Diagnosis of in vivo vertical root fracture using deep learning on cone-beam CT images

Automatic Vertical Root Fracture Detection on Intraoral Periapical Radiographs With Artificial Intelligence-Based Image Enhancement

BACKGROUND/AIM

To explore transfer learning (TL) techniques for enhancing vertical root fracture (VRF) diagnosis accuracy and to assess the impact of artificial intelligence (AI) on image enhancement for VRF detection on both extracted teeth images and intraoral images taken from patients.

MATERIALS AND METHODS

A dataset of 378 intraoral periapical radiographs comprising 195 teeth with fractures and 183 teeth without fractures serving as controls was included. DenseNet, ConvNext, Inception121, and MobileNetV2 were employed with model fusion. Prior to evaluation, Particle Swarm Optimization (PSO) and Deep Learning (DL) image enhancement were applied. Performance assessment included accuracy rate, precision, recall, F1-score, AUC, and kappa values. Intra- and inter-observer agreement, according to the Gold Standard (GS), were assessed using ICC and t-tests. Statistical significance was set at p < 0.05.

RESULTS

The DenseNet + Inception fusion model achieved the highest accuracy rate of 0.80, with commendable recall, F1-score, and AUC values, supported by precision (0.81) and kappa (0.60) values. Molar tooth examination yielded an accuracy rate, precision, recall, and F1-score of 0.80, with an AUC of 0.84 and kappa of 0.60. For premolar teeth, the fusion network showed an accuracy rate of 0.78, an AUC of 0.78, and notable metrics, including F1-score (0.80), recall (0.85), precision (0.71), and kappa (0.55). ICC results demonstrated acceptable agreement (≥ 0.57 for molars, ≥ 0.52 for premolars).

CONCLUSION

TL methods have demonstrated significant potential in enhancing diagnostic accuracy for VRFs in radiographic imaging. TL is emerging as a valuable tool in the development of robust, automated diagnostic systems for VRF identification, ultimately supporting clinicians in delivering more accurate diagnoses.

Diagnosis of In Vivo Vertical Root Fracture in Endodontically Treated Teeth Using Machine Learning Techniques

INTRODUCTION

This study aimed to diagnose vertical root fracture (VRF) of endodontically treated teeth using clinical features and bone loss information from cone beam computed tomography with machine learning models.

METHODS

A total of 887 patients with 941 teeth undergoing endodontic surgery were included in this retrospective study. The clinical factors and bone defects detected via cone beam computed tomography were measured and recorded. Linear machine learning models, logistic regression model and nonlinear models, including XGBoost, LightGBM, and CatBoost were used to diagnose VRF. Model performance was evaluated using 5-fold cross-validation and based on various performance parameters, including the area under the curve, sensitivity, specificity, precision, and F score. Model interpretations were visualized by Shapley Additive Explanations.

RESULTS

Of the 941 teeth, 112 VRF teeth (11.9%) were identified during endodontic surgery or after tooth extraction. XGBoost and LightGBM showed excellent performance with area under the curves of 0.98 [0.96, 0.99], specificity of 0.978 and 0.983, sensitivity of 0.883 and 0.803, and precision of 0.846 and 0.865, respectively. Shapley Additive Explanations values showed that lingual/buccal bone defect, the ratio of bone defect height above the root apex to the defect total height, width of bone defect and age were the top 5 contributors.

CONCLUSIONS

Machine learning models for the diagnosis of VRF using age, sex, tooth type, the quality of root canal filling and bone loss position, height, width, and depth are valuable for clinical decision making after root canal treatment.

Artificial intelligence applications in smile design dentistry: A scoping review

PURPOSE

Artificial intelligence (AI) applications are growing in smile design and aesthetic procedures. The current expansion and performance of AI models in digital smile design applications have not yet been systematically documented and analyzed. The purpose of this review was to assess the performance of AI models in smile design, assess the criteria of points of reference using AI analysis, and assess different AI software performance.

METHODS

An electronic review was completed in five databases: MEDLINE/PubMed, EMBASE, World of Science, Cochrane, and Scopus. Studies that developed AI models for smile design were included. The search strategy included articles published until November 1, 2024. Two investigators independently evaluated the quality of the studies by applying the Joanna Briggs Institute (JBI) Critical Appraisal Checklist for Quasi-Experimental Studies and Textual Evidence: Expert Opinion Results.

RESULTS

The search resulted in 2653 articles. A total of 2649 were excluded according to the exclusion criteria after reading the title, abstract, and/or full-text review. Four articles published between 2023 and 2024 were included in the present investigation. Two articles compared 2D and 3D points while one article compared the outcome of satisfaction between dentists and patients, and the last article emphasized the ethical components of using AI.

CONCLUSION

The results of the studies reviewed in this paper suggest that AI-generated smile designs are not significantly different from manually created designs in terms of esthetic perception. 3D designs are more accurate than 2D designs and offer more advantages. More articles are needed in the field of AI and smile design.

Fracture detection of distal radius using deep- learning-based dual-channel feature fusion algorithm

PURPOSE

Distal radius fracture is a common trauma fracture and timely preoperative diagnosis is crucial for the patient's recovery. With the rise of deep-learning applications in the medical field, utilizing deep-learning for diagnosing distal radius fractures has become a significant topic. However, previous research has suffered from low detection accuracy and poor identification of occult fractures. This study aims to design an improved deep-learning model to assist surgeons in diagnosing distal radius fractures more quickly and accurately.

METHODS

This study, inspired by the comprehensive analysis of anteroposterior and lateral X-ray images by surgeons in diagnosing distal radius fractures, designs a dual-channel feature fusion network for detecting distal radius fractures. Based on the Faster region-based convolutional neural network framework, an additional Residual Network 50, which is integrated with the Deformable and Separable Attention mechanism, was introduced to extract semantic information from lateral X-ray images of the distal radius. The features extracted from the 2 channels were then combined via feature fusion, thus enriching the network's feature information. The focal loss function was also employed to address the sample imbalance problem during the training process.The selection of cases in this study was based on distal radius X-ray images retrieved from the hospital's imaging database, which met the following criteria: inclusion criteria comprised clear anteroposterior and lateral X-ray images, which were diagnosed as distal radius fractures by experienced radiologists. The exclusion criteria encompassed poor image quality, the presence of severe multiple or complex fractures, as well as non-adult or special populations (e.g., pregnant women). All cases meeting the inclusion criteria were labeled as distal radius fracture cases for model training and evaluation. To assess the model's performance, this study employed several metrics, including accuracy, precision, recall, area under the precision-recall curve, and intersection over union.

RESULTS

The proposed dual-channel feature fusion network achieved an average precision (AP)50 of 98.5%, an AP75 of 78.4%, an accuracy of 96.5%, and a recall of 94.7%. When compared to traditional models, such as Faster region-based convolutional neural network, which achieved an AP50 of 94.1%, an AP75 of 70.6%, a precision of 91.1%, and a recall of 92.3%, our method shows notable improvements in all key metrics. Similarly, when compared to other classic object detection networks like You Only Look Once version 4 (AP50=95.2%, AP75=72.2 %, precision=91.2%, recall=92.4%) and You Only Look Once version 5s (AP50=95.1%, AP75=73.8%, precision=93.7%, recall=92.8%), the dual-channel feature fusion network outperforms them in precision, recall, and AP scores. These results highlight the superior accuracy and reliability of the proposed method, particularly in identifying both apparent and occult distal radius fractures, demonstrating its effectiveness in clinical applications where precise detection of subtle fractures is critical.

CONCLUSION

This study found that combining anteroposterior and lateral X-ray images of the distal radius as input for deep-learning algorithms can more accurately and efficiently identify distal radius fractures, providing a reference for research on distal radius fractures.

Mapping the Use of Artificial Intelligence-Based Image Analysis for Clinical Decision-Making in Dentistry: A Scoping Review

OBJECTIVES

Artificial intelligence (AI) is an emerging field in dentistry. AI is gradually being integrated into dentistry to improve clinical dental practice. The aims of this scoping review were to investigate the application of AI in image analysis for decision-making in clinical dentistry and identify trends and research gaps in the current literature.

MATERIAL AND METHODS

This review followed the guidelines provided by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews (PRISMA-ScR). An electronic literature search was performed through PubMed and Scopus. After removing duplicates, a preliminary screening based on titles and abstracts was performed. A full-text review and analysis were performed according to predefined inclusion criteria, and data were extracted from eligible articles.

RESULTS

Of the 1334 articles returned, 276 met the inclusion criteria (consisting of 601,122 images in total) and were included in the qualitative synthesis. Most of the included studies utilized convolutional neural networks (CNNs) on dental radiographs such as orthopantomograms (OPGs) and intraoral radiographs (bitewings and periapicals). AI was applied across all fields of dentistry - particularly oral medicine, oral surgery, and orthodontics - for direct clinical inference and segmentation. AI-based image analysis was use in several components of the clinical decision-making process, including diagnosis, detection or classification, prediction, and management.

CONCLUSIONS

A variety of machine learning and deep learning techniques are being used for dental image analysis to assist clinicians in making accurate diagnoses and choosing appropriate interventions in a timely manner.

Applications of artificial intelligence in dentomaxillofacial imaging: a systematic review

BACKGROUND

Artificial intelligence (AI) technology has been increasingly developed in oral and maxillofacial imaging. The aim of this systematic review was to assess the applications and performance of the developed algorithms in different dentomaxillofacial imaging modalities.

STUDY DESIGN

A systematic search of PubMed and Scopus databases was performed. The search strategy was set as a combination of the following keywords: "Artificial Intelligence," "Machine Learning," "Deep Learning," "Neural Networks," "Head and Neck Imaging," and "Maxillofacial Imaging." Full-text screening and data extraction were independently conducted by two independent reviewers; any mismatch was resolved by discussion. The risk of bias was assessed by one reviewer and validated by another.

RESULTS

The search returned a total of 3,392 articles. After careful evaluation of the titles, abstracts, and full texts, a total number of 194 articles were included. Most studies focused on AI applications for tooth and implant classification and identification, 3-dimensional cephalometric landmark detection, lesion detection (periapical, jaws, and bone), and osteoporosis detection.

CONCLUSION

Despite the AI models' limitations, they showed promising results. Further studies are needed to explore specific applications and real-world scenarios before confidently integrating these models into dental practice.

Embracing technological revolution: A panorama of machine learning in dentistry

BACKGROUND

The overarching aim of this study is to furnish dental experts and researchers with a comprehensive understanding of the role of machine learning in dentistry. This entails a nuanced understanding of prevailing technologies, discerning emerging trends, and providing strategic guidance for future research endeavors and practical implementations.

MATERIAL AND METHODS

We assessed the literature by looking for papers related to the issue after 2019 in the Pubmed, Web of Science, and Google Scholar databases. A narrative review of 29 papers satisfying the search criteria was undertaken, with an emphasis on the application of machine learning in dentistry.

RESULTS

A review was conducted, including 29 publications. The advent of emerging technologies holds promise for enhancing the accuracy and efficiency of dental diagnosis, treatment, and prognosis. Nevertheless, the intricate nature of oral disease diagnosis and outcome prediction mandates acknowledgment of variables such as individual idiosyncrasies, lifestyle, genetics, image quality, and tooth morphology. These factors may impact the precision of machine learning models. Dental professionals should not rely solely on AI-based results but rather use them as references. Integrating these findings with clinical examinations, assessing the patient's overall health, and oral condition is crucial for informed decision-making.

CONCLUSIONS

This review explores the clinical applications of machine learning in dentistry, encompassing disciplines like cariology, endodontics, periodontology, oral medicine, oral and maxillofacial surgery, prosthodontics and orthodontics. It serves as a valuable resource for dental practitioners and scholars in understanding the computer algorithms employed in each study, facilitating the clinical translation of machine learning research outcomes.

Artificial intelligent-driven decision-making for automating root fracture detection in periapical radiographs

BACKGROUND

The detection and early diagnosis of root fractures can be challenging; this difficulty applies particularly to newly qualified dentists. Aside from clinical examination, diagnosis often requires radiographic assessment. Nonetheless, human fallibility can introduce errors due to a lack of experience.

AIM

The proposed system aimed to assist in detecting root fractures through the integration of artificial intelligence techniques into the diagnosis process as a step for automating dental diagnosis and decision-making processes.

MATERIALS AND METHOD

A total of 400 radiographic images of fractured and unfractured teeth were obtained for the present research. Data handling techniques were implemented to balance the distribution of the samples. The AI-based system used the voting technique for five different pretrained models namely, VGG16, VGG19, ResNet50. DenseNet121, and DenseNet169 to perform the analysis. The parameters used for the analysis of the models are loss and accuracy curves.

RESULTS

VGG16 exhibited notable success with low training and validation losses (0.09% and 0.18%, respectively), high specificity, sensitivity, and positive predictive value (PPV). VGG19 showed potential overfitting concerns, while ResNet50 displayed progress in minimizing loss but exhibited bias toward unfractured cases. DenseNet121 effectively addressed overfitting and noise issues, achieving balanced metrics and impressive PPVs for both fractured and unfractured cases (0.933 and 0.898 respectively). With increased depth, DenseNet169 demonstrated enhanced generalization capability.

CONCLUSION

The proposed AI- based system demonstrated high precision and sensitivity for detecting root fractures in endodontically treated teeth by utilizing the voting method.

Empowering Modern Dentistry: The Impact of Artificial Intelligence on Patient Care and Clinical Decision Making

Advancements in artificial intelligence (AI) are poised to catalyze a transformative shift across diverse dental disciplines including endodontics, oral radiology, orthodontics, pediatric dentistry, periodontology, prosthodontics, and restorative dentistry. This narrative review delineates the burgeoning role of AI in enhancing diagnostic precision, streamlining treatment planning, and potentially unveiling innovative therapeutic modalities, thereby elevating patient care standards. Recent analyses corroborate the superiority of AI-assisted methodologies over conventional techniques, affirming their capacity for personalization, accuracy, and efficiency in dental care. Central to these AI applications are convolutional neural networks and deep learning models, which have demonstrated efficacy in diagnosis, prognosis, and therapeutic decision making, in some instances surpassing traditional methods in complex cases. Despite these advancements, the integration of AI into clinical practice is accompanied by challenges, such as data security concerns, the demand for transparency in AI-generated outcomes, and the imperative for ongoing validation to establish the reliability and applicability of AI tools. This review underscores the prospective benefits of AI in dental practice, envisioning AI not as a replacement for dental professionals but as an adjunctive tool that fortifies the dental profession. While AI heralds improvements in diagnostics, treatment planning, and personalized care, ethical and practical considerations must be meticulously navigated to ensure responsible development of AI in dentistry.

Artificial intelligence vs. semi-automated segmentation for assessment of dental periapical lesion volume index score: A cone-beam CT study

INTRODUCTION

Cone beam computed tomography periapical volume index (CBCTPAVI) is a categorisation tool to assess periapical lesion size in three-dimensions and predict treatment outcomes. This index was determined using a time-consuming semi-automatic segmentation technique. This study compared artificial intelligence (AI) with semi-automated segmentation to determine AI's ability to accurately determine CBCTPAVI score.

METHODS

CBCTPAVI scores for 500 tooth roots were determined using both the semi-automatic segmentation technique in three-dimensional imaging analysis software (Mimics Research™) and AI (Diagnocat™). A confusion matrix was created to compare the CBCTPAVI score by the AI with the semi-automatic segmentation technique. Evaluation metrics, precision, recall, F1-score (2×precision×recallprecision+recall), and overall accuracy were determined.

RESULTS

In 84.4 % (n = 422) of cases the AI classified CBCTPAVI score the same as the semi-automated technique. AI was unable to classify any lesion as index 1 or 2, due to its limitation in small volume measurement. When lesions classified as index 1 and 2 by the semi-automatic segmentation technique were excluded, the AI demonstrated levels of precision, recall and F1-score, all above 0.85, for indices 0, 3-6; and accuracy over 90 %.

CONCLUSIONS

Diagnocat™ with its ability to determine CBCTPAVI score in approximately 2 min following upload of the CBCT could be an excellent and efficient tool to facilitate better monitoring and assessment of periapical lesions in everyday clinical practice and/or radiographic reporting. However, to assess three-dimensional healing of smaller lesions (with scores 1 and 2), further advancements in AI technologies are needed.

Deep learning for determining the difficulty of endodontic treatment: a pilot study

BACKGROUND

To develop and validate a deep learning model for automated assessment of endodontic case difficulty from periapical radiographs.

METHODS

A dataset of 1,386 periapical radiographs was compiled from two clinical sites. Two dentists and two endodontists annotated the radiographs for difficulty using the "simple assessment" criteria from the American Association of Endodontists' case difficulty assessment form in the Endocase application. A classification task labeled cases as "easy" or "hard", while regression predicted overall difficulty scores. Convolutional neural networks (i.e. VGG16, ResNet18, ResNet50, ResNext50, and Inception v2) were used, with a baseline model trained via transfer learning from ImageNet weights. Other models was pre-trained using self-supervised contrastive learning (i.e. BYOL, SimCLR, MoCo, and DINO) on 20,295 unlabeled dental radiographs to learn representation without manual labels. Both models were evaluated using 10-fold cross-validation, with performance compared to seven human examiners (three general dentists and four endodontists) on a hold-out test set.

RESULTS

The baseline VGG16 model attained 87.62% accuracy in classifying difficulty. Self-supervised pretraining did not improve performance. Regression predicted scores with ± 3.21 score error. All models outperformed human raters, with poor inter-examiner reliability.

CONCLUSION

This pilot study demonstrated the feasibility of automated endodontic difficulty assessment via deep learning models.

Application of deep learning in isolated tooth identification

BACKGROUND

Teeth identification has a pivotal role in the dental curriculum and provides one of the important foundations of clinical practice. Accurately identifying teeth is a vital aspect of dental education and clinical practice, but can be challenging due to the anatomical similarities between categories. In this study, we aim to explore the possibility of using a deep learning model to classify isolated tooth by a set of photographs.

METHODS

A collection of 5,100 photographs from 850 isolated human tooth specimens were assembled to serve as the dataset for this study. Each tooth was carefully labeled during the data collection phase through direct observation. We developed a deep learning model that incorporates the state-of-the-art feature extractor and attention mechanism to classify each tooth based on a set of 6 photographs captured from multiple angles. To increase the validity of model evaluation, a voting-based strategy was applied to refine the test set to generate a more reliable label, and the model was evaluated under different types of classification granularities.

RESULTS

This deep learning model achieved top-3 accuracies of over 90% in all classification types, with an average AUC of 0.95. The Cohen's Kappa demonstrated good agreement between model prediction and the test set.

CONCLUSIONS

This deep learning model can achieve performance comparable to that of human experts and has the potential to become a valuable tool for dental education and various applications in accurately identifying isolated tooth.

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.

Establishment of a 3D esthetic analysis workflow on 3D virtual patient and preliminary evaluation

BACKGROUND

In esthetic dentistry, a thorough esthetic analysis holds significant role in both diagnosing diseases and designing treatment plans. This study established a 3D esthetic analysis workflow based on 3D facial and dental models, and aimed to provide an imperative foundation for the artificial intelligent 3D analysis in future esthetic dentistry.

METHODS

The established 3D esthetic analysis workflow includes the following steps: 1) key point detection, 2) coordinate system redetermination and 3) esthetic parameter calculation. The accuracy and reproducibility of this established workflow were evaluated by a self-controlled experiment (n = 15) in which 2D esthetic analysis and direct measurement were taken as control. Measurement differences between 3D and 2D analysis were evaluated with paired t-tests.

RESULTS

3D esthetic analysis demonstrated high consistency and reliability (0.973 < ICC < 1.000). Compared with 2D measurements, the results from 3D esthetic measurements were closer to direct measurements regarding tooth-related esthetic parameters (P<0.05).

CONCLUSIONS

The 3D esthetic analysis workflow established for 3D virtual patients demonstrated a high level of consistency and reliability, better than 2D measurements in the precision of tooth-related parameter analysis. These findings indicate a highly promising outlook for achieving an objective, precise, and efficient esthetic analysis in the future, which is expected to result in a more streamlined and user-friendly digital design process. This study was registered with the Ethics Committee of Peking University School of Stomatology in September 2021 with the registration number PKUSSIRB-202168136.

Semantic segmentation for tooth cracks using improved DeepLabv3+ model

Objective

Accurate and prompt detection of cracked teeth plays a critical role for human oral health. The aim of this paper is to evaluate the performance of a tooth crack segmentation model (namely, FDB-DeepLabv3+) on optical microscopic images.

Method

The FDB-DeepLabv3+ model proposed here improves feature learning by replacing the backbone with ResNet50. Feature pyramid network (FPN) is introduced to fuse muti-level features. Densely linked atrous spatial pyramid pooling (Dense ASPP) is applied to achieve denser pixel sampling and wider receptive field. Bottleneck attention module (BAM) is embedded to enhance local feature extraction.

Results

Through testing on a self-made hidden cracked tooth dataset, the proposed method outperforms four classical networks (FCN, U-Net, SegNet, DeepLabv3+) on segmentation results in terms of mean pixel accuracy (MPA) and mean intersection over union (MIoU). The network achieves an increase of 11.41% in MPA and 12.14% in MIoU compared to DeepLabv3+. Ablation experiments shows that all the modifications are beneficial.

Conclusion

An improved network is designed for segmenting tooth surface cracks with good overall performance and robustness, which may hold significant potential in computer-aided diagnosis of cracked teeth.

Artificial Intelligence in Endodontics: A Scoping Review

Artificial intelligence (AI) is transforming the diagnostic methods and treatment approaches in the constantly evolving field of endodontics. The current review discusses the recent advancements in AI; with a specific focus on convolutional and artificial neural networks. Apparently, AI models have proved to be highly beneficial in the analysis of root canal anatomy, detecting periapical lesions in early stages as well as providing accurate working-length determination. Moreover, they seem to be effective in predicting the treatment success next to identifying various conditions e.g., dental caries, pulpal inflammation, vertical root fractures, and expression of second opinions for non-surgical root canal treatments. Furthermore, AI has demonstrated an exceptional ability to recognize landmarks and lesions in cone-beam computed tomography scans with consistently high precision rates. While AI has significantly promoted the accuracy and efficiency of endodontic procedures, it is of high importance to continue validating the reliability and practicality of AI for possible widespread integration into daily clinical practice. Additionally, ethical considerations related to patient privacy, data security, and potential bias should be carefully examined to ensure the ethical and responsible implementation of AI in endodontics.

Machine learning techniques for periodontitis and dental caries detection: A narrative review

OBJECTIVES

In recent years, periodontitis, and dental caries have become common in humans and need to be diagnosed in the early stage to prevent severe complications and tooth loss. These dental issues are diagnosed by visual inspection, measuring pocket probing depth, and radiographs findings from experienced dentists. Though a glut of machine learning (ML) algorithms has been proposed for the automated detection of periodontitis, and dental caries, determining which ML techniques are suitable for clinical practice remains under debate. This review aims to identify the research challenges by analyzing the limitations of current methods and how to address these to obtain robust systems suitable for clinical use or point-of-care testing.

METHODS

An extensive search of the literature published from 2015 to 2022 written in English, related to the subject of study was sought by searching the electronic databases: PubMed, Institute of Electrical and Electronics Engineers (IEEE) Xplore, and ScienceDirect.

RESULTS

The initial electronic search yielded 1743 titles, and 55 studies were eventually included based on the selection criteria adopted in this review. Studies selected were on ML applications for the automatic detection of periodontitis and dental caries and related dental issues: Apical lessons, Periodontal bone loss, and Vertical root fracture.

CONCLUSION

While most of the ML-based studies use radiograph images for the detection of periodontitis and dental caries, few pieces of the literature revealed that good diagnostic accuracy could be achieved by training the ML model even with mobile photos representing the images of dental issues. Nowadays smartphones are used in every sector for different applications. Training the ML model with as many images of dental issues captured by the smartphone can achieve good accuracy, reduce the cost of clinical diagnosis, and provide user interaction.

The Application of Deep Learning on CBCT in Dentistry

Cone beam computed tomography (CBCT) has become an essential tool in modern dentistry, allowing dentists to analyze the relationship between teeth and the surrounding tissues. However, traditional manual analysis can be time-consuming and its accuracy depends on the user's proficiency. To address these limitations, deep learning (DL) systems have been integrated into CBCT analysis to improve accuracy and efficiency. Numerous DL models have been developed for tasks such as automatic diagnosis, segmentation, classification of teeth, inferior alveolar nerve, bone, airway, and preoperative planning. All research articles summarized were from Pubmed, IEEE, Google Scholar, and Web of Science up to December 2022. Many studies have demonstrated that the application of deep learning technology in CBCT examination in dentistry has achieved significant progress, and its accuracy in radiology image analysis has reached the level of clinicians. However, in some fields, its accuracy still needs to be improved. Furthermore, ethical issues and CBCT device differences may prohibit its extensive use. DL models have the potential to be used clinically as medical decision-making aids. The combination of DL and CBCT can highly reduce the workload of image reading. This review provides an up-to-date overview of the current applications of DL on CBCT images in dentistry, highlighting its potential and suggesting directions for future research.

Diagnostic efficacy of cone-beam computed tomography for detection of vertical root fractures in endodontically treated teeth: a systematic review

BACKGROUND

Vertical root fractures (VRFs) sometimes occur in endodontically treated teeth. They have a difficult diagnosis and a dismal result. The objective of this review was to evaluate the diagnostic performance of cone-beam computed tomography (CBCT) for detecting VRFs in teeth that had undergone endodontic treatment.

METHODS

Literature was reviewed from Web of Science, PubMed, Cochrane Review, SCOPUS, and Embase databases between 2000 and 2022. The searched keywords included "endodontically treated teeth," "cone-beam computed tomography," "CBCT," "tooth fracture," "vertical root fracture," "VRF," "accuracy," "sensitivity," and "specificity." Only articles in the English language were included. The final analysis included 20 papers that satisfied the eligibility requirements.

RESULTS

The overall mean ± SD values (%) for the diagnostic sensitivity and specificity of CBCT for detection of VRFs in endodontically treated teeth in the presence of root-filling materials without an intracanal post were 71.50 ± 22.19 and 75.64 ± 19.41, respectively. The overall mean (SD) value (%) for the sensitivity of CBCT for the detection of VRFs in the presence of root-filling materials and intracanal posts was 72.76 (18.73), while the mean (SD) specificity was 75.44 (18.26). The accuracy of CBCT (mean ± SD) was 78.47 ± 17.19% and 74.02 ± 10.64%, respectively, for teeth without intracanal posts and those with posts.

CONCLUSIONS

Further clinical research is needed to validate the optimum efficiency of CBCT as a diagnostic technique for detecting VRFs in teeth that have had endodontic treatment, given the low sensitivity, significant heterogeneity of studies, and lack of in-vivo studies on the subject.

Detection of vertical root fractures by cone-beam computed tomography based on deep learning

OBJECTIVES

This study aims to evaluate the performance of ResNet models in the detection of in vitro and in vivo vertical root fractures (VRF) in Cone-beam Computed Tomography (CBCT) images.

METHODS

A CBCT image dataset consisting of 28 teeth (14 intact and 14 teeth with VRF, 1641 slices) from 14 patients, and another dataset containing 60 teeth (30 intact and 30 teeth with VRF, 3665 slices) from an in vitro model were used for the establishment of VRFconvolutional neural network (CNN) models. The most popular CNN architecture ResNet with different layers was fine-tuned for the detection of VRF. Sensitivity, specificity, accuracy, PPV (positive predictive value), NPV (negative predictive value), and AUC (the area under the receiver operating characteristic curve) of the VRF slices classified by the CNN in the test set were compared. Two oral and maxillofacial radiologists independently reviewed all the CBCT images of the test set, and intraclass correlation coefficients (ICCs) were calculated to assess the interobserver agreement for the oral maxillofacial radiologists.

RESULTS

The AUC of the models on the patient data were: 0.827(ResNet-18), 0.929(ResNet-50), and 0.882(ResNet-101). The AUC of the models on the mixed data get improved as:0.927(ResNet-18), 0.936(ResNet-50), and 0.893(ResNet-101). The maximum AUC were: 0.929 (0.908-0.950, 95% CI) and 0.936 (0.924-0.948, 95% CI) for the patient data and mixed data from ResNet-50, which is comparable to the AUC (0.937 and 0.950) for patient data and (0.915 and 0.935) for the mixed data obtained from the two oral and maxillofacial radiologists, respectively.

CONCLUSIONS

Deep-learning models showed high accuracy in the detection of VRF using CBCT images. The data obtained from the in vitro VRF model increases the data scale, which is beneficial to the training of deep-learning models.

Developments and Performance of Artificial Intelligence Models Designed for Application in Endodontics: A Systematic Review

Technological advancements in health sciences have led to enormous developments in artificial intelligence (AI) models designed for application in health sectors. This article aimed at reporting on the application and performances of AI models that have been designed for application in endodontics. Renowned online databases, primarily PubMed, Scopus, Web of Science, Embase, and Cochrane and secondarily Google Scholar and the Saudi Digital Library, were accessed for articles relevant to the research question that were published from 1 January 2000 to 30 November 2022. In the last 5 years, there has been a significant increase in the number of articles reporting on AI models applied for endodontics. AI models have been developed for determining working length, vertical root fractures, root canal failures, root morphology, and thrust force and torque in canal preparation; detecting pulpal diseases; detecting and diagnosing periapical lesions; predicting postoperative pain, curative effect after treatment, and case difficulty; and segmenting pulp cavities. Most of the included studies (n = 21) were developed using convolutional neural networks. Among the included studies. datasets that were used were mostly cone-beam computed tomography images, followed by periapical radiographs and panoramic radiographs. Thirty-seven original research articles that fulfilled the eligibility criteria were critically assessed in accordance with QUADAS-2 guidelines, which revealed a low risk of bias in the patient selection domain in most of the studies (risk of bias: 90%; applicability: 70%). The certainty of the evidence was assessed using the GRADE approach. These models can be used as supplementary tools in clinical practice in order to expedite the clinical decision-making process and enhance the treatment modality and clinical operation.