Dental Radiographs Research Articles

Etiology, clinical presentation, and outcome of temporomandibular joint luxation in cats: 21 cases (2000-2018).

This study was performed to report etiology, clinical presentation, treatment, and outcome of temporomandibular joint (TMJ) luxation in 21 cats (2000-2018). TMJ luxation was diagnosed by either dental radiography or computed tomography (CT). Causes for TMJ luxation included altercation with a dog (38%), hit-by-car (19%), unknown (19%), fall (19%), and ran into inanimate object (5%). The most common complication was development of permanent malocclusion (24%), followed by reduced vertical mandibular range of motion (5%). Reduction of TMJ luxation was successful in 94.7% of the cases. Reduction of the TMJ luxation was significantly associated with time from injury to treatment. No other significant associations were observed between cause of injury, type of TMJ stabilization, and outcome. TMJ luxation in cats carries an excellent prognosis with early joint reduction and supportive care. Concurrent dental trauma and maxillofacial injuries are common, and all patients with head trauma must be stabilized and evaluated with a thorough examination.

Digital Subtraction Radiography in Dentistry: An Overview

Digital imaging has transformed the field of medical and dental imaging by overcoming many of the drawbacks ofconventional film-based radiographs. These alterations, however, are hidden behind a background of typical anatomical featuresand are not readily visible, making it difficult for the human eye to discern them. One image-enhancing method or tool that solvesthis issue is digital subtraction radiography, which has recently found usage in a number of dental specialties. Quantitative measuresfrom digital subtraction pictures have been validated as a result of the development of digital imaging technology. Moreover,changes made to computer software have made the clinical use of subtraction imaging more adaptable. Digital subtraction radiologyhas significantly improved the ability to find oral and maxillofacial lesions. Comparing standardised radiographs acquired duringsuccessive examination sessions is done using digital subtraction radiography. Any structures that have not changed are removed,and these regions are shown in the subtraction picture as neutral grey areas, whilst parts that have altered are shown as deeperor brighter shades of grey. Yet, there is not much analysis of this often used method in the literature. The purpose of this articleis to examine this technology, its applications in different areas of dentistry, and its potential for detecting a range of disorders.This review article highlights its basic principles, applications, technique, advantages, disadvantages and future scope of digitalsubtraction radiography in dentistry.

Changes in periapical status of root canal-treated teeth after head and neck IMRT: a retrospective study.

The objective of this retrospective study was to analyze the effect of the intensity modulated radiotherapy (IMRT) of the head and neck region on root canal-treated teeth and their periapical changes due to radiation. Patients undergoing IMRT of the head and neck region were evaluated. Different types of teeth (molars, premolars, incisors and canines) were compared. Panoramic and dental radiographs were used to assess the periapical region of root canal-treated teeth using the periapical index (PAI) before and after radiotherapy (RT) and put in relation to the radiation dose per tooth. Further parameters (patient-, therapy- and tooth-related factors) were included in this study. One hundred and twenty-four root canal-treated teeth (maxilla and mandible) of 51 patients were observed. A radiolucency in the periapical region was seen in 34.7% of the samples before IMRT and an increasing number of 46% after IMRT (p-value 0.092). Clinical or radiological signs of osteoradionecrosis could not be determined. The only statistically significant difference was detected in regions irradiated with less than 40 Gy (p-value 0.045). In regions irradiated with higher doses (>40 Gy), comparable pathologies increased in non-significant numbers. A statistically significant increase of periapical pathologies was detected in premolars of the maxilla. The observations reported here suggest that a high radiation dose during IMRT has no significant consequences on root canal-treated teeth. To compare the success of endodontic treatment before versus after IMRT, further research needs to be done.

Hard tissue density of permanent teeth as a radiometric biomarker of their quality

Background. The use of radiographic methods for examining patients with the ability to determine the density of examined tissues opens up wide opportunities for individually predicting damage to hard tissues of teeth, as well as determining risk groups for the development of their pathology at early stages, which requires a detailed study of the mineral density indicators of hard tissues of teeth of different groups within the norm, taking into account their age, sex, ethnic, and other characteristics. The aim of the study was to determine the density indicators of hard tissues of permanent teeth of different groups in the cervical area and compare them. Material and Methods. The density of hard tissues was determined by analyzing radiographs of 320 intact permanent teeth of 10 patients. Radiographs were made using a Siemens Sirona (XIOS XG Supreme, Germany) dental radiography unit with Trophy Radiologie software (version 7.0, Slovakia). The images were processed using Gendex VixWin Pro software (version 3.5, USA). The density was measured in Conditional Unit of Grayness (CUG). The results were presented as Me (25%; 75%), where Me represents the median, 25% denotes the 25th percentile (first quartile), and 75% denotes the 75th percentile (third quartile). Results. Incisors, canines, and premolars exhibited the highest density in the dentin area; the lowest density was observed in incisors and premolars for cementum and in canines for enamel. In molars, enamel had the highest density, while cementum had the lowest. Significant differences in enamel density were observed among incisors and premolars, incisors and molars, canines and premolars, canines and molars, and premolars and molars. Significant differences in dentin density were found among incisors and canines, incisors and premolars, incisors and molars, canines and premolars, and canines and molars. Significant differences in cementum density were observed among incisors and canines, incisors and premolars, incisors and molars, canines and premolars, and premolars and molars. Conclusion. The density indicators of enamel, dentin, and cementum established for groups of permanent incisors, canines, premolars, and molars are different for different tooth groups with characteristic density patterns. Keywords: enamel, dentin, cementum, incisors, canines, premolars, molars.

Radiographic Outcome of Vital Teeth Treated with Prosthodontic Crowns in Dogs: 26 Cases (2015-2017).

This retrospective study evaluated the effects of prosthodontic crown placement on tooth vitality. Prosthodontic crown placement may be indicated for vital teeth affected by attrition, abrasion, uncomplicated crown fractures, enamel defects, and enamel hypoplasia. This study evaluated 26 vital teeth in 17 patients at the time of crown placement and after 1-year following crown placement. Dental radiographs were used to determine vitality of the 26 teeth. Twenty-five teeth were found to be vital and 1 tooth was non-vital 1-year after crown placement. These results demonstrated that tooth vitality was maintained after titanium alloy crown placement to treat crown attrition, abrasion, uncomplicated crown fractures, and enamel defects.

Early diagnosis of aortic calcification through dental X-ray examination for dental pulp stones

Vascular calcification, an ectopic calcification exacerbated by aging and renal dysfunction, is closely associated with cardiovascular disease. However, early detection indicators are limited. This study focused on dental pulp stones, ectopic calcifications found in oral tissues that are easily identifiable on dental radiographs. Our investigation explored the frequency and timing of these calcifications in different locations and their relationship to aortic calcification. In cadavers, we examined the association between the frequency of dental pulp stones and aortic calcification, revealing a significant association. Notably, dental pulp stones appeared prior to aortic calcification. Using a rat model of hyperphosphatemia, we confirmed that dental pulp stones formed earlier than calcification in the aortic arch. Interestingly, there were very few instances of aortic calcification without dental pulp stones. Additionally, we conducted cell culture experiments with vascular smooth muscle cells (SMCs) and dental pulp cells (DPCs) to explore the regulatory mechanism underlying high phosphate-mediated calcification. We found that DPCs produced calcification deposits more rapidly and exhibited a stronger augmentation of osteoblast differentiation markers compared with SMCs. In conclusion, the observation of dental pulp stones through X-ray examination during dental checkups could be a valuable method for early diagnosis of aortic calcification risk.

Deep learning architectures in dental diagnostics: a systematic comparison of techniques for accurate prediction of dental disease through x-ray imaging

PurposeThe purpose of the study is to classify the radiographic images into three categories such as fillings, cavity and implant to identify dental diseases because dental disease is a very common dental health problem for all people. The detection of dental issues and the selection of the most suitable method of treatment are both determined by the results of a radiological examination. Dental x-rays provide important information about the insides of teeth and their surrounding cells, which helps dentists detect dental issues that are not immediately visible. The analysis of dental x-rays, which is typically done by dentists, is a time-consuming process that can become an error-prone technique due to the wide variations in the structure of teeth and the dentist's lack of expertise. The workload of a dental professional and the chance of misinterpretation can be decreased by the availability of such a system, which can interpret the result of an x-ray automatically.Design/methodology/approachThis study uses deep learning (DL) models to identify dental diseases in order to tackle this issue. Four different DL models, such as ResNet-101, Xception, DenseNet-201 and EfficientNet-B0, were evaluated in order to determine which one would be the most useful for the detection of dental diseases (such as fillings, cavity and implant).FindingsLoss and accuracy curves have been used to analyze the model. However, the EfficientNet-B0 model performed better compared to Xception, DenseNet-201 and ResNet-101. The accuracy, recall, F1-score and AUC values for this model were 98.91, 98.91, 98.74 and 99.98%, respectively. The accuracy rates for the Xception, ResNet-101 and DenseNet-201 are 96.74, 93.48 and 95.65%, respectively.Practical implicationsThe present study can benefit dentists from using the DL model to more accurately diagnose dental problems.Originality/valueThis study is conducted to evaluate dental diseases using Convolutional neural network (CNN) techniques to assist dentists in selecting the most effective technique for a particular clinical condition.

Interactions between X-/gamma rays and alloys used in dental braces: A study on theory and simulations

In the field of dentistry, the utilization of dental X-rays plays a pivotal role in ensuring accurate diagnoses for various dental conditions. A crucial aspect of this practice involves understanding how these X-ray emissions interact with dental braces. In the presented study, the details of how X-rays and gamma rays interact with different materials used in dental braces, namely stainless steel, nitinol, elgiloy, and beta-titanium alloys, were examined. This investigation was carried out through a combination of advanced simulation codes such as FLUKA and GEANT4, alongside theoretical calculations using the WinXCOM approach. A comprehensive analysis was conducted at fourteen distinct energy levels, ranging from 20 to 150 keV with 10 keV increments. The primary focus of this study revolves around quantifying the shielding characteristics of gamma and X rays as they traverse through these dental brace materials. To achieve this, some gamma/X-ray shielding parameters, buildup-factors, and kerma relative to air were meticulously simulated and calculated. Additionally, the energy deposits within these materials and the subsequent generation of secondary radiations are thoroughly explored. Significantly, these results highlight that elgiloy alloy demonstrates the highest attenuation of X-ray and gamma ray intensities compared to the other considered materials. This comprehensive study thus offers valuable insights into the behavior of dental braces when subjected to ionizing radiation, with potential implications for patient safety and diagnostic accuracy in dental radiology.

An Assessment of the Public's Perceptions of Radiation Exposure and Risk Associated With Dental Radiographs: A Cross-Sectional Study.

Ionizing radiation exposure is an ever-present part of the dental diagnostic process. A public concern often exists due to the misunderstanding of the stochastic effects of dental X-rays. This information can be difficult to explain to the patient since many patients are apprehensive about the subject matter. This article aims to assess the public's knowledge of radiation exposure and estimate the general concern or apprehension about these diagnostic imaging modalitiesin an effort to understand and therefore ensure patient reassurance during treatment. A questionnaire was conducted asking adults between the ages of 18 to 74 in Jeddah, Saudi Arabia questions pertaining to radiation risk. There were 105 respondents; 21.9% showed concerns towarddental imaging, while 20% were skeptical. approximately 74% of respondents believed there was a limit to the amount of radiation exposure a patient could receive for diagnostic purposes, while only eight percentcorrectly identified that there was no set limit. Only 21.9% knew that a breastfeeding mother could have dental X-rays if need be; 33.3% understoodthat ionizing radiation from an intra-oral dental X-raycaused less exposure than natural background radiation froma return flight from Jeddah to Dammam. Conclusions: Patients are not aware of ionizing radiation exposure equivalencies between different imaging modalities. A more effective approach to convey exposure risk would be relating the radiation doses to natural background radiation as comparators.

A Multi-featured Approach by Integrating Digital Hand and Dental X-Ray for Human Age Estimation

Traditionally, human bone age is estimated manually by inspecting the multiple body part X-ray images, which is extremely time-consuming and prone to error. The accuracy of the human estimate depends on the experience of the medical practitioner, and thus it suffers from intra- and inter-observer variability. Hence, efficient automatic approaches are required to determine human age with high accuracy. In this work, we propose a human age estimation technique using Deep Learning (DL) technique based on hand X-ray images combined with dental orthopantomographs (OPGs) is proposed. Here, the input X-ray image is pre-processed first using Non-Local Means (NLM) first, followed by Region of Interest (RoI) extraction. Later, color and position image augmentation are performed in order to balance the dataset. Thereafter, the salient features in the image are determined, and based on these features, human age estimation is carried out using the Deep Residual Network (DRN). Here, the DRN is trained using the Beluga whale lion optimization (BWLO) algorithm. Furthermore, the BWLO_DRN is examined for its superiority considering the model accuracy and is found to obtain value of 90.1% on hand-wrist and 89.9% OPG real time dataset, thus showing superior performance for hand-wrist images.

Coarse-to-fine prior-guided attention network for multi-structure segmentation on dental panoramic radiographs

Objective. Accurate segmentation of various anatomical structures from dental panoramic radiographs is essential for the diagnosis and treatment planning of various diseases in digital dentistry. In this paper, we propose a novel deep learning-based method for accurate and fully automatic segmentation of the maxillary sinus, mandibular condyle, mandibular nerve, alveolar bone and teeth on panoramic radiographs. Approach. A two-stage coarse-to-fine prior-guided segmentation framework is proposed to segment multiple structures on dental panoramic radiographs. In the coarse stage, a multi-label segmentation network is used to generate the coarse segmentation mask, and in the fine-tuning stage, a prior-guided attention network with an encoder-decoder architecture is proposed to precisely predict the mask of each anatomical structure. First, a prior-guided edge fusion module is incorporated into the network at the input of each convolution level of the encode path to generate edge-enhanced image feature maps. Second, a prior-guided spatial attention module is proposed to guide the network to extract relevant spatial features from foreground regions based on the combination of the prior information and the spatial attention mechanism. Finally, a prior-guided hybrid attention module is integrated at the bottleneck of the network to explore global context from both spatial and category perspectives. Main results. We evaluated the segmentation performance of our method on a testing dataset that contains 150 panoramic radiographs collected from real-world clinical scenarios. The segmentation results indicate that our proposed method achieves more accurate segmentation performance compared with state-of-the-art methods. The average Jaccard scores are 87.91%, 85.25%, 63.94%, 93.46% and 88.96% for the maxillary sinus, mandibular condyle, mandibular nerve, alveolar bone and teeth, respectively. Significance. The proposed method was able to accurately segment multiple structures on panoramic radiographs. This method has the potential to be part of the process of automatic pathology diagnosis from dental panoramic radiographs.

Occurrence of radiopaque and mixed lesions at periapical region in patients with rheumatoid arthritis and ankylosing spondylitis: a retrospective study

BackgroundRheumatoid arthritis (RA) and ankylosing spondylitis (AS) have different effects on bones, cartilage and joints, sometimes destroying the spine and joints, and other times causing new bone formation. This study aimed to evaluate the effects of RA and AS on the types (radiolucent, radiopaque and mixed) of periapical lesions in jaw bones.MethodsThis study included 708 individuals (97 with AS, 327 with RA and 284 healthy controls (C)) and a total of 17,118 teeth (AS: 2,442; RA: 7,638; C: 7,038). The number of teeth, extracted teeth and teeth with root canal treatment and the presence of radiopaque, radiolucent and mixed periapical lesions were recorded from dental panoramic radiographs. Kruskal–Wallis and chi-square tests were used for statistical analysis.ResultsThe frequency of radiopaque lesions in the AS and RA groups was similar (p > 0.05) and significantly higher than in the C group (p < 0.05) (AS: 13.4%; RA: 6.1%; C: 2%). Mixed lesions (AS: 3.1%; RA: 4.0%; C: 0.4%) were statistically significantly higher for the RA group compared to the C group (p < 0.05), while the AS–C and AS–RA groups were similar (p > 0.05). There was no significant difference in terms of radiolucent lesions among groups (p > 0.05).ConclusionRadiopaque apical lesions were frequent in RA and AS patients, while mixed lesions were significantly higher in RA patients.

Panoramic dental X-ray exposure leads to oxidative stress, inflammation and apoptosis-mediated developmental defects in zebrafish embryos

Panoramic x-ray units are widely used in dental radiodiagnostics. Patients are exposed to relatively low radiation doses with panoramic imaging, but considering lifetime frequency of exposure, even a small risk can have serious health consequences. Our aim was to assess the effects of panoramic x-rays at two different exposure times on developing zebrafish embryos, focusing on oxidative stress, inflammation, apoptotic pathways, and development. Zebrafish embryos were divided into three groups: control, standard panoramic (SPE, 5.5 s exposure time) and pedodontic panoramic x-ray group (PPE, 4.8 s exposure time). Optically stimulated luminescence dosimeters were used to measure absorbed doses. Mean radiation doses for SPE and PPE were 7.83 mSv and 5.83 mSv respectively. At the end of 96 h post-fertilization, lipid peroxidation (LPO), nitric oxide (NO), reduced glutathione (GSH), glutathione S-transferase and superoxide dismutase were measured in the embryos. Expressions of genes related with inflammation (tnfα, il6, ill15, il21), immunoregulation (ifng) and apoptosis (p53, bax, casp2, casp3, casp8) were determined by RT-PCR. Even at reduced doses at high-speed mode, developmental toxicity was observed in both groups as evidenced by decreased pigmentation, yolk sac oedema, and spinal curvature. While deterioration of oxidant-antioxidant balance, suppression of immune response, induction of inflammation and apoptosis were observed through increased LPO, NO, decreased GSH, ifng, and increased expressions of genes related with inflammation and apoptosis, these effects were more pronounced in the SPE group. These results demonstrate the influence of exposure time and indicate the need for further consideration of optimal panoramic modes from a radiation-induced damage perspective.

Single Crown Restorations Supported by One-Piece Zirconia Dental Implants: Case Series with a Mean Follow-Up of 58 Months

The main aim of this case series was to report the clinical and radiographic outcomes of 22 one-piece zirconia dental implants positioned in 19 patients to restore single edentulisms and followed up for at least 2 years. The mean observation period was 58.18 months. At the last follow-up visit, no issues, such as foreign body sensation, discomfort or pain, were reported by the patients. No implant showed signs of infection with suppuration or implant mobility. Marginal bone levels (MBL) were assessed through standardized dental radiographs and a dedicated software. The mean distance between the implant head and the first detectable bone to implant contact was calculated at the mesial and distal aspect of each implant. The mean MBL at baseline was 1.82 ± 0.63 mm while the mean MBL at the last follow-up visit was 2.57 ± 0.72 mm. The results obtained in the present case series over a mean follow-up period of 58.18 months (range 27–96) showed that one-piece zirconia dental implants could be an alternative option to support single crowns in patients requiring metal-free restorations. Nevertheless, further better-controlled research featuring an adequate study design and longer follow-up is needed in order to clarify advantages and limitations which are related to this treatment modality.

Assessment of Panoramic Radiograph Errors: An Evaluation of Patient Preparation and Positioning Quality at Soelastri Dental and Oral Hospital

Background: Dental radiography is an important component of comprehensive patient care. One of the most frequently used extraoral radiographic techniques is panoramic radiography. Panoramic radiographs lose value when the quality of the diagnostic radiographs is poor. In their application, there are still many errors that occur in panoramic radiographs. Objective: This study aimed to determine the quality and frequency of panoramic radiograph errors due to patient preparation and positioning errors at the Soelastri Dental and Oral Hospital, Universitas Muhammadiyah Surakarta (RSGM Soelastri UMS), Indonesia. Methods: This type of research is descriptive, quantitative, observational, and takes a retrospective approach. A total of 312 panoramic radiographs of patients at RSGM Soelastri UMS from January to December 2021 who met the inclusion and exclusion criteria were taken at random stages and evaluated. Radiographs were compared with ideal quality, given a National Radiological Protection Board (NRPB) rating, and the frequency of errors that occurred was evaluated. Results: The most common rating obtained on panoramic radiographs in this study was 2 of 173 (55.45%). The frequency of preparation errors was 1.22%, while the frequency of patient position errors was 98.78%. The highest patient position error was not placing the tongue on the palate, which was 49.68%. Conclusion: The quality of panoramic radiographs at RSGM Soelastri UMS in the period from January–December 2021 generally has a rating of 2, according to the NRPB, and there are still errors in patient preparation and positioning.

Convolutional-neural-network-based radiographs evaluation assisting in early diagnosis of the periodontal bone loss via periapical radiograph

Background/PurposeThe preciseness of detecting periodontal bone loss is examiners dependent, and this leads to low reliability. The need for automated assistance systems on dental radiographic images has been increased. To the best of our knowledge, no studies have quantitatively and automatically staged periodontitis using dental periapical radiographs. The purpose of this study was to evaluate periodontal bone loss and periodontitis stage on dental periapical radiographs using deep convolutional neural networks (CNNs). Materials and methods336 periapical radiographic images (teeth: 390) between January 2017 and December 2019 were collected and de-identified. All periapical radiographic image datasets were divided into training dataset (n = 82, teeth: 123) and test dataset (n = 336, teeth: 390). For creating an optimal deep CNN algorithm model, the training datasets were directly used for the segmentation and individual tooth detection. To evaluate the diagnostic power, we calculated the degree of alveolar bone loss deviation between our proposed method and ground truth, the Pearson correlation coefficients (PCC), and the diagnostic accuracy of the proposed method in the test datasets. ResultsThe periodontal bone loss degree deviation between our proposed method and the ground truth drawn by the three periodontists was 6.5 %. In addition, the overall PCC value of our proposed system and the periodontists’ diagnoses was 0.828 (P < 0.01). The total diagnostic accuracy of our proposed method was 72.8 %. The diagnostic accuracy was highest for stage III (97.0 %). ConclusionThis tool helps with diagnosis and prevents omission, and this may be especially helpful for inexperienced younger doctors and doctors in underdeveloped countries. It could also dramatically reduce the workload of clinicians and timely access to periodontist care for people requiring advanced periodontal treatment.

Insights on the digitalisation of dental practices: A cross-sectional pilot study in Hesse.

Even though today, many fields in dentistry allow digital processes, analogue procedures are still widely used. This cross-sectional pilot study aimed to survey insights on the digitalisation of dental practices using the example of Hesse. Between April and June 2022, 4840 active practicing dentists registered by the State Dental Association of Hesse were invited via e-mail to fill out an online questionnaire regarding their technical requirements in dental practice, dental treatment procedures, and attitude towards digitalisation in dentistry. Demographic questions were asked. Besides descriptive statistics, correlations were analyzed (P < 0.05). Questionnaires of 937 dentists (279 female, 410 male, four inter/divers, 244 no answer; mean age of 51.4 ± 10.4 years) were examined representing a respond rate of 19.36%. In the area of practice administration and dental radiography, the majority of the dentists surveyed is already working digitally, which is predominantly assessed as a positive development. Already one third of the respondents state that they use an intraoral scanner for dental treatments, but the indication is mainly limited to smaller restorations. However, many dentists rate the use of social media accounts and telemedicine rather negative. Within the limitation of this pilot study, many processes especially in dental treatments are still analogue. However, 60% of the participants plan digitalisation of their dental practices within the next five years, which indicates a clear shift from analogue to digital dentistry.

Diagnostic Study of Mandibular Cortical Index Classification Using Dental Cone-Beam Computed Tomography Findings: A Preliminary Cross-Sectional Study

The prevalence of osteoporosis is high, reportedly affecting 200 million people worldwide. A major problem associated with osteoporosis is that femoral fractures cause a decline in general function and loss of independence, greatly reducing patients’ quality of life. Notably, osteoporosis is an asymptomatic chronic metabolic disease, and its detection is thus often delayed. Interestingly, mandibular cortical index (MCI) classification using dental panoramic radiography is reportedly useful for early detection of osteoporosis. However, this visual classification method can lead to differences in diagnoses among surgeons. The aim of this preliminary study was to analyze cone-beam computed tomography (CBCT) data and examine an objective MCI classification using the findings obtained. MCI classification (classified as C1, C2, or C3) was performed by three examiners on 70 women (91 sites) aged ≥20 years. The mandibular cortical width of all sites was measured using CBCT images. The results showed that the mandibular cortical width was not particularly correlated with age or number of present teeth, and no significant quantitative differences were found between C1 and C2. However, coronal CBCT images of C2 revealed multiple characteristic trabecular bone structures. These structures may be an important finding affecting the classification of two-dimensional dental panoramic radiography images.

Pseudo labelling workflow, margin losses, hard triplet mining, and PENViT backbone for explainable age and biological gender estimation using dental panoramic radiographs

This study aimed to estimate human age and gender from panoramic radiographs using various deep learning techniques while using explainability to have a novel hybrid unsupervised model explain the decision-making process. The classification task involved training neural networks and vision transformers on 706 panoramic radiographs using different loss functions and backbone architectures namely ArcFace, a triplet network named TriplePENViT, and the subsequently developed model called PENViT. Pseudo labeling techniques were applied to train the models using unlabeled data. FullGrad Explainable AI was used to gain insights into the decision-making process of the developed PENViT model. The ViT Large 32 model achieved a validation accuracy of 68.21% without ArcFace, demonstrating its effectiveness in the classification task. The PENViT model outperformed other backbones, achieving the same validation accuracy without ArcFace and an improved accuracy of 70.54% with ArcFace. The TriplePENViT model achieved a validation accuracy of 67.44% using hard triplet mining techniques. Pseudo labeling techniques yielded poor performance, with a validation accuracy of 64.34%. Validation accuracy without ArcFace was established at 67.44% for Age and 84.49% for gender. The unsupervised model considered developing tooth buds, tooth proximity and mandibular shape for estimating age within deciduous and mixed dentitions. For ages 20–29, it factored permanent dentition, alveolar bone density, root apices, and third molars. Above 30, it notes occlusal deformity resulting from missing dentition and the temporomandibular joint complex as predictors for age estimation from panoramic radiographs.Graphical abstract

Teeth and prostheses detection in dental panoramic X-rays using CNN-based object detector and a priori knowledge-based algorithm

Deep learning techniques for automatically detecting teeth in dental X-rays have gained popularity, providing valuable assistance to healthcare professionals. However, teeth detection in X-ray images is often hindered by alterations in tooth appearance caused by dental prostheses. To address this challenge, our paper proposes a novel method for teeth detection and numbering in dental panoramic X-rays, leveraging two separate CNN-based object detectors, namely YOLOv7, for detecting teeth and prostheses, alongside an optimization algorithm to refine the outcomes. The study utilizes a dataset of 3138 radiographs, of which 2553 images contain prostheses, to build a robust model. The tooth and prosthesis detection algorithms perform excellently, achieving mean average precisions of 0.982 and 0.983, respectively. Additionally, the trained tooth detection model is verified using an external dataset, and six-fold cross-validation is conducted to demonstrate the proposed method’s feasibility and robustness. Moreover, the investigation of performance improvement resulting from the inclusion of prosthesis information in the teeth detection process reveals a marginal increase in the average F1-score, rising from 0.985 to 0.987 compared to the sole teeth detection method. The proposed method is unique in its approach to numbering teeth as it incorporates prosthesis information and considers complete restorations such as dental implants and dentures of fixed bridges during the teeth enumeration process, which follows the universal tooth numbering system. These advancements hold promise for automating dental charting processes.