meta-analyses, letters, congress abstracts or commentaries. Of the initial 1006 unique articles, 11 studies were finally included. In four studies, an intervention involving the musculature was performed with subsequent assessment of skeletal relapse. The changes in the morphological and functional properties of the muscles after BSSO were studied in seven studies. The findings of this review demonstrate that the perimandibular musculature plays a role in skeletal relapse after advancement BSSO and may serve as a target for preventive strategies to reduce this complication. However, further research is necessary to (i) develop a better understanding of the role of each muscle group, (ii) to develop new therapeutic strategies and (iii) to define criteria that allow identification of patients at risk.

Is the pattern of mandibular asymmetry in mild craniofacial microsomia comparable to non-syndromic class II asymmetry?

OBJECTIVES: To compare the characteristics of mandibular asymmetry in patients with unilateral craniofacial microsomia (CFM) and class II asymmetry. MATERIALS AND METHODS: Pretreatment cone-beam computed tomography of consecutive adults with Pruzansky-Kaban type I and IIA CFM (CFM group) was analyzed by 3D cephalometry. Fourteen mandibular landmarks and two dental landmarks were identified. The mandibular size and positional asymmetry were calculated by using landmark-based linear and volumetric measurements, in terms of asymmetry ratios (affected/non-affected side) and absolute differences (affected - non-affected side). Results were compared with non-syndromic class II with matched severity of chin deviation (Class II group). Statistical analyses included independent t test, paired t test, chi-square test, and ANOVA. RESULTS: CFM group (n, 21; mean age, 20.4 ± 2.5 years) showed significantly larger size asymmetry in regions of mandibular body, ramus, and condyle compared to Class II group (n, 21; mean age, 27.8 ± 5.9 years) (p < 0.05). The curvature of mandibular body was asymmetric in CFM. Regarding the positional asymmetry of mandibular body, while a comparable transverse shift and a negligible yaw rotation were found among the two groups, the roll rotation in CFM was significantly greater as well as the occlusal (6.06° vs. 4.17°) and mandibular (7.84° vs. 2.80°) plane cants (p < 0.05). CONCLUSIONS: Mild CFM showed significantly more severe size asymmetry and roll rotation in mandible than non-CFM class II asymmetry. CLINICAL RELEVANCE: To improve the mandibular size and positional asymmetry in CFM, adjunct hard tissue augmentation or reduction in addition to OGS orthodontics with a meticulous roll and yaw planning is compulsory, which is expected to be distinct from treating non-CFM class II asymmetry.

A Survey on the Use of Artificial Intelligence by Clinicians in Dentistry and Oral and Maxillofacial Surgery.

Background: Applications of artificial intelligence (AI) in medicine and dentistry have been on the rise in recent years. In dental radiology, deep learning approaches have improved diagnostics, outperforming clinicians in accuracy and efficiency. This study aimed to provide information on clinicians' knowledge and perceptions regarding AI. Methods: A 21-item questionnaire was used to study the views of dentistry professionals on AI use in clinical practice. Results: In total, 302 questionnaires were answered and assessed. Most of the respondents rated their knowledge of AI as average (37.1%), below average (22.2%) or very poor (23.2%). The participants were largely convinced that AI would improve and bring about uniformity in diagnostics (mean Likert ± standard deviation 3.7 ± 1.27). Among the most serious concerns were the responsibility for machine errors (3.7 ± 1.3), data security or privacy issues (3.5 ± 1.24) and the divestment of healthcare to large technology companies (3.5 ± 1.28). Conclusions: Within the limitations of this study, insights into the acceptance and use of AI in dentistry are revealed for the first time.

Three-Dimensional Analysis of the Condylar Hypoplasia and Facial Asymmetry in Craniofacial Microsomia Using Cone-Beam Computed Tomography.

PURPOSE: To assess the condylar hypoplasia and its correlation with craniofacial deformities in adults with unilateral craniofacial microsomia (CFM). METHODS: Pretreatment cone-beam computed tomography scans of consecutive adults (mean age: 20.4 ± 3.0 years; range: 17.3 to 31.4 years) with Pruzansky-Kaban type I and IIA CFM were reconstructed in 3D. Both condyles were segmented. Asymmetry ratios (affected side/contralateral side) of condylar volume were calculated to indicate the extent of condylar hypoplasia. 3D cephalometry was performed to quantify the maxillomandibular morphology and facial asymmetry. The correlations in between were assessed by using Pearson's or Spearman's correlation coefficients. RESULTS: Thirty-six subjects were enrolled, consisting of 22 subjects with Pruzansky-Kaban type I and 14 subjects with type IIA. The condyles in type IIA group were significantly more hypoplastic in height (asymmetry ratio: 40.69 vs 59.95%, P = .006) and volume (18.16 vs 47.84%, P < .001) compared to type I group. Type IIA group had a significantly smaller SNB value than type I group (72.94° vs 77.41°, P = .012), and a significantly greater facial asymmetry (P < .05). The hypoplastic extent of condylar volume and Pruzansky-Kaban types were significantly correlated with SNB (r = 0.457 and ρ = -0.411, respectively), upper incisor deviation (r = -0.446 and ρ = 0.362), chin deviation (r = -0.477 and ρ = 0.527), upper occlusal plane cant (r = -0.672 and ρ = 0.631), and mandibular plane cant (r = -0.557 and ρ = 0.357, P < .05). CONCLUSION: For unilateral CFM adults, greater condylar hypoplasia in volume along with more severe mandibular retrusion and facial asymmetry objectively indicated a higher scale of Pruzansky-Kaban classification (type IIA). These quantitative distinctions are expected to enhance the diagnostic reliability of CFM.

Median Lingual Foramen, a new midmandibular cephalometric landmark.

PURPOSE: In asymmetrical mandibles, it is often challenging to identify the mandibular midline. The median lingual foramen (MLF) is located at the midline of the anterior mandible. The purpose of this study is to evaluate the reproducibility of identifying the MLF compared to conventional landmarks on cone beam computed tomography's (CBCT's) to mark the mandibular midline. MATERIAL AND METHODS: Ten symmetrical class II, 10 symmetrical class III, ten asymmetrical class II and 10 asymmetrical class III patients were included. On CBCTs, the cephalometric landmarks menton, pogonion, genial tubercle and MLF were identified twice by two observers. RESULTS: A high intra- and interobserver reproducibility was found for all landmarks, the highest being the MLF. The gain in accuracy is 0.998 mm, 0.824 mm and 0.361 mm compared to pogonion, genial tubercle and menton, respectively (P-value <.05). CONCLUSION: MLF is a reliable and reproducible landmark to indicate the midline of the mandible, particularly in Class II asymmetric mandibles.

Fully automated landmarking and facial segmentation on 3D photographs.

Three-dimensional facial stereophotogrammetry provides a detailed representation of craniofacial soft tissue without the use of ionizing radiation. While manual annotation of landmarks serves as the current gold standard for cephalometric analysis, it is a time-consuming process and is prone to human error. The aim in this study was to develop and evaluate an automated cephalometric annotation method using a deep learning-based approach. Ten landmarks were manually annotated on 2897 3D facial photographs. The automated landmarking workflow involved two successive DiffusionNet models. The dataset was randomly divided into a training and test dataset. The precision of the workflow was evaluated by calculating the Euclidean distances between the automated and manual landmarks and compared to the intra-observer and inter-observer variability of manual annotation and a semi-automated landmarking method. The workflow was successful in 98.6% of all test cases. The deep learning-based landmarking method achieved precise and consistent landmark annotation. The mean precision of 1.69 ± 1.15 mm was comparable to the inter-observer variability (1.31 ± 0.91 mm) of manual annotation. Automated landmark annotation on 3D photographs was achieved with the DiffusionNet-based approach. The proposed method allows quantitative analysis of large datasets and may be used in diagnosis, follow-up, and virtual surgical planning.

Complex Craniofacial Cases through Augmented Reality Guidance in Surgical Oncology: A Technical Report.

Augmented reality (AR) is a promising technology to enhance image guided surgery and represents the perfect bridge to combine precise virtual planning with computer-aided execution of surgical maneuvers in the operating room. In craniofacial surgical oncology, AR brings to the surgeon's sight a digital, three-dimensional representation of the anatomy and helps to identify tumor boundaries and optimal surgical paths. Intraoperatively, real-time AR guidance provides surgeons with accurate spatial information, ensuring accurate tumor resection and preservation of critical structures. In this paper, the authors review current evidence of AR applications in craniofacial surgery, focusing on real surgical applications, and compare existing literature with their experience during an AR and navigation guided craniofacial resection, to subsequently analyze which technological trajectories will represent the future of AR and define new perspectives of application for this revolutionizing technology.

The Perfect Timing-Immediate versus Delayed Microvascular Reconstruction of the Mandible.

In this retrospective study, the clinical and economic implications of microvascular reconstruction of the mandible were assessed, comparing immediate versus delayed surgical approaches. Utilizing data from two German university departments for oral and maxillofacial surgery, the study included patients who underwent mandibular reconstruction following continuity resection. The data assessed included demographic information, reconstruction details, medical history, dental rehabilitation status, and flap survival rates. In total, 177 cases (131 male and 46 females; mean age: 59 years) of bony free flap reconstruction (72 immediate and 105 delayed) were included. Most patients received adjuvant treatment (81% with radiotherapy and 51% combined radiochemotherapy), primarily for tumor resection. Flap survival was not significantly influenced by the timing of reconstruction, radiotherapy status, or the mean interval (14.5 months) between resection and reconstruction. However, immediate reconstruction had consumed significantly fewer resources. The rate of implant-supported masticatory rehabilitation was only 18% overall. This study suggests that immediate jaw reconstruction is economically advantageous without impacting flap survival rates. It emphasizes patient welfare as paramount over financial aspects in clinical decisions. Furthermore, this study highlights the need for improved pathways for masticatory rehabilitation, as evidenced by only 18% of patients with implant-supported dentures, to enhance quality of life and social integration.

Detection of caries around restorations on bitewings using deep learning.

OBJECTIVE: Secondary caries lesions adjacent to restorations, a leading cause of restoration failure, require accurate diagnostic methods to ensure an optimal treatment outcome. Traditional diagnostic strategies rely on visual inspection complemented by radiographs. Recent advancements in artificial intelligence (AI), particularly deep learning, provide potential improvements in caries detection. This study aimed to develop a convolutional neural network (CNN)-based algorithm for detecting primary caries and secondary caries around restorations using bitewings. METHODS: Clinical data from 7 general dental practices in the Netherlands, comprising 425 bitewings of 383 patients, were utilized. The study used the Mask-RCNN architecture, for instance, segmentation, supported by the Swin Transformer backbone. After data augmentation, model training was performed through a ten-fold cross-validation. The diagnostic accuracy of the algorithm was evaluated by calculating the area under the Free-Response Receiver Operating Characteristics curve, sensitivity, precision, and F1 scores. RESULTS: The model achieved areas under FROC curves of 0.806 and 0.804, and F1-scores of 0.689 and 0.719 for primary and secondary caries detection, respectively. CONCLUSION: An accurate CNN-based automated system was developed to detect primary and secondary caries lesions on bitewings, highlighting a significant advancement in automated caries diagnostics. CLINICAL SIGNIFICANCE: An accurate algorithm that integrates the detection of both primary and secondary caries will permit the development of automated systems to aid clinicians in their daily clinical practice.

Skeletal structure of asymmetric mandibular prognathism and retrognathism.

BACKGROUND: This study aimed to compare the skeletal structures between mandibular prognathism and retrognathism among patients with facial asymmetry. RESULTS: Patients who had mandibular asymmetry with retrognathism (Group A) in The Netherlands were compared with those with deviated mandibular prognathism (Group B) in Korea. All the data were obtained from 3D-reformatted cone-beam computed tomography images from each institute. The right and left condylar heads were located more posteriorly, inferiorly, and medially in Group B than in Group A. The deviated side of Group A and the contralateral side of Group B showed similar condylar width and height, ramus-proper height, and ramus height. Interestingly, there were no inter-group differences in the ramus-proper heights. Asymmetric mandibular body length was the most significantly correlated with chin asymmetry in retrognathic asymmetry patients whereas asymmetric elongation of condylar process was the most important factor for chin asymmetry in deviated mandibular prognathism. CONCLUSION: Considering the 3D positional difference of gonion and large individual variations of frontal ramal inclination, significant structural deformation in deviated mandibular prognathism need to be considered in asymmetric prognathism patients. Therefore, Individually planned surgical procedures that also correct the malpositioning of the mandibular ramus are recommended especially in patients with asymmetric prognathism.

Deep learning for automated segmentation of the temporomandibular joint.

OBJECTIVE: Quantitative analysis of the volume and shape of the temporomandibular joint (TMJ) using cone-beam computed tomography (CBCT) requires accurate segmentation of the mandibular condyles and the glenoid fossae. This study aimed to develop and validate an automated segmentation tool based on a deep learning algorithm for accurate 3D reconstruction of the TMJ. MATERIALS AND METHODS: A three-step deep-learning approach based on a 3D U-net was developed to segment the condyles and glenoid fossae on CBCT datasets. Three 3D U-Nets were utilized for region of interest (ROI) determination, bone segmentation, and TMJ classification. The AI-based algorithm was trained and validated on 154 manually segmented CBCT images. Two independent observers and the AI algorithm segmented the TMJs of a test set of 8 CBCTs. The time required for the segmentation and accuracy metrics (intersection of union, DICE, etc.) was calculated to quantify the degree of similarity between the manual segmentations (ground truth) and the performances of the AI models. RESULTS: The AI segmentation achieved an intersection over union (IoU) of 0.955 and 0.935 for the condyles and glenoid fossa, respectively. The IoU of the two independent observers for manual condyle segmentation were 0.895 and 0.928, respectively (p<0.05). The mean time required for the AI segmentation was 3.6 s (SD 0.9), whereas the two observers needed 378.9 s (SD 204.9) and 571.6 s (SD 257.4), respectively (p<0.001). CONCLUSION: The AI-based automated segmentation tool segmented the mandibular condyles and glenoid fossae with high accuracy, speed, and consistency. Potential limited robustness and generalizability are risks that cannot be ruled out, as the algorithms were trained on scans from orthognathic surgery patients derived from just one type of CBCT scanner. CLINICAL SIGNIFICANCE: The incorporation of the AI-based segmentation tool into diagnostic software could facilitate 3D qualitative and quantitative analysis of TMJs in a clinical setting, particularly for the diagnosis of TMJ disorders and longitudinal follow-up.

Detection of oral squamous cell carcinoma in clinical photographs using a vision transformer.

Oral squamous cell carcinoma (OSCC) is amongst the most common malignancies, with an estimated incidence of 377,000 and 177,000 deaths worldwide. The interval between the onset of symptoms and the start of adequate treatment is directly related to tumor stage and 5-year-survival rates of patients. Early detection is therefore crucial for efficient cancer therapy. This study aims to detect OSCC on clinical photographs (CP) automatically. 1406 CP(s) were manually annotated and labeled as a reference. A deep-learning approach based on Swin-Transformer was trained and validated on 1265 CP(s). Subsequently, the trained algorithm was applied to a test set consisting of 141 CP(s). The classification accuracy and the area-under-the-curve (AUC) were calculated. The proposed method achieved a classification accuracy of 0.986 and an AUC of 0.99 for classifying OSCC on clinical photographs. Deep learning-based assistance of clinicians may raise the rate of early detection of oral cancer and hence the survival rate and quality of life of patients.

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.

Optimizing efficiency in the creation of patient-specific plates through field-driven generative design in maxillofacial surgery.

Field driven design is a novel approach that allows to define through equations geometrical entities known as implicit bodies. This technology does not rely upon conventional geometry subunits, such as polygons or edges, rather it represents spatial shapes through mathematical functions within a geometrical field. The advantages in terms of computational speed and automation are conspicuous, and well acknowledged in engineering, especially for lattice structures. Moreover, field-driven design amplifies the possibilities for generative design, facilitating the creation of shapes generated by the software on the basis of user-defined constraints. Given such potential, this paper suggests the possibility to use the software nTopology, which is currently the only software for field-driven generative design, in the context of patient-specific implant creation for maxillofacial surgery. Clinical scenarios of applicability, including trauma and orthognathic surgery, are discussed, as well as the integration of this new technology with current workflows of virtual surgical planning. This paper represents the first application of field-driven design in maxillofacial surgery and, although its results are very preliminary as it is limited in considering only the distance field elaborated from specific points of reconstructed anatomy, it introduces the importance of this new technology for the future of personalized implant design in surgery.

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.

Positional assessment of lower third molar and mandibular canal using explainable artificial intelligence.

OBJECTIVE: The aim of this study is to automatically assess the positional relationship between lower third molars (M3i) and the mandibular canal (MC) based on the panoramic radiograph(s) (PR(s)). MATERIAL AND METHODS: A total of 1444 M3s were manually annotated and labeled on 863 PRs as a reference. A deep-learning approach, based on MobileNet-V2 combination with a skeletonization algorithm and a signed distance method, was trained and validated on 733 PRs with 1227 M3s to classify the positional relationship between M3i and MC into three categories. Subsequently, the trained algorithm was applied to a test set consisting of 130 PRs (217 M3s). Accuracy, precision, sensitivity, specificity, negative predictive value, and F1-score were calculated. RESULTS: The proposed method achieved a weighted accuracy of 0.951, precision of 0.943, sensitivity of 0.941, specificity of 0.800, negative predictive value of 0.865 and an F1-score of 0.938. CONCLUSION: AI-enhanced assessment of PRs can objectively, accurately, and reproducibly determine the positional relationship between M3i and MC. CLINICAL SIGNIFICANCE: The use of such an explainable AI system can assist clinicians in the intuitive positional assessment of lower third molars and mandibular canals. Further research is required to automatically assess the risk of alveolar nerve injury on panoramic radiographs.

Patients' perspectives on the use of artificial intelligence in dentistry: a regional survey.

The use of artificial intelligence (AI) in dentistry is rapidly evolving and could play a major role in a variety of dental fields. This study assessed patients' perceptions and expectations regarding AI use in dentistry. An 18-item questionnaire survey focused on demographics, expectancy, accountability, trust, interaction, advantages and disadvantages was responded to by 330 patients; 265 completed questionnaires were included in this study. Frequencies and differences between age groups were analysed using a two-sided chi-squared or Fisher's exact tests with Monte Carlo approximation. Patients' perceived top three disadvantages of AI use in dentistry were (1) the impact on workforce needs (37.7%), (2) new challenges on doctor-patient relationships (36.2%) and (3) increased dental care costs (31.7%). Major expected advantages were improved diagnostic confidence (60.8%), time reduction (48.3%) and more personalised and evidencebased disease management (43.0%). Most patients expected AI to be part of the dental workflow in 1-5 (42.3%) or 5-10 (46.8%) years. Older patients (> 35 years) expected higher AI performance standards than younger patients (18-35 years) (p < 0.05). Overall, patients showed a positive attitude towards AI in dentistry. Understanding patients' perceptions may allow professionals to shape AI-driven dentistry in the future.