Research Appraisals
Evidence-based critical appraisals of the latest medical research, systematically evaluated using Oxford CEBM methodology.
Showing 9 appraisals
Acta odontologica Scandinavica
Predicting gingival embrasure risk after invisible orthodontics using multimodal data and machine learning
OBJECTIVE: To develop and validate a risk prediction model for gingival embrasures after clear aligner therapy using multimodal oral data. METHODS: A retrospective study of 340 patients (December 2022-June 2025) was randomly divided into training (n = 238) and validation (n = 102) sets (7:3). Univariate analysis, multivariate logistic regression, and least absolute shrinkage and selection operator regression were applied to identify independent risk factors. Three machine learning models - random forest (RF), logistic regression, and support vector machine - were constructed based on seven core variables. Model performance was assessed using area under the receiver operating characteristic curve (AUC), calibration curves, decision curve analysis, and Shapley Additive Explanations (SHAP) values for interpretability. RESULTS: No significant baseline differences existed between sets (p > 0.05). Seven indicators were identified (p < 0.05). Multivariate analysis confirmed percentage of bleeding on probing-positive sites, interproximal alveolar bone height, and relative movement of adjacent teeth at target site as independent risk factors, while gingival thickness, proximal contact area, interdental papilla height, and buccal/lingual bone plate thickness were protective factors (p < 0.05). The RF model performed best: training AUC = 0.849 (95% CI: 0.786-0.912), validation AUC = 0.815 (95% CI: 0.720-0.910), with good calibration and net benefit. SHAP analysis highlighted gingival thickness and interproximal alveolar bone height as key predictors. CONCLUSION: A risk prediction model for post-clear aligner gingival embrasures was successfully developed and validated using multimodal oral data, with RF as the optimal algorithm. The model exhibits good discrimination, calibration, and clinical utility, which can be used as an objective auxiliary tool for individualized risk prediction following clear aligner therapy and supplement traditional clinical empirical judgment.
23 July 2026
Read appraisal →Clinical oral investigations
Deep learning-based automated assessment of alveolar bone loss in CBCT for periodontitis
OBJECTIVES: This study aimed to develop and validate a CBCT-based automated system for assessing radiographic alveolar bone loss (RBL) to improve the accuracy and efficiency of periodontitis diagnosis. METHODS: A total of 110 patients (2,796 teeth) with Stage I-IV periodontitis from four center were included. The nnU-Net framework was used to segment teeth, alveolar bone, and the cemento-enamel junction (CEJ). RBL was calculated automatically using an edge-constrained shortest path algorithm. The model was trained on data from Center A and externally validated with datasets from Centers B-D. Linear periodontal measurements from 11 CBCT scans were compared between manual and CAD-based segmentation. An independent validation set was used to assess automated staging accuracy and time efficiency. RESULTS: The CAD system achieved Dice similarity coefficients (DSC) of 95.85% for teeth, 95.75% for alveolar bone, and 86.18% for the CEJ. External validation showed alveolar bone and tooth DSC values both above 95% and CEJ DSC values above 77%. Linear measurements showed strong agreement with manual segmentation (Spearman's ρ = 0.9187; ICC = 0.9266). For staging, the CAD system reached an overall accuracy of 87.31%, with 97.01% for Stage Ⅰ, 88.06% for Stage Ⅱ and 89.55% for Stage Ⅲ/Ⅳ. The CAD system represented a 13.42-fold acceleration compared with the manual workflow. CONCLUSION: The CAD system enables accurate automated segmentation and RBL quantification on CBCT images, with robust multicenter performance and substantial gains in efficiency. CLINICAL SIGNIFICANCE: This system offers a fast and reliable method for RBL assessment, supporting consistent diagnosis and monitoring of periodontitis.
19 July 2026
Read appraisal →Dento maxillo facial radiology
JADE: jawbone lesion diagnosis and decision supporting system
OBJECTIVES: To develop and evaluate JADE, a proof-of-concept retrieval-augmented generation (RAG) diagnostic assistive system, designed to enhance large language model (LLM) reasoning for jawbone lesion assessment. This study examined whether RAG improves diagnostic accuracy and stability compared with standalone LLMs and ORAD, a supervised learning-based system. METHODS: JADE was developed as a cloud-based RAG system integrating an expert-curated oral radiology database embedded using text-embedding-3-large and indexed in Qdrant Cloud. Structured clinical inputs were encoded as prioritized vector queries. Hybrid semantic and keyword-based retrieval appended relevant evidence to prompts for differential diagnosis generation across LLM backbones. Performance was evaluated in 25 validation cases and compared with GPT-5, Claude Sonnet 4.5, DeepSeek-R1, Gemini 2.5 Flash, their RAG configurations, and ORAD. Diagnostic accuracy was analyzed using Cochran's Q test, with post-hoc McNemar's tests and Bonferroni correction. Intra-model stability and response time were assessed. RESULTS: RAG-GPT-5 achieved the highest diagnostic accuracy (20/25), followed by RAG-Claude Sonnet 4.5 (18/25), RAG-DeepSeek R1 (17/25), and RAG-Gemini 2.5 Flash (15/25). Standalone models achieved 9-13/25 correct diagnoses, ORAD achieved 17/25. No significant differences were observed among standalone models or RAG-based models and ORAD. A significant improvement was observed for GPT-5 when integrated with RAG (P = .002). RAG configurations showed higher intra-model stability, with RAG-GPT-5 achieving mean stability of 0.90 ± 0.11. Mean response times ranged from 3 to 10 s. CONCLUSIONS: JADE improved diagnostic accuracy and stability compared with standalone LLMs, underscoring value of RAG reasoning in jawbone lesion assessment and marking the first RAG application in dentomaxillofacial radiology.
3 July 2026
Read appraisal →The Journal of prosthetic dentistry
Deep learning-based object detection of dental implant systems in panoramic and periapical radiographs
STATEMENT OF PROBLEM: The manual identification of dental implant systems on radiographs is time-consuming, operator-dependent, and prone to diagnostic inaccuracies, particularly for patients where clinical documentation is lacking. The increasing variety of implant designs further complicates identification in prosthetic and surgical practice. PURPOSE: The purpose of this study was to develop and evaluate a deep learning-based model for the automated identification of 7 implant systems (Adin, Dentium, Dionavi, Make It Simple (MIS), Nobel, Noris, and Osstem) using panoramic radiographs and periapical radiographs in an effort to enhance diagnostic efficiency and support clinical decision-making in prosthodontic care. MATERIAL AND METHODS: A total of 4677 anonymized radiographic images with 8189 implants were curated and annotated using Roboflow with bounding boxes outlining fixture components. The preprocessing involved normalization, resizing to 640×640 pixels, and geometric augmentation (rotation, cropping, and blurring) to handle class imbalances. You Only Look Once (YOLO) v10 architecture, implemented with PyTorch, using CSPDarknet and PANet for multiscale feature fusion, was used to optimize real-time detection. Transfer learning used pretrained weights, with training for over 500 epochs (batch size: 32) on NVIDIA T4 GPUs. Data partitioning involved an 80:10:10 ratio (training: validation: testing), with performance evaluated using precision, recall, F1-score, and mean average precision (mAP). RESULTS: The model achieved a mAP of 98.3%, with mean precision, recall, and F1-score values of 93%, 86%, and 89%, respectively. Osstem implants demonstrated maximum discriminability (99% precision, 95% recall). In contrast, Nobel implants exhibited low recall (72.7%), attributed to the sparsity of the dataset (564 samples for Nobel compared with 2320 for Osstem) and similar radiopacity patterns. CONCLUSIONS: The YOLOv10 model demonstrated good performance in identifying dental implants, showing clinical promise for minimizing prosthetic mismatches. Subject to ethics and regulatory approvals, additional improvements involving 3-dimensional imaging and heterogeneous datasets may add precision and validate artificial intelligence as an evidence-based advance in implant dentistry.
2 July 2026
Read appraisal →Progress in orthodontics
Evaluation of artificial intelligence-based cephalometric analysis compared to the manual method.
BACKGROUND: The application of artificial intelligence (AI) to cephalometry has aroused growing interest due to its potential to optimize clinical workflows and improve efficiency. This study aimed to compare the differences and reproducibility of cephalometric measurements obtained using three methods: manual, WebCeph™ and AngelAligner™. METHODOLOGY: 30 patients were analysed, evaluating 9 measurements generated by the three methods at two different timepoints (T1 and T2). RESULTS: WebCeph™ presented significant differences as compared to the manual method in 7 of the 9 measurements (SNA, SNB, ANB, U1-APg, L1-APg, GoGn-SN, U1-L1), while AngelAligner™ displayed significant differences with the manual method in 4 measurements (SNA, U1-APg, L1-APg and U1-L1). Significant differences were found between the two AI-based systems in 7 of the 9 measurements. Reproducibility was high at all angles with the manual method, except for U1-L1, NLA, and LFH. The ICC was 1 for all measurements for both AI-based systems, indicating optimal reproducibility. When grouping by malocclusions, however, WebCeph™ presented greater differences with the manual method than AngelAligner™. CONCLUSION: AI-based methods offered greater reproducibility than the manual method, but still require human supervision.
31 May 2026
Read appraisal →Dental materials journal
Domain shift analysis of deep learning models for tooth detection in pediatric panoramic radiographs
The aim of this study was to externally validate object detection models for comprehensive tooth detection on pediatric panoramic radiographs, quantify the impact of domain shift across institutions and imaging protocols, and compare YOLOv8 and YOLOv10. Two datasets were used: an internal set of 200 images of early mixed dentition without bite blocks, and an external, open-source set of 192 images acquired with bite blocks. Performance was assessed using mean average precision (mAP), per‑class AP, precision‑recall curves, and confusion matrix. Overall performance was high but domain dependent. mAP was 0.958/0.941 (YOLOv8/v10) in Experiment 1; 0.910/0.906 in Experiment 2; and 0.906/0.901 in Experiment 3. Error analysis revealed domain‑specific failure modes: bite blocks were occasionally mistaken for the primary central incisor. Deep learning enables accurate and comprehensive detection of teeth on pediatric panoramic radiographs. However, domain shift alters error patterns and impairs the detection of rare yet clinically important classes.
25 May 2026
Read appraisal →Dental press journal of orthodontics
Accuracy of cephalometric landmark identification on artificial intelligence-based software: a comparative study
INTRODUCTION: Cephalometry has long been a cornerstone of orthodontic diagnosis and treatment planning. While traditionally performed manually, cephalometric analysis has increasingly shifted toward semi-automated or fully automated computer-based software, due to advances in technology. OBJECTIVE: This study aims to evaluate and compare the accuracy of anatomical landmarks identified manually by orthodontists using digital tools versus those identified automatically by various artificial intelligence (AI)-based cephalometric software. METHODS: A total of 25 lateral cephalometric radiographs were selected for analysis. Manual landmark identification was independently performed by two orthodontists using the digital cephalometric software NemoCeph. These were compared to the automatic landmark identifications generated by two AI-based programs: CephNinja and WebCeph™. A total of 17 anatomical landmarks, 15 hard tissue and 2 soft tissue points, were evaluated. The 17 cephalometric landmarks selected in this study are among the most commonly used and reliably identifiable reference points in the literature. To facilitate direct visual comparison, landmark positions derived from all three methods were overlaid using semi-transparent image layers. One-way ANOVA and intraclass correlation coefficient (ICC) tests were conducted to assess inter-method differences. The threshold for statistical significance was set at p < 0.05 and p < 0.01. RESULTS: Statistically significant differences were observed at the porion, basion, gonion, and the apexes of the upper and lower incisors (p < 0.05). Compared to the manual tracings in NemoCeph, AI-based software generally placed these landmarks in more distal and superior positions. Among the AI tools, CephNinja showed greater similarity to NemoCeph for the gonion and basion points, while WebCeph™ was closer at the incisor apexes. CONCLUSION: Although AI-based cephalometric software offers rapid and reproducible assessments, certain anatomical landmarks - particularly porion, basion, gonion, and the apexes of the upper and lower incisors - demonstrate statistically significant positional discrepancies, compared to manual tracings. Clinicians should be cautious when relying exclusively on automated analyses for diagnostic or treatment planning purposes, since AI struggles with certain landmarks, due to algorithmic limitations.
21 May 2026
Read appraisal →Medical image analysis
MICCAI STS 2024 challenge: Semi-supervised instance-level tooth segmentation in panoramic X-ray and CBCT images
Orthopantomogram (OPGs) and Cone-Beam Computed Tomography (CBCT) are vital for dentistry, but creating large datasets for automated tooth segmentation is hindered by the labor-intensive process of manual instance-level annotation. This research aimed to benchmark and advance semi-supervised learning (SSL) as a solution for this data scarcity problem. We organized the 2nd Semi-supervised Teeth Segmentation (STS 2024) Challenge at MICCAI 2024. We provided a large-scale dataset comprising over 90,000 2D images and 3D axial slices, which includes 2380 OPG images and 330 CBCT scans, all featuring detailed instance-level FDI annotations on part of the data. The challenge attracted 114 (OPG) and 106 (CBCT) registered teams. To ensure algorithmic excellence and full transparency, we rigorously evaluated the valid, open-source submissions from the top 10 (OPG) and top 5 (CBCT) teams, respectively. All successful submissions were deep learning-based SSL methods. The winning semi-supervised models demonstrated impressive performance gains over a fully-supervised nnU-Net baseline trained only on the labeled data. For the 2D OPG track, the top method improved the Instance Affinity (IA) score by over 44 percentage points. For the 3D CBCT track, the winning approach boosted the Instance Dice score by 61 percentage points. This challenge demonstrates the potential benefit benefit of SSL for complex, instance-level medical image segmentation tasks where labeled data is scarce. The most effective approaches consistently leveraged hybrid semi-supervised frameworks that combined knowledge from foundational models like SAM with multi-stage, coarse-to-fine refinement pipelines. Both the challenge dataset and the participants' submitted code have been made publicly available on GitHub (https://github.com/ricoleehduu/STS-Challenge-2024), ensuring transparency and reproducibility.
9 May 2026
Read appraisal →American journal of orthodontics and dentofacial orthopedics : official publication of the American Association of Orthodontists, its constituent societies, and the American Board of Orthodontics
Prediction of skeletal maturity using machine learning based on multiple biological indicators
INTRODUCTION: Accurate assessment of skeletal maturity is essential for orthodontic treatment planning and growth modulation. This study aimed to develop a comprehensive machine learning (ML) model to predict skeletal maturation stages using cervical vertebral morphology, dental maturation stage (DMS), gender, and age. METHODS: A total of 860 patients with lateral cephalograms, panoramic, and hand-wrist radiographs were included. A baseline model using 6 cervical morphologic parameters was compared with enhanced models incorporating DMS, gender, and age. Six ML algorithms were evaluated, with CatBoost showing the highest performance. Feature importance was analyzed using 2 methods, and the final model was assessed using a confusion matrix, receiver operating characteristic curves, and calibration curves. RESULTS: Incorporating DMS, gender, and age significantly improved predictive performance. The CatBoost model achieved an area under the receiver operating characteristic curve of 0.924, an area under the precision-recall curve of 0.806, and an F1-score of 0.752. Feature importance analysis confirmed the strong predictive contribution of DMS, especially the mandibular second molar. The final model demonstrated high accuracy, strong discrimination, and good calibration across all skeletal maturity stages. CONCLUSIONS: This study presents a novel and clinically practical ML model for skeletal maturity prediction, integrating routinely acquired orthodontic records. The results demonstrate that combining DMS with cervical morphology, age, and gender enhances prediction accuracy, offering a reliable alternative to traditional hand-wrist assessments.
5 May 2026
Read appraisal →