Dental Images Research Articles

Ethical aspects and applications of artificial intelligence in maxillofacial imaging

Artificial intelligence (AI) seeks to develop algorithms and software capable of emulating intelligent human actions. AI applications in dentistry hold considerable promise for enhancing the precision and effectiveness of diverse dental imaging techniques. While this domain is still relatively young, it demands thorough exploration. Human supervision remains essential to mitigate potential adverse consequences. This article endeavors to shed light on the prevailing ethical considerations stemming from the integration of artificial intelligence into dental practice. It seeks to stimulate discourse surrounding potential ethical pitfalls and encourages critical examination of these issues.

Radiation protection in dental imaging: Evaluating the impact of the SABA thyroid shield during panoramic and cone beam computed tomography

BackgroundDental panoramic radiography (DPR) and cone beam computed tomography (CBCT) are an imaging modalities in dentistry. However, these procedures involve exposure to ionising radiation, raising concerns about radiation-induced thyroid damage. To mitigate these risks, thyroid shields have been introduced. This study aimed to evaluate the effectiveness of the SABA thyroid shield in reducing thyroid radiation exposure during DPR and CBCT scans. MethodAn experimental study to measure surface dose in the thyroid region during (DPR) and (CBCT) scans using an Anthropomorphic Phantom and a Solid-State Detector. The attenuation percentage was calculated between radiation surface doses with and without shielding. The significance of the Saba thyroid shield was calculated using a t-test. ResultsFor DPR scans, the average surface doses in the thyroid reduced significantly from 301.1 μGy (at 85 kVp) and 170.3 μGy (at 60 kVp) to 64.43 μGy and 12.94 μGy, respectively, when the Saba thyroid shield was used. For CBCT scans, the average surface doses in the thyroid decreased significantly from 814.43 μGy (for 11 × 13 cm2 field size) and 32.40 μGy (for 5 × 5 cm2 field size) to 62.91 μGy and 12.07 μGy, respectively, with the application of the Saba thyroid shield. The maximum attenuation percentages in the thyroid for the DPR and CBCT scans were 92.40% and 92.27%, respectively. Statistically significant differences between the surface dose reduction with the Saba shield and without it was observed in both DPR scans (p < 0.0000) and CBCT scans (p < 0.0003). ConclusionThe study demonstrates that Saba thyroid shields effectively reduce doses in the thyroid region during dental DPR and CBCT scans. The dose reduction depends on tube voltage for DPR scans, field size, and its position for CBCT scans. Findings highlight the importance of using Saba thyroid shields to minimise radiation exposure and protect the thyroid gland during dental scans.

Accuracy of measurements on CBCT-generated digital models using different exposure parameters (in vitro study).

Dental imaging comprises a wide range of techniques and modalities, each with different diagnostic features influenced by numerous parameters, all of which contribute to the precision and effectiveness of dental evaluations and treatments. This study examined the dimensional reproducibility of arch measurements from CBCT scans with different voxel sizes and exposure parameters compared to a reference model from Extra-oral scanners (EOS) and how these parameters affected digital matching and diagnostic image quality. A diagnostic observational study of arch dimensions, including inter-canine, inter-premolar, intermolar, arch width and arch length, was conducted on digital models (DMs) created from 65 CBCT scans of a full dentate epoxy maxillary model. The measurements obtained from EOS scans served as the control for the study. Normality was tested with the Shapiro-Wilk test, comparisons used the Kruskal-Wallis test with Bonferroni-adjusted pairwise comparisons for significant results, and data were analysed using IBM SPSS (Version 26.0), with significance set at p < .05. Significant deviations were revealed among study groups, with group I (smallest voxel size) consistently displaying the lowest values, mean (SD) deviation was reported as 0.01 (0.006) and group IV (lowest kV value) exhibiting the highest deviations, mean (SD) deviation of 0.16 (0.17). Combining a small voxel size (0.12 mm) with high milliampere (8 mA) and kilovoltage (90 kV) settings in CBCT ensured detailed anatomical visualization and accurate linear measurements, crucial for precise dental assessments, and emphasizing the necessity for strict control over CBCT parameters in dental applications.

Ex vivo human teeth imaging with various photoacoustic imaging systems.

Dental caries cause pain and if not diagnosed, it may lead to the loss of teeth in extreme cases. Dental X-ray imaging is the gold standard for caries detection; however, it cannot detect hidden caries. In addition, the ionizing nature of X-ray radiation is another concern. Hence, other alternate imaging modalities like photoacoustic (PA) imaging are being explored for dental imaging. Here, we demonstrate the feasibility of acoustic resolution photoacoustic microscopy (ARPAM) to image a tooth with metal filling, circular photoacoustic computed tomography (cPACT) to acquire images of teeth with caries and pigmentation, and linear array-based photoacoustic imaging (lPACT) of teeth with caries and pigmentation. The cavity measured with lPACT imaging is compared with the X-ray computed tomography image. The metal filling and its boundaries are clearly seen in the ARPAM image. cPACT images at 1064 nm were a better representative of the tooth surface compared to the images acquired at 532 nm. It was possible to detect the cavities present in the dentine when lPACT imaging was used. The PA signal from the pigmented caries on the lateral surface (occlusion view) of the tooth was high when imaged using the lPACT system.

A Review of Radiology and Optical Coherence Tomography in Dental Diagnostics

Background: The rapid evolution of imaging technologies has significantly transformed dentistry, improving the diagnosis and treatment of various dental conditions. While traditional radiological methods such as X-rays and Cone Beam Computed Tomography (CBCT) remain the cornerstone of dental imaging, the emergence of Optical Coherence Tomography (OCT) offers a non-invasive, radiation-free alternative. This review examines the comparative efficacy of X-ray and OCT technologies in diagnosing hard tissue conditions in the oral cavity. Methods: The review analyzed the application of OCT and radiology in dental diagnostics through a custom-built Swept-Source OCT system, assessing its performance against traditional radiological methods. The study utilized both in vivo and ex vivo evaluations, considering variables such as resolution, depth penetration, and image clarity across different dental conditions. Results: OCT demonstrated superior capabilities in detecting early-stage cavities, fractures, and minor imperfections, particularly when high-resolution imaging was required. However, its limited penetration compared to X-rays restricted its application in comprehensive bone assessments. Radiology, especially CBCT, provided essential volumetric data crucial for evaluating bone density, fractures, and post-surgical conditions, though it posed concerns related to radiation exposure. Combining both modalities proved beneficial, offering complementary insights that enhanced overall diagnostic accuracy. Conclusion: While X-ray-based radiology remains indispensable for broad evaluations of dental and bone structures, OCT presents a valuable adjunct, particularly for high-precision assessments of soft tissues and small surface defects. The synergistic use of these technologies can provide a more holistic approach to dental diagnostics, potentially improving patient care and treatment outcomes. Future advancements in OCT could further expand its role in routine dental practice, enhancing diagnostic accuracy while minimizing radiation exposure risks.

Estimation of human age using machine learning on panoramic radiographs for Brazilian patients

This paper addresses a relevant problem in Forensic Sciences by integrating radiological techniques with advanced machine learning methodologies to create a non-invasive, efficient, and less examiner-dependent approach to age estimation. Our study includes a new dataset of 12,827 dental panoramic X-ray images representing the Brazilian population, covering an age range from 2.25 to 96.50 years. To analyze these exams, we employed a model adapted from InceptionV4, enhanced with data augmentation techniques. The proposed approach achieved robust and reliable results, with a Test Mean Absolute Error of 3.1 years and an R-squared value of 95.5%. Professional radiologists have validated that our model focuses on critical features for age assessment used in odontology, such as pulp chamber dimensions and stages of permanent teeth calcification. Importantly, the model also relies on anatomical information from the mandible, maxillary sinus, and vertebrae, which enables it to perform well even in edentulous cases. This study demonstrates the significant potential of machine learning to revolutionize age estimation in Forensic Science, offering a more accurate, efficient, and universally applicable solution.

Diagnostic accuracy of artificial intelligence models in detecting osteoporosis using dental images: a systematic review and meta-analysis.

The current study aimed to systematically review the literature on the accuracy of artificial intelligence (AI) models for osteoporosis (OP) diagnosis using dental images. A thorough literature search was executed in October 2022 and updated in November 2023 across multiple databases, including PubMed, Scopus, Web of Science, and Google Scholar. The research targeted studies using AI models for OP diagnosis from dental radiographs. The main outcomes were the sensitivity and specificity of AI models regarding OP diagnosis. The "meta" package from the R Foundation was selected for statistical analysis. A random-effects model, along with 95% confidence intervals, was utilized to estimate pooled values. The Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) tool was employed for risk of bias and applicability assessment. Among 640 records, 22 studies were included in the qualitative analysis and 12 in the meta-analysis. The overall sensitivity for AI-assisted OP diagnosis was 0.85 (95% CI, 0.70-0.93), while the pooled specificity equaled 0.95 (95% CI, 0.91-0.97). Conventional algorithms led to a pooled sensitivity of 0.82 (95% CI, 0.57-0.94) and a pooled specificity of 0.96 (95% CI, 0.93-0.97). Deep convolutional neural networks exhibited a pooled sensitivity of 0.87 (95% CI, 0.68-0.95) and a pooled specificity of 0.92 (95% CI, 0.83-0.96). This systematic review corroborates the accuracy of AI in OP diagnosis using dental images. Future research should expand sample sizes in test and training datasets and standardize imaging techniques to establish the reliability of AI-assisted methods in OP diagnosis through dental images.

MAPPNet: A Multi-Scale Attention Pyramid Pooling Network for Dental Calculus Segmentation

Dental diseases are among the most prevalent diseases globally, and accurate segmentation of dental calculus images plays a crucial role in periodontal disease diagnosis and treatment planning. However, the current methods are not stable and reliable enough due to the variable morphology of dental calculus and the blurring of the boundaries between the dental edges and the surrounding tissues; therefore, our hope is to propose an accurate and reliable calculus segmentation algorithm to improve the efficiency of clinical detection. We propose a multi-scale attention pyramid pooling network (MAPPNet) to enhance the performance of dental calculus segmentation. The network incorporates a multi-scale fusion strategy in both the encoder and decoder, forming a model with a dual-ended multi-scale structure. This design, in contrast to employing a multi-scale fusion scheme at a single end, enables more effective capturing of features from diverse scales. Furthermore, the attention pyramid pooling module (APPM) reconstructs the features on this map by leveraging a spatial-first and channel-second attention mechanism. APPM enables the network to adaptively adjust the weights of different locations and channels in the feature map, thereby enhancing the perception of important regions and key features. Experimental evaluation of our collected dental calculus segmentation dataset demonstrates the superior performance of MAPPNet, which achieves an intersection-over-union of 81.46% and an accuracy rate of 98.35%. Additionally, on two publicly available datasets, ISIC2018 (skin lesion dataset) and Kvasir-SEG (gastrointestinal polyp segmentation dataset), MAPPNet achieved an intersection-over-union of 76.48% and 91.38%, respectively. These results validate the effectiveness of our proposed network in accurately segmenting lesion regions and achieving high accuracy rates, surpassing many existing segmentation methods.

Clinical boundary conditions for propagation-based X-ray phase contrast imaging: from bio-sample models targeting to clinical applications.

Typical propagation-based X-ray phase contrast imaging (PB-PCI) experiments using polyenergetic sources are tested in very ideal conditions: low-energy spectrum (mainly characteristic X-rays), small thickness and homogeneous materials considered weakly absorbing objects, large object-to-detector distance, long exposure times and non-clinical detector. Explore PB-PCI features using boundary conditions imposed by a low power polychromatic X-ray source (X-ray spectrum without characteristic X-rays), thick and heterogenous materials and a small area imaging detector with high low-detection radiation threshold, elements commonly found in a clinical scenario. A PB-PCI setup implemented using a microfocus X-ray source and a dental imaging detector was characterized in terms of different spectra and geometric parameters on the acquired images. Test phantoms containing fibers and homogeneous materials with close attenuation characteristics and animal bone and mixed soft tissues (bio-sample models) were analyzed. Contrast to Noise Ratio (CNR), system spatial resolution and Kerma values were obtained for all images. Phase contrast images showed CNR up to 15% higher than conventional contact images. Moreover, it is better seen when large magnifications (>3) and object-to-detector distances (>13 cm) were used. The influence of the spectrum was not appreciable due to the low efficiency of the detector (thin scintillator screen) at high energies. Despite the clinical boundary condition used in this work, regarding the X-ray spectrum, thick samples, and detection system, it was possible to acquire phase contrast images of biological samples.

A Review of Dental Imaging and Treatment Protocol Prior to Open Heart Valve Surgery: An Evidence-Based Approach

Context: Open heart valve surgery is associated with a significant risk of infective endocarditis (IE), often linked to oral infections. Despite existing guidelines recommending dental evaluations before surgery, there is no standardized protocol for dental imaging and treatment in this high-risk population. This study aims to present the first evidence-based protocol for dental imaging and treatment to systematically identify and address oral infections, thereby reducing perioperative risks and improving outcomes. Evidence Acquisition: A comprehensive review of current literature on the relationship between oral infections and infective endocarditis was conducted. Existing guidelines from the American Heart Association (AHA) and European Society of Cardiology (ESC) were analyzed. Imaging modalities were integrated into the proposed protocol for preoperative dental evaluation. A risk-based classification system for common dental conditions and antibiotic prophylaxis was developed. Results: The protocol utilizes Imaging techniques to detect oral infections and conditions that could lead to bacteremia, such as dental caries, periodontal disease, and impacted teeth. It also introduces a risk-based approach for managing dental conditions, prioritizing treatments based on urgency, and recommending appropriate antibiotic prophylaxis. This interdisciplinary approach enhances collaboration between dental and cardiac care teams. Conclusions: This study introduces a novel dental imaging and treatment protocol for patients undergoing open heart valve surgery, significantly reducing the risk of infective endocarditis. By addressing oral health systematically before surgery, this protocol improves surgical outcomes and sets a new standard of care for cardiac patients. Future research should validate its effectiveness across diverse healthcare settings and evaluate its long-term impact on patient outcomes.

RadImageNet and ImageNet as Datasets for Transfer Learning in the Assessment of Dental Radiographs: A Comparative Study.

Transfer learning (TL) is an alternative approach to the full training of deep learning (DL) models from scratch and can transfer knowledge gained from large-scale data to solve different problems. ImageNet, which is a publicly available large-scale dataset, is a commonly used dataset for TL-based image analysis; many studies have applied pre-trained models from ImageNet to clinical prediction tasks and have reported promising results. However, some have questioned the effectiveness of using ImageNet, which consists solely of natural images, for medical image analysis. The aim of this study was to evaluate whether pre-trained models using RadImageNet, which is a large-scale medical image dataset, could achieve superior performance in classification tasks in dental imaging modalities compared with ImageNet pre-trained models. To evaluate the classification performance of RadImageNet and ImageNet pre-trained models for TL, two dental imaging datasets were used. The tasks were (1) classifying the presence or absence of supernumerary teeth from a dataset of panoramic radiographs and (2) classifying sex from a dataset of lateral cephalometric radiographs. Performance was evaluated by comparing the area under the curve (AUC). On the panoramic radiograph dataset, the RadImageNet models gave average AUCs of 0.68 ± 0.15 (p < 0.01), and the ImageNet models had values of 0.74 ± 0.19. In contrast, on the lateral cephalometric dataset, the RadImageNet models demonstrated average AUCs of 0.76 ± 0.09, and the ImageNet models achieved values of 0.75 ± 0.17. The difference in performance between RadImageNet and ImageNet models in TL depends on the dental image dataset used.

Dental cavity analysis, prediction, localization, and quantification using computer vision

Dental health assessment is a critical component of overall well-being, and advancements in computer vision and deep learning have opened new avenues for automating and enhancing this process. In this study, we present a comprehensive approach to dental cavity analysis, spanning localization, quantification, and visualization. Our methodology leveraged a diverse dataset of colored dental images that had been meticulously augmented and annotated. The You Only Look Once model was employed for precise dental cavity localization, providing bounding box predictions. Remarkably, these results were obtained based on images from standard device cameras. Subsequently, we introduced the use of the segment anything model segmentation model, known for its zero-shot generalization capabilities, to focus on the exact areas of dental cavities. This approach enhanced the granularity of our analysis, providing dental professionals with detailed visualizations for precise diagnosis. During the quantification phase, we extracted cavity areas from bounding box coordinates, enabling accurate measurement of cavity sizes. The model achieved a notable mean average precision of 0.732, an accuracy of 0.789, and a recall of 0.701. Moreover, the model converged quickly, with most metrics achieving near-optimal results after 100 iterations. This quantitative data augments traditional diagnosis methods, facilitating more informed treatment decisions.

DeMambaNet: Deformable Convolution and Mamba Integration Network for High-Precision Segmentation of Ambiguously Defined Dental Radicular Boundaries.

The incorporation of automatic segmentation methodologies into dental X-ray images refined the paradigms of clinical diagnostics and therapeutic planning by facilitating meticulous, pixel-level articulation of both dental structures and proximate tissues. This underpins the pillars of early pathological detection and meticulous disease progression monitoring. Nonetheless, conventional segmentation frameworks often encounter significant setbacks attributable to the intrinsic limitations of X-ray imaging, including compromised image fidelity, obscured delineation of structural boundaries, and the intricate anatomical structures of dental constituents such as pulp, enamel, and dentin. To surmount these impediments, we propose the Deformable Convolution and Mamba Integration Network, an innovative 2D dental X-ray image segmentation architecture, which amalgamates a Coalescent Structural Deformable Encoder, a Cognitively-Optimized Semantic Enhance Module, and a Hierarchical Convergence Decoder. Collectively, these components bolster the management of multi-scale global features, fortify the stability of feature representation, and refine the amalgamation of feature vectors. A comparative assessment against 14 baselines underscores its efficacy, registering a 0.95% enhancement in the Dice Coefficient and a diminution of the 95th percentile Hausdorff Distance to 7.494.

The Role of Artificial Intelligence in Dental Diagnosis

This paper discusses the potential role of artificial intelligence in enhancing dental diagnosis and care delivery. It describes how AI shows promise in automatic detection of abnormalities in dental images, personalised disease risk assessment through predictive modelling, and generation of comprehensive examination reports. The ability of AI to analyse patterns across medical imaging, genetics, lifestyle factors and health records enables more holistic understanding of oral health in relation to systemic conditions. Clinical decision support through differential diagnoses and evidence-based treatment recommendations aims to augment dentists' decision making. Seamless integration of AI into clinical workflows through interfaces is emphasised as important for adoption. Challenges around data and model validation are also addressed. Continued development of AI aims to realise benefits like earlier disease identification and more proactive, personalised care approaches

Teeth, Tech, and AI: A Revolutionizing approach in Dentistry

Abstract Introduction Digital dentistry equipped with AI based integrated platforms enabled dentists to provide accurate diagnosis, prompt treatment, increased efficiency of evidence-based healthcare service provision. Objectives To explore the role of AI in dentistry with emphasis on public health aspects as oral health awareness, tobacco cessation counselling and tele Dentistry to reconnoitre the role of AI in dental healthcare services and public health surveillance. Methodology Electronic search in various databases with the keywords AI, Digital dentistry, oral health awareness, surveillance, Communication, Dental imaging, were performed such as PubMed/MEDLINE (National Library of Medicine), Scopus (Elsevier), ScienceDirect databases (Elsevier), Web of Science (Clarivate Analytics), and the Cochrane Collaboration (Wiley). 30 full-text articles were selected and systematically analysed and the relevant data were extracted. Result and Conclusion AI-driven platform enhanced oral health awareness, services and facilitated automated evidence synthesis in the field of dentistry making it affordable, accessible and available to the public. Integration of AI and digital dentistry holds the promise of revolutionizing the way we approach oral health care.

The eye of artificial intelligence - Convolutional Neural Networks

Inspired by the biological visual system, the convolutional neural network has been widely studied and invented in the field of artificial intelligence. As one of the important algorithms in artificial neural networks, convolutional neural networks have shown outstanding application potential in fields such as image recognition, computer vision, and natural language processing. This article will focus on exploring the powerful capabilities of convolutional neural networks in image processing. By delving into the implementation process of a convolutional neural network, readers will gain a deeper understanding of its working principles. In addition, this article will briefly introduce three classic models of convolutional neural networks, providing readers with more background knowledge. Next, this paper will analyze in detail two typical application cases of convolutional neural networks in the field of image processing: intelligent transportation systems and dental imaging technology. These cases demonstrate the successful application of convolutional neural networks in practical scenarios, pointing the way for their future development. In the future, convolutional neural networks will be more widely used in fields such as image and video processing as data scale increases and computing power improves. By using techniques such as model compression and hardware optimization, it is made more suitable for low-power and high-efficiency environments, and its interpretability and applicability are enhanced through data augmentation and model interpretation techniques.

Accuracy of automated forensic dental age estimation lab (F-DentEst Lab) on large Malaysian dataset

When a disaster occurs, the authority must prioritise two things. First, the search and rescue of lives, and second, the identification and management of deceased individuals. However, with thousands of dead bodies to be individually identified in mass disasters, forensic teams face challenges such as long working hours resulting in a delayed identification process and a public health concern caused by the decomposition of the body. Using dental panoramic imaging, teeth have been used in forensics as a physical marker to estimate the age of an individual. Traditionally, dental age estimation has been performed manually by experts. Although the procedure is fairly simple, the large number of victims and the limited amount of time available to complete the assessment during large-scale disasters make forensic work even more challenging. The emergence of artificial intelligence (AI) in the fields of medicine and dentistry has led to the suggestion of automating the current process as an alternative to the conventional method. This study aims to test the accuracy and performance of the developed deep convolutional neural network system for age estimation in large, out-of-sample Malaysian children dataset using digital dental panoramic imaging. Forensic Dental Estimation Lab (F-DentEst Lab) is a computer application developed to perform the dental age estimation digitally. The introduction of this system is to improve the conventional method of age estimation that significantly increase the efficiency of the age estimation process based on the AI approach. A total number of one-thousand-eight-hundred-and-ninety-two digital dental panoramic images were retrospectively collected to test the F-DentEst Lab. Data training, validation, and testing have been conducted in the early stage of the development of F-DentEst Lab, where the allocation involved 80 % training and the remaining 20 % for testing. The methodology was comprised of four major steps: image preprocessing, which adheres to the inclusion criteria for panoramic dental imaging, segmentation, and classification of mandibular premolars using the Dynamic Programming-Active Contour (DP-AC) method and Deep Convolutional Neural Network (DCNN), respectively, and statistical analysis. The suggested DCNN approach underestimated chronological age with a small ME of 0.03 and 0.05 for females and males, respectively.

Radiation safety of dentomaxillofacial radiology in Finland: Reported abnormal irradiation-related incidents during 2012–2022

IntroductionThis study evaluates adverse events (AEs) in dentomaxillofacial radiology (DMFR) in Finland, including their quantity, nature, and outcomes. It also compares reporting activity between public and private healthcare organizations, as they share same legal obligations to report AEs. Two-thirds of dental images are taken in public, the rest in private healthcare. MethodsIn Finland, radiation-related AEs are reported to the Radiation and Nuclear Safety Authority (STUK). We categorized DMFR-related AEs by nature, outcome, and imaging modality. We also submitted a questionnaire to STUK to gather information on their observations and remarks made during inspections of establishments providing dental imaging services. ResultsDuring 2012–2022, STUK received reports of 1343 DMFR-related AEs, mostly causing patient harm (92.9%) leading almost always to excessive radiation doses (99.7%). Private healthcare reported most AEs (65.2%), with municipal institutions reporting the remainder (34.8%). Intraoral-related AEs (20.0%) slightly outnumbered panoramic-related ones (18.7%), but the majority (56.7%) didn't specify the imaging modality. CBCT-related incidents were least reported (4.5%). During STUK's inspections, remarks mostly concerned deficiencies in practical quality assurance (31.3%) or technical quality assurance (32.9%). ConclusionDMFR-related AEs may be underreported despite legal obligations, with most stemming from human error, highlighting the need to enhance patient safety culture. There's a notable reporting gap between private and public healthcare providers relative to the number of dental radiographs conducted. Organizational deficiencies in practical and technical quality assurance underscore the importance of ongoing education, as well as monitoring by STUK. Implications for practiceEnhancing patient safety culture in dentistry, including dentomaxillofacial radiology, demands customized multiprofessional development, as unaltered patient safety procedures from medicine and secondary care may not align with dentistry's requirements.

B-MAR: bidirectional artifact representations learning framework for metal artifact reduction in dental CBCT

Objective. Metal artifacts in computed tomography (CT) images hinder diagnosis and treatment significantly. Specifically, dental cone-beam computed tomography (Dental CBCT) images are seriously contaminated by metal artifacts due to the widespread use of low tube voltages and the presence of various high-attenuation materials in dental structures. Existing supervised metal artifact reduction (MAR) methods mainly learn the mapping of artifact-affected images to clean images, while ignoring the modeling of the metal artifact generation process. Therefore, we propose the bidirectional artifact representations learning framework to adaptively encode metal artifacts caused by various dental implants and model the generation and elimination of metal artifacts, thereby improving MAR performance. Approach. Specifically, we introduce an efficient artifact encoder to extract multi-scale representations of metal artifacts from artifact-affected images. These extracted metal artifact representations are then bidirectionally embedded into both the metal artifact generator and the metal artifact eliminator, which can simultaneously improve the performance of artifact removal and artifact generation. The artifact eliminator learns artifact removal in a supervised manner, while the artifact generator learns artifact generation in an adversarial manner. To further improve the performance of the bidirectional task networks, we propose artifact consistency loss to align the consistency of images generated by the eliminator and the generator with or without embedding artifact representations. Main results. To validate the effectiveness of our algorithm, experiments are conducted on simulated and clinical datasets containing various dental metal morphologies. Quantitative metrics are calculated to evaluate the results of the simulation tests, which demonstrate b-MAR improvements of >1.4131 dB in PSNR, >0.3473 HU decrements in RMSE, and >0.0025 promotion in structural similarity index measurement over the current state-of-the-art MAR methods. All results indicate that the proposed b-MAR method can remove artifacts caused by various metal morphologies and restore the structural integrity of dental tissues effectively. Significance. The proposed b-MAR method strengthens the joint learning of the artifact removal process and the artifact generation process by bidirectionally embedding artifact representations, thereby improving the model’s artifact removal performance. Compared with other comparison methods, b-MAR can robustly and effectively correct metal artifacts in dental CBCT images caused by different dental metals.

Artificial intelligence methods in dental radiography

Introduction: The application of artificial intelligence in dentistry can significantly improve the outcome of patient treatment and reduce human error. Computer systems and certain methods that use complex algorithms are able to encode digital data about the radiographic image, obtained by X-ray radiation, and transfer them to a computer language for processing a large set of data. Applied as an auxiliary tool, it can give information about the patient and diagnostics of teeth and surrounding dental structures on radiographic images (two-dimensional and three-dimensional). The aim of the paper: The aim of the paper is to present the development and application of artificial intelligence in dentistry and at the same time emphasizing the importance of segmentation methods and methods of improving the quality of dental radiographic images. Discussion: The common goal of the development and application of artificial intelligence methods in dental radiography is directed on obtaining radiograms from which diagnostically valuable information will be collected faster and easier for the purpose of successful treatment of the patients. Methods of segmentation of a dental structures have been developed to automatically number and localize the tooth or independently mark and display the desired structures on the image. Methods of image quality enhancement tend to work on, for example, improving image resolution or reducing metal artifacts. Research has shown that artificial intelligence can be applied in everyday clinical practice, but it has certain limitations. Conclusion: Artificial intelligence methods in dental radiography are an useful auxiliary tool in clinical dental practice because they can speed up the process of retrieving patient data. Automatic localization of dental structure reduces the dentist's time for manual image analysis, but cannot completely replace human knowledge. For the full implementation of the mentioned methods, more development standards and resources are needed to overcome the ethical and legal problems of replacing humans with computer systems.