RESUMEN
INTRODUCTION: Cone-beam computed tomography (CBCT) is widely used to detect jaw lesions, although CBCT interpretation is time-consuming and challenging. Artificial intelligence (AI) for CBCT segmentation may improve lesion detection accuracy. However, consistent automated lesion detection remains difficult, especially with limited training data. This study aimed to assess the applicability of pre-trained transformer-based architectures for semantic segmentation of CBCT volumes when applied to periapical lesion detection. METHODS: CBCT volumes (n=138) were collected and annotated by expert clinicians using five labels - "lesion", "restorative material", "bone", "tooth structure", and "background". U-Net (convolutional neural network (CNN)-based) and Swin-UNETR (transformer-based) models, pre-trained (Swin-UNETR-PRETRAIN) and from scratch (Swin-UNETR-SCRATCH), were trained with subsets of the annotated CBCTs. These models were then evaluated for semantic segmentation performance using the Sørensen-Dice coefficient (DICE), lesion detection performance using sensitivity and specificity, and training sample size requirements by comparing models trained with 20, 40, 60, or 103 samples. RESULTS: Trained with 103 samples, Swin-UNETR-PRETRAIN achieved a DICE of 0.8512 for "lesion", 0.8282 for "restorative materials", 0.9178 for "bone", 0.9029 for "tooth structure", and 0.9901 for "background". "Lesion" DICE was statistically similar between Swin-UNETR-PRETRAIN trained with 103 and 60 images (P>.05), with the latter achieving 1.00 sensitivity and 0.94 specificity in lesion detection. With small training sets, Swin-UNETR-PRETRAIN outperformed Swin-UNETR-SCRATCH in DICE over all labels (P<.001 [n=20], P<.001 [n=40]), and U-Net in lesion detection specificity (P=.006 [n=20], P=.031 [n=40]). CONCLUSIONS: Transformer-based Swin-UNETR architectures allowed for excellent semantic segmentation and periapical lesion detection. Pre-trained, it may provide an alternative with smaller training datasets compared to classic U-Net architectures.
RESUMEN
AIM: Post-operative pain may occur after non-surgical root canal treatment (NSRCT). The root filling technique and choice of sealer may be contributing factors. This randomized controlled clinical trial compared post-operative pain and analgesics intake after NSRCT using a sealer-based obturation technique (SBO) with single-cone gutta-percha and calcium silicate sealer (CSS) versus a warm-vertical compaction (WVC) technique with gutta-percha and a resin-based sealer (RBS). METHODOLOGY: This study was designed as a parallel-two arm, double-blind, randomized superiority clinical trial registered at www. CLINICALTRIALS: gov (NCT04753138). Patients referred for NSRCT fulfilling the inclusion criteria participated in this study. Pre-operative periapical radiographs and CBCT scans were taken and numerical rating scale (NRS) pain scores were recorded. Upon completion of canal instrumentation, participants were randomly allocated to either Group SBO: SBO with CSS or Group WVC: Warm-vertical compaction with RBS. Post-operative pain and analgesics intake were recorded at 1, 3 and 7 days after endodontic treatment completion. Non-parametric Mann-Whitney U and Friedman tests and a generalized estimating equation were used to assess differences in pain scores between the groups, within each treatment group at different time points and for correlations, respectively. RESULTS: The study included 195 participants (212 teeth). One participant declined to submit the NRS form. Therefore, 194 participants (211 teeth) were included in the final analysis (99.5% response rate). No statistically significant differences in post-operative pain or analgesic intake existed between the two groups at any time point (p > .05). Age, pre-operative pain, apical diagnosis and post-operative analgesic intake were significantly related to post-operative pain (p < .05). CONCLUSIONS: Sealer-based obturation technique with CSS was associated with similar post-operative pain levels and analgesics intake as WVC with RBS. Regarding post-operative pain, SBO with CSSs may be a suitable clinical alternative.
Asunto(s)
Gutapercha , Dolor Postoperatorio , Materiales de Obturación del Conducto Radicular , Obturación del Conducto Radicular , Tratamiento del Conducto Radicular , Humanos , Dolor Postoperatorio/tratamiento farmacológico , Dolor Postoperatorio/prevención & control , Obturación del Conducto Radicular/métodos , Materiales de Obturación del Conducto Radicular/uso terapéutico , Método Doble Ciego , Femenino , Masculino , Adulto , Gutapercha/uso terapéutico , Persona de Mediana Edad , Tratamiento del Conducto Radicular/métodos , Compuestos de Calcio/uso terapéutico , Silicatos/uso terapéutico , Dimensión del Dolor , Analgésicos/uso terapéuticoRESUMEN
BACKGROUND/AIM: The knowledge of dental students about managing traumatic dental injuries (TDIs) may not be uniform, depending on global location and dental education. The aim of this study was to evaluate the level of knowledge of undergraduate and postgraduate students specializing in endodontics and pediatric dentistry at 10 dental schools in 10 countries about the 2020 International Association of Dental Traumatology (IADT) guidelines regarding the management of TDIs. MATERIALS & METHODS: A previously published questionnaire was used in the current survey. It was an online survey with 12 questions regarding the management of TDIs and some additional questions regarding sociodemographic and professional profiles of the participants were added. The survey was distributed to final-year undergraduate students and postgraduate students in pediatric dentistry and endodontics from 10 dental schools. Simple frequency distributions and descriptive statistics were predominantly used to describe the data. Differences in the median percentage scores among the student categories were assessed using the Kruskal-Wallis test followed by Dwass-Steel-Critchlow-Fligner pairwise comparisons. RESULTS: A total of 347 undergraduates, 126 postgraduates in endodontics, and 72 postgraduates in pediatric dentistry from 10 dental schools participated in this survey. The postgraduates had a significantly higher percentage score for correct responses compared with the undergraduates. No significant difference was observed between the endodontic and pediatric dentistry postgraduates. CONCLUSION: The knowledge possessed by undergraduate and postgraduate students concerning the IADT-recommended management of TDIs varied across the globe and some aspects were found to be deficient. This study emphasizes the critical importance of reassessing the teaching and learning activities pertaining to the management of TDIs.
RESUMEN
Studies investigating the accuracy of diagnostic tests should provide data on how effectively they identify or exclude disease in order to inform clinicians responsible for managing patients. This consensus-based project was undertaken to develop reporting guidelines for authors submitting manuscripts, which describe studies that have evaluated the accuracy of diagnostic tests in endodontics. These guidelines are known as the Preferred Reporting Items for Diagnostic Accuracy Studies in Endodontics (PRIDASE) 2024 guidelines. A nine-member steering committee created an initial checklist by integrating and modifying items from the Standards for Reporting of Diagnostic Accuracy (STARD) 2015 checklist and the Clinical and Laboratory Images in Publications (CLIP) principles, as well as adding a number of new items specific to the specialty of endodontics. Thereafter, the steering committee formed the PRIDASE Delphi Group (PDG) and the PRIDASE Online Meeting Group (POMG) in order to collect expert feedback on the preliminary draft checklist. Members of the Delphi group engaged in an online Delphi process to reach consensus on the clarity and suitability of the items in the checklist. The online meeting group then held an in-depth discussion on the online Delphi-generated items via the Zoom platform on 20 October 2023. According to the feedback obtained, the steering committee revised the PRIDASE checklist, which was then piloted by several authors when preparing manuscripts describing diagnostic accuracy studies in endodontics. Feedback from this process resulted in the final version of the PRIDASE 2024 checklist, which has 11 sections and 66 items. Authors are encouraged to use the PRIDASE 2024 guidelines when developing manuscripts on diagnostic accuracy in endodontics in order to improve the quality of reporting in this area. Editors of relevant journals will be invited to include these guidelines in their instructions to authors.
Asunto(s)
Lista de Verificación , Consenso , Técnica Delphi , Endodoncia , Humanos , Endodoncia/normas , Pruebas Diagnósticas de Rutina/normasRESUMEN
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.
Asunto(s)
Inteligencia Artificial , Curriculum , Educación en Odontología , Endodoncia , Endodoncia/educación , Humanos , Educación en Odontología/métodos , Competencia ClínicaRESUMEN
INTRODUCTION: Training of Artificial Intelligence (AI) for biomedical image analysis depends on large annotated datasets. This study assessed the efficacy of Active Learning (AL) strategies training AI models for accurate multilabel segmentation and detection of periapical lesions in cone-beam CTs (CBCTs) using a limited dataset. METHODS: Limited field-of-view CBCT volumes (n = 20) were segmented by clinicians (clinician segmentation [CS]) and Bayesian U-Net-based AL strategies. Two AL functions, Bayesian Active Learning by Disagreement [BALD] and Max_Entropy [ME], were used for multilabel segmentation ("Lesion"-"Tooth Structure"-"Bone"-"Restorative Materials"-"Background"), and compared to a non-AL benchmark Bayesian U-Net function. The training-to-testing set ratio was 4:1. Comparisons between the AL and Bayesian U-Net functions versus CS were made by evaluating the segmentation accuracy with the Dice indices and lesion detection accuracy. The Kruskal-Wallis test was used to assess statistically significant differences. RESULTS: The final training set contained 26 images. After 8 AL iterations, lesion detection sensitivity was 84.0% for BALD, 76.0% for ME, and 32.0% for Bayesian U-Net, which was significantly different (P < .0001; H = 16.989). The mean Dice index for all labels was 0.680 ± 0.155 for Bayesian U-Net and 0.703 ± 0.166 for ME after eight AL iterations, compared to 0.601 ± 0.267 for Bayesian U-Net over the mean of all iterations. The Dice index for "Lesion" was 0.504 for BALD and 0.501 for ME after 8 AL iterations, and at a maximum 0.288 for Bayesian U-Net. CONCLUSIONS: Both AL strategies based on uncertainty quantification from Bayesian U-Net BALD, and ME, provided improved segmentation and lesion detection accuracy for CBCTs. AL may contribute to reducing extensive labeling needs for training AI algorithms for biomedical image analysis in dentistry.
Asunto(s)
Algoritmos , Inteligencia Artificial , Teorema de Bayes , Incertidumbre , Tomografía Computarizada de Haz Cónico , Materiales Dentales , Procesamiento de Imagen Asistido por ComputadorRESUMEN
OBJECTIVE: The aim was to compare the 'reverse sandwich restoration' to resin composite restorations re- garding marginal adaptation, fracture resistance, favourable/unfavourable fractures in the management of external cervical resorption. METHODS: Forty-eight extracted maxillary central incisors were selected and endodontically treated. Cervical regions of the labial root surfaces received simulated resorptive defects and were restored as three randomly allocated groups: Reverse Sandwich Restoration (resin composite + resin-modified glass ionomer) (RSR); resin composite restoration (COMP), and no restoration (NR). Each group was further divided into two subgroups (n=8 each): Thermomechanical Aging (TA) (equivalent to one year) and No Aging (NA). Marginal adaptation was scored by scanning electron microscopy. Fracture resistance was tested using a universal testing machine. Favourable versus unfavourable fractures were classified based on fracture extent. RESULTS: TA decreased the marginal adaptation for both RSR and COMP. Mean fracture resistance per groups were: RSR-NA 1522.4+-94.9N, RSR-TA 939.6+-72.9N, COMP-NA 1197.6+-95.7N, COMP-TA 870.4+-86.3N, NR-NA 1057.1+-88.1N, and NR-TA 836.6+-81.9N, respectively. Fracture resistance was the highest for RSR- NA compared to all other groups (p<0.05). TA decreased the fracture resistance in all groups (p<0.05), there was no significant difference between RSR and COMP regarding fracture resistance and favourable/ unfavourable fractures (p>0.05). CONCLUSION: RSR provided comparable results to resin composite fillings to restore artificial cervical defects pertaining to marginal adaptation, fracture resistance, and favourable versus unfavourable fractures. RSR is preferable due to its inherent biocompatibility to the periodontium. (EEJ-2023-04-050).
Asunto(s)
Resinas Compuestas , Restauración Dental Permanente , Restauración Dental Permanente/métodos , Resinas Compuestas/uso terapéuticoRESUMEN
The objectives of this study were to evaluate the stress distribution and risk of fracture of a non-vital immature maxillary central incisor subjected to various clinical procedures using finite element analysis (FEA). A three-dimensional model of an immature central incisor was developed, from which six main models were designed: untreated immature tooth (C), standard apical plug (AP), resin composite (RC), glass-fibre post (GFP), regeneration procedure (RET), and regeneration with induced root maturation (RRM). Mineral trioxide aggregate (MTA) or Biodentine® were used as an apical or coronal plug. All models simulated masticatory forces in a quasi-static approach with an oblique force of 240 Newton at a 120° to the longitudinal tooth axis. The maximum principal stress, maximum shear stress, risk of fracture, and the strengthening percentage were evaluated. The mean maximum principal stress values were highest in model C [90.3 MPa (SD = 4.4)] and lowest in the GFP models treated with either MTA and Biodentine®; 64.1 (SD = 1.7) and 64.0 (SD = 1.6) MPa, respectively. Regarding the shear stress values, the dentine tooth structure in model C [14.4 MPa (SD = 0.8)] and GFP models [15.4 MPa (SD = 1.1)] reported significantly higher maximum shear stress values compared to other tested models (p < 0.001), while no significant differences were reported between the other models (p > 0.05). No significant differences between MTA and Biodentine® regarding maximum principal stress and maximum shear stress values for each tested model (p > 0.05). A maximum strain value of 4.07E-03 and maximum displacement magnitude of 0.128 mm was recorded in model C. In terms of strengthening percentage, the GFP models were associated with the highest increase (22%). The use of a GFP improved the biomechanical performance and resulted in a lower risk of fracture of a non-vital immature maxillary central incisor in a FEA model.