Intraoral Imaging Research Articles

Influence of zirconia crown artifacts on cone beam computed tomography scans and image superimposition of tomographic image and tooth surface scan: An in vitro study

Statement of problemZirconia restorations create significant artifacts on 3D cone beam computed tomography (CBCT) imaging. Static computer-assisted implant surgery (s-CAIS) relies on the accuracy of superimposition between an intraoral surface scan and CBCT imaging. However, how the artifacts from zirconia on the tomographic image might affect the predictability of s-CAIS is unclear. PurposeThe purpose of this in vitro study was to evaluate the effect of zirconia crown restorations on the superimposition process for s-CAIS. Material and methodsFour stone casts generated 4 groups: a control group (CG) with no crowns and 3 experimental groups with 4 (TG4), 7 (TG7), and 13 (TG13) zirconia crowns. A total of 40 CBCT scans were made for the 4 groups (n=10). All CBCTs were imported into a computer planning software program, and the casts from all 4 groups were scanned by using a high-resolution laboratory scanner. The standard tessellation language (STL) files were imported, segmented, and the 3 files superimposed for all groups. The accuracy of the superimposition was assessed, in millimeters, in 3 planes corresponding to anterior-posterior, horizontal, and vertical, as well as the overall measurement, and the results were analyzed statistically (α=.05). ResultsThe overall analysis demonstrated statistically significant differences between all groups (P<.001), except between CG and TG4. The anterior-posterior dimension demonstrated significant differences between CG and TG7 (P<.001), CG and TG13 (P<.001), TG4 and TG7 (P=.004), and TG4 and TG13 (P=.001). For the vertical dimension analysis, significant differences were found between CG and TG7 (P=.001), CG and TG13 (P<.001), and TG4 and TG13 (P<.001). For the horizontal variable, statistically significant differences were found between CG and TG7 (P=.049), CG and TG13 (P<.001), TG4 and TG13 (P<.001), and TG7 and TG13 (P=.003). ConclusionsThe accuracy of the superimposition of the images was influenced by the number of zirconia crowns, with an increased number reducing the superimposition accuracy.

Accuracy of intraoral scan images in full arch with orthodontic brackets: a retrospective in vivo study

ObjectivesThe purpose of this retrospective study was to evaluate the accuracy of intraoral scan (IOS) images in the maxillary and mandibular arches with orthodontic brackets.Material and methodsFrom digital impressions of 140 patients who underwent orthodontic treatment, consecutive IOS images were selected based on standardized inclusion criteria: Two pre-orthodontic IOS images (IOS1 and IOS2) of permanent dentition with fully erupted second molars and IOS images obtained immediately after orthodontic bracket bonding (IOSb). Superimpositions were performed to evaluate the reproducibility of repeated IOS images. Accuracy of IOSb images was analyzed by comparing the average surface errors between IOS1c and IOS2c images, which were IOS images cut based on the same region of the interest as between IOS1 and IOSb images.ResultsA total of 84 IOS images was analyzed. The average surface errors between IOS1 and IOS2 images were 57 ± 8 μm and 59 ± 14 μm in the maxillary and mandibular arch, respectively, and their reliability was almost perfect. The average errors between IOSb and IOS1c images exhibited an increase, which measured 97 ± 28 μm in the maxillary arch and 95 ± 29 μm in the mandibular arch. These surface deviations between IOSb and IOS1c images were significantly larger in each region as well as entire dentition (P < 0.001) compared to those between IOS1c and IOS2c images.ConclusionsThe average surface errors of the scans with brackets showed increased values compared with those without brackets. This suggests that orthodontic brackets could affect the trueness of intraoral scan images.Clinical relevanceIt is necessary for clinicians to consider the effect of brackets on digital impression when using IOS images in orthodontic patients.

Deep learning convolutional neural network algorithms for the early detection and diagnosis of dental caries on periapical radiographs: A systematic review.

PurposeThe aim of this study was to analyse and review deep learning convolutional neural networks for detecting and diagnosing early-stage dental caries on periapical radiographs.Materials and MethodsIn order to conduct this review, the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines were followed. Studies published from 2015 to 2021 under the keywords (deep convolutional neural network) AND (caries), (deep learning caries) AND (convolutional neural network) AND (caries) were systematically reviewed.ResultsWhen dental caries is improperly diagnosed, the lesion may eventually invade the enamel, dentin, and pulp tissue, leading to loss of tooth function. Rapid and precise detection and diagnosis are vital for implementing appropriate prevention and treatment of dental caries. Radiography and intraoral images are considered to play a vital role in detecting dental caries; nevertheless, studies have shown that 20% of suspicious areas are mistakenly diagnosed as dental caries using this technique; hence, diagnosis via radiography alone without an objective assessment is inaccurate. Identifying caries with a deep convolutional neural network-based detector enables the operator to distinguish changes in the location and morphological features of dental caries lesions. Deep learning algorithms have broader and more profound layers and are continually being developed, remarkably enhancing their precision in detecting and segmenting objects.ConclusionClinical applications of deep learning convolutional neural networks in the dental field have shown significant accuracy in detecting and diagnosing dental caries, and these models hold promise in supporting dental practitioners to improve patient outcomes.

Use of intraoral three-dimensional images for the identification of dental morphological traits related to ancestry estimation

Victim identification through dental features is one of the main objectives of forensic dentistry. In circumstances where information regarding antemortem dental records is missing, reconstruction of a biological profile can be useful as a first step toward personal identification. This reconstructive method provides valuable information, namely regarding the individual's ancestry, through the detection and degree of expression of dental morphological traits, which may help to restrict the number of candidates for identification. Technological advances allowed the development of alternative methods for dental evaluation, that complement or substitute those already in use in clinical practice. Among these, intraoral three-dimensional (3D) images are increasingly used in dentistry, as they have a high level of accuracy and are easy to obtain and store. However, a fundamental question regarding forensic dentistry is whether they allow recognition and analysis of dental morphological traits in detail, namely those related to ancestry. In this study, we evaluated 20 teeth morphological features using intraoral 3D imaging from 77 individuals from Northern Portugal. Our results showed that it was possible to identify and classify a large part of the main morphological traits used in the estimation of ancestry. As these 3D images present sufficient morphological detail to be classified, we believe that future applications of this technique can be expected in forensic dentistry.

Classification of Intraoral X-ray Images Using Artificial Intelligence

Classification of Intraoral X-ray Images Using Artificial Intelligence

A Novel Phantom and a Dedicated Developed Software for Image Quality Controls in X-Ray Intraoral Devices.

Background: Periodic quality control (QC) procedures are important in order to guarantee the image quality of radiological equipment and are also conducted using phantoms simulating human body. Objective: To perform (QC) measurements in intraoral imaging devices, a new and simple phantom was manufactured. Besides, to simplify QC procedures, computerized LabView-based software has been devised, enabling determination of image quantitative parameters in real time or during post processing.Material and Methods: In this experimental study, the novel developed phantom consists of a Polymethyl methacrylate (PMMA) circular insert. It is able to perform a complete QC image program of X-ray intraoral equipment and also causes the evaluation of image uniformity, high and low contrast spatial resolution, image linearity and artefacts, with only two exposures. Results: Three raters analyzed the images using the LabView dedicated software and determined the quantitative and qualitative parameters in an innovative and accurate way. Statistical analysis evaluated the reliability of this study. Good accuracy of the quantitative and qualitative measurements for the different intraoral systems was obtained and no statistical differences were found using the inter-rater analysis. Conclusion: The achieved results and the related statistical analysis showed the validity of this methodology, which could be proposed as an alternative to the commonly adopted procedures, and suggested that the novel phantom, coupled with the LabView based software, could be considered as an effective tool to carry out a QC image program in a reproducible manner.

Fabrication of a 3-dimensionally printed definitive cast for an obturator prosthesis by merging intraoral scan image with cone beam computed tomography data: A clinical report

This clinical report focuses on the production of a 3-dimensionally printed definitive cast by using digital scans to fabricate an obturator prosthesis for a completely edentulous patient with a maxillary defect. A merged intraoral scanner image of the soft tissue anatomy of maxillary structures with volumetric data of the craniofacial hard and soft tissues obtained from a cone beam computed tomography scan was used to generate a virtual cast, producing the polyurethane definitive cast manufactured by using a 3-dimensional printer.

Does Caries in the Adjacent Tooth Justify Prophylactic Odontectomy of Impacted Tooth?

Purpose: To ascertain the influence of impacted lower wisdom tooth on the initiation and progression of distal caries in the lower second molar and whether prophylactic removal of lower wisdom tooth is justified. Materials and Methods: A retrospective study was conducted based on the records of the patients who underwent surgical removal of lower wisdom tooth in 3 dental clinics of the different cities during December 2015 to August 2019. The orthopantomogram and intraoral periapical radiograph images of 470 patients were considered to evaluate the impact of impacted lower wisdom tooth on the initiation and progression of distal caries in the lower second molar and whether prophylactic removal of lower wisdom tooth is justified. In order to achieve this, parameters evaluated on the radiographs are the presence of dental caries in relation to lower second molar, angulation of the lower wisdom tooth, depth of impaction, presence of pericoronitis, and patient characteristics like age and gender. The results were analyzed to ascertain whether prophylactic removal of lower wisdom tooth is justified to preserve the long term health of the lower second molar. Results: The results of this study reveal that 28% of the patients developed dental caries in relation to lower second molar in the presence of a wisdom tooth. The mandibular third molars were impacted unilaterally in 63% of cases and bilaterally in 37% of cases. Caries in lower second molars was noticed when the wisdom tooth was in Class I position in 53.8% of the cases and 69% in position A. This finding is commonly encountered in the presence of a mesioangular impacted third molar. It was observed that when a wisdom tooth was in Class I relationship for any kind of angulation it increased the chances for the initiation and progression of caries in lower second molars. Conclusion: It can be concluded that to preserve the lower second molars, it is advisable to prophylactically remove an impacted third molar that have a mesial angulation and Class I and Level A position.

The Effect of Submucosal Injection of Plasma Rich Platelets on the Rate of Orthodontically Induced Canine Retraction in Subject with Bimaxillary Protrusion

Objective: The present study aimed to reveal the effects of submucosal injection of Plasma Rich Platelets (PRP) on the rate of canine retraction. &#x0D; Methods: Eighteen females with bimaxillary protrusion were selected from patients seeking orthodontic treatment, College of the Dentistry/University of Sulaimani, whose maxillary and mandibular first premolars were decided to be extracted after proper diagnosis. It’s a split-mouth design; the upper left side was the control side while the upper right side served as the intervention side (submucosal injection of PRP); after aligning and leveling, the retraction phase was initiated on .017× .025 Stainless steel archwire with power chain from the canine bracket to temporary anchorage device inserted between the upper 2nd premolar and 1st molar at the same level for both sides. The elastomeric chains were changed every two weeks. Scanned intraoral images were obtained by intra- oral CEREC omnicam scanner before retraction and at the end of retraction to measure the amount of canine movement using inLab CAM 15.0 software 2015.&#x0D; Results: A highly significant acceleration of canine retraction on the intervention side compared with the control side at p &lt;.0001 with a rate of 29.1% higher overall retraction phase (108 days). &#x0D; Conclusions: Submucosal injection of PRP is a minimally-invasive and low-cost method that can be used for accelerating orthodontic tooth movement.

Mussel-Inspired Adhesive Double-Network Hydrogel for Intraoral Ultrasound Imaging.

Periodontal diseases could be diagnosed through intraoral ultrasound imaging with the advantages of simple operation procedures, low cost, and low safety risks. A couplant is normally placed between transducers and tissues for better ultrasound image quality. If applied intraorally, the couplants should possess good stability in water and robust mechanical properties, as well as strong adhesiveness to transducers and tissues. However, commercial couplants, such as Aquaflex (AF) cannot fulfill these requirements. In this work, inspired by the mussel adhesion mechanism, we reported a poly(vinyl alcohol)-polyacrylamide-polydopamine (PVA-PAM-PDA) hydrogel synthesized by incorporating PDA into the PAM-PVA double-network for intraoral ultrasound imaging. The hydrogel maintains good stability in water as well as exceptional mechanical properties and can adhere to different substrates (i.e., metal, glass, and porcine skin) without losing the original adhesion strength after multiple adhesion-strip cycles. Besides, when applied to porcine mandibular incisor imaging, the PVA-PAM-PDA hydrogel possesses good image quality for diagnosis as AF does. This work provides practical insights into the fabrication of multifunctional hydrogel-based interfaces between human tissues and medical devices for disease diagnosis applications.

An artificial intelligence proposal to automatic teeth detection and numbering in dental bite-wing radiographs

Objectives Radiological examination has an important place in dental practice, and it is frequently used in intraoral imaging. The correct numbering of teeth on radiographs is a routine practice that takes time for the dentist. This study aimed to propose an automatic detection system for the numbering of teeth in bitewing images using a faster Region-based Convolutional Neural Networks (R-CNN) method. Methods The study included 1125 bite-wing radiographs of patients who attended the Faculty of Dentistry of Ordu University from 2018 to 2019. A faster R-CNN an advanced object identification method was used to identify the teeth. The confusion matrix was used as a metric and to evaluate the success of the model. Results The deep CNN system (CranioCatch, Eskisehir, Turkey) was used to detect and number teeth in bitewing radiographs. Of 715 teeth in 109 bite-wing images, 697 were correctly numbered in the test data set. The F1 score, precision and sensitivity were 0.9515, 0.9293 and 0.9748, respectively. Conclusions A CNN approach for the analysis of bitewing images shows promise for detecting and numbering teeth. This method can save dentists time by automatically preparing dental charts.

The analysis of the subjective image quality of a dental radiogram obtained with a digital and film detector

Introduction: Today film-based technologies are largely abandoned in dental radiology. New technologies based on digital detectors are being adopted. This change is part of the global digitalization of radiological procedures. A person operating the radiological device needs to be educated and well informed about radiological procedures and technologies, as well as to know how the system operates and to be sure about the dose required for the adequate image quality. The awareness of radiation doses received by patients is necessary for the estimation of risk from ionizing radiation.Material and methods: The aim of this study was to evaluate whether patient doses in intraoral dental radiology affect image quality. The study included subjective analysis of radiological image quality for an intraoral x-ray device with possibility to use both digital and film detector. Subjective assessment of image quality is performed according to criteria taken from the literature and is expressed using the Likert scale, grade 1–5. In order to improve the quality of inferential statistics related ratings to detector quality and image quality were collected, thus introducing two variables: the detector quality index and the technique quality index.Results: Z-test proportions column shows that the number of grades four given as a contrast score is significantly higher in the RVG detector (65%) compared to the film (40%). There is also a significant difference in the number of given grades five as a rating of the apex preview. The share of grades five in film is 19%, while in RVG it is 42%. It is similar in the evaluation of the preview of the dental canal, where there is also a significant difference in the number of given grades five, so that the share in the film is 11% and in RVG 35%.Conclusion: Analysis of the subjective quality of the radiological image in devices for intraoral radiography with the possibility of recording using digital and film detectors indicates that the digital detector is superior to film in low contrast resolution, apex and dental canal preview.

Template-guided endodontic access

Background and OverviewEndodontics is a rewarding and challenging profession. The tasks of negotiating, cleaning, and shaping canals can frustrate even the most talented clinicians, particularly if the canals are blocked by calcifications. Circumventing these obstructions can be a slow, tedious process, and at times it is not possible. Various techniques have been developed to assist with this problem but have met with limited success. Given that the success of endodontic treatment is dependent on the elimination or reduction of bacterial influences on periapical tissues, this is a critical issue. Case DescriptionA new application of an existing technology is available. This technique uses 3-dimensional (3D) radiography (cone-beam computed tomography) and 3D intraoral imaging. Through the integration of the data sets, pathways are planned to locate blocked canals. Templates can be designed and fabricated on 3D printers that fit intimately on the dentition. Special drills are selected that fit through the sleeves embedded in the template. These pathways align the drill and guide the depth of the access. Once the canals are accessed, conventional root canal treatment is possible. Conclusions and Practical ImplicationsThis case report reviews a new technique for treating a necrotic calcified root canal. The canal was identified in the apical one-third of the root. A pathway was planned to negotiate the canal using digital technology. The blockage was bypassed, and the root canal was successfully treated using a “template-guided access technique” with minimal tooth loss.

A Deep Learning-Based Approach for the Detection of Early Signs of Gingivitis in Orthodontic Patients Using Faster Region-Based Convolutional Neural Networks.

Computer-based technologies play a central role in the dentistry field, as they present many methods for diagnosing and detecting various diseases, such as periodontitis. The current study aimed to develop and evaluate the state-of-the-art object detection and recognition techniques and deep learning algorithms for the automatic detection of periodontal disease in orthodontic patients using intraoral images. In this study, a total of 134 intraoral images were divided into a training dataset (n = 107 [80%]) and a test dataset (n = 27 [20%]). Two Faster Region-based Convolutional Neural Network (R-CNN) models using ResNet-50 Convolutional Neural Network (CNN) were developed. The first model detects the teeth to locate the region of interest (ROI), while the second model detects gingival inflammation. The detection accuracy, precision, recall, and mean average precision (mAP) were calculated to verify the significance of the proposed model. The teeth detection model achieved an accuracy, precision, recall, and mAP of 100 %, 100%, 51.85%, and 100%, respectively. The inflammation detection model achieved an accuracy, precision, recall, and mAP of 77.12%, 88.02%, 41.75%, and 68.19%, respectively. This study proved the viability of deep learning models for the detection and diagnosis of gingivitis in intraoral images. Hence, this highlights its potential usability in the field of dentistry and aiding in reducing the severity of periodontal disease globally through preemptive non-invasive diagnosis.

Diagnosis of Occlusal Tooth Wear Using 3D Imaging of Optical Coherence Tomography Ex Vivo.

The aim of this study was to assess the utility of 3D imaging of optical coherence tomography (OCT) for the diagnosis of occlusal tooth wear ex vivo. Sixty-three extracted human molars with or without visible tooth wear were collected to take digital intraoral radiography and 3D OCT images. The degree of tooth wear was evaluated by 12 examiners and scored using 4-rank scale: 1—slight enamel wear; 2—distinct enamel wear; 3—tooth wear with slight dentin exposure; 4—tooth wear with distinct involvement of dentin. The degree of tooth wear was validated by the histological view of confocal laser scanning microscopy (CLSM). The sensitivity, specificity, and area under the curve (AUC) of receiver operating characteristic analysis were calculated. Diagnostic accuracy was compared with the agreement with CLSM observation using weighted kappa. The results were statistically analyzed at a significance level of α = 0.05. Three-dimensional OCT showed significantly higher sensitivity (p < 0.05) for all the diagnostic thresholds of enamel wear and dentin exposure than digital radiography (0.82, 0.85, and 0.79 vs. 0.56, 0.52, and 0.57, respectively). Three-dimensional OCT showed higher AUC and kappa coefficients than digital radiography (p < 0.05), where mean AUC and Kappa values were 0.95 and 0.76 for OCT and 0.92 and 0.47 for radiography, respectively. No significant difference of specificity was observed (p > 0.05). Three-dimensional OCT could visualize and estimate the degree of tooth wear and detect the dentin exposure at the tooth wear surface accurately and reproducibly. Consequently, a new guideline for tooth wear assessment can be proposed using OCT.

Awareness of intraoral photography among dental students

Precise shade matching is one of the most testing parts of dental restorations and stylish dentistry. Because of the incredible assortment of regular tooth shading accomplishing a nearby shade match of a fake rebuilding with normal dentition is a mind-boggling process. Photography has been utilized for a long time trying to improve correspondence among dental specialists and dental professionals. The aim of the study was to evaluate the awareness of intra oral photography among dental graduates. This was a questionnaire based cross-sectional type of study comprising 100 dental college students in Chennai. A self-designed questionnaire contains 10 questions based on the knowledge, awareness on intra oral photography among dental college students. Questionnaires were circulated through an online website survey planet. After the responses were received from 100 participants, data were collected and analysed.76% of dental graduates stated the intraoral photographs were taken in automatic mode and only 24% of dental graduates set the camera in manual mode to capture the intraoral image. 97% of Dental graduates use the grey card as the contrast to the subject taken, and 98% of dental graduates are unaware about the use of white balance in-camera setting, and 1% of dental graduates are only aware of the 18% grey card and the use of grey card to pre-set the white balance. 100% of the people recommend the intraoral photography courses included in the dental course. 33% of the graduates have attended the photography courses to learn the technique to capture the intraoral photographs. The awareness of intraoral photography among dental students is moderately adequate. More than 30 of the graduates have attended the photography courses to learn the technique to capture the intraoral photographs. The documentation of the clinical cases is being encouraged in dental schools, but the proper training for the intraoral photographs is not being met. 100% of graduates recommend the photography courses be included in the dental curriculum.

Comparison of intraoral digital radiography and cone-beam computed tomography in the measurement of periodontal bone defects.

Periodontal disease has a high prevalence in many countries. Thus, the early detection of periodontal disease is important in order to obtain the most appropriate treatment plan to prevent tooth loss, and subsequently, to maintain the patient's general health. The aim of this study was to compare the accuracy of cone-beam computed tomography (CBCT) and intraoral parallel digital radiography in measuring the dimensions of periodontal bone defects. In this in vitro study, 236 periodontal bone defects were artificially created in dry human mandibles using a burr. Defects included horizontal, one-, two-, and three-wall defects, craters, dehiscences, and fenestrations. Intraoral digital radiographs were obtained using the parallel technique with photostimulable phosphor plates (PSP) and CBCT scans were performed. Two calibrated observers evaluated the images and measured the dimensions of the defects. Clinical probing was performed and considered as the gold standard. The measurements of digital radiography and CBCT were compared to those achieved by probing to evaluate their accuracy. Cone-beam computed tomography had a significantly stronger correlation with the gold standard than intraoral parallel digital imaging. In the total assessment of the periodontal defects, the intraclass correlation coefficient (ICC) was calculated at 0.93 for CBCT-probe and at 0.78 for PSP-probe (p < 0.05). The accuracy of CBCT was superior to that of intraoral digital radiography for measuring horizontal, one-, two-, and three-wall defects, craters, dehiscences, and fenestrations.

Intraoral radiographs: A comparison of dose and risk reduction with collimation and thyroid shielding

BackgroundThe National Commission on Radiation Protection and Measurements has reinforced its recommendation for the use of rectangular collimation for intraoral radiography in its Report No. 177 published in 2019. This study compared effective dose (E) using circular and rectangular collimator (RC) modalities. MethodsThe authors exposed 18 projections for adult and 12 projections for child full-mouth series using an original equipment 6 centimeter diameter circular collimator (circular), original equipment rectangular positioning indicator device (Focus-RC), and 5 universal RC modalities (JadRad-RC, Rinn-RC, Durr-RC, DEXshield-RC, and TruAlign-RC) for adult and child phantoms. The authors acquired dosimetry using optically stimulated luminescence dosimeters. Exposures were made with a Focus (Instrumentarium) intraoral source using 70 peak kilovoltage and total milliamperes of 5.34 (adult) and 2.7 (child). ResultsAdult E was lowest for Focus-RC (54 microsieverts), which also produced the greatest exposure area reduction (51%) compared with circular, followed by JadRad-RC (55 μSv), Durr-RC (58 μSv), Rinn-RC (62 μSv), DEXshield-RC (70 μSv), TruAlign-RC (85 μSv), and circular (86 μSv). Child E followed a similar trend: Focus-RC (44 μSv), JadRad-RC (44 μSv), Durr-RC (45 μSv), Rinn-RC (48 μSv), DEXshield-RC (53 μSv), TruAlign-RC (85 μSv), and circular (89 μSv). When used with thyroid shielding, circular collimation thyroid dose was reduced by as much as 59%. ConclusionsFocus-RC techniques yielded the greatest dose reduction compared with alternative RC and circular. In addition to shape, collimator dimensions should be considered as significant factors affecting patient E. RC alone yielded a greater reduction in thyroid dose than did circular with thyroid shielding. Practical ImplicationsThis study’s findings underscore the updated recommendations of the National Commission on Radiation Protection and Measurements Report No.177, which emphasized the benefits and important practical considerations of RC with intraoral imaging.

A deep learning algorithm for detection of oral cavity squamous cell carcinoma from photographic images: A retrospective study.

BackgroundThe overall prognosis of oral cancer remains poor because over half of patients are diagnosed at advanced-stages. Previously reported screening and earlier detection methods for oral cancer still largely rely on health workers’ clinical experience and as yet there is no established method. We aimed to develop a rapid, non-invasive, cost-effective, and easy-to-use deep learning approach for identifying oral cavity squamous cell carcinoma (OCSCC) patients using photographic images.MethodsWe developed an automated deep learning algorithm using cascaded convolutional neural networks to detect OCSCC from photographic images. We included all biopsy-proven OCSCC photographs and normal controls of 44,409 clinical images collected from 11 hospitals around China between April 12, 2006, and Nov 25, 2019. We trained the algorithm on a randomly selected part of this dataset (development dataset) and used the rest for testing (internal validation dataset). Additionally, we curated an external validation dataset comprising clinical photographs from six representative journals in the field of dentistry and oral surgery. We also compared the performance of the algorithm with that of seven oral cancer specialists on a clinical validation dataset. We used the pathological reports as gold standard for OCSCC identification. We evaluated the algorithm performance on the internal, external, and clinical validation datasets by calculating the area under the receiver operating characteristic curves (AUCs), accuracy, sensitivity, and specificity with two-sided 95% CIs.Findings1469 intraoral photographic images were used to validate our approach. The deep learning algorithm achieved an AUC of 0·983 (95% CI 0·973–0·991), sensitivity of 94·9% (0·915–0·978), and specificity of 88·7% (0·845–0·926) on the internal validation dataset (n = 401), and an AUC of 0·935 (0·910–0·957), sensitivity of 89·6% (0·847–0·942) and specificity of 80·6% (0·757–0·853) on the external validation dataset (n = 402). For a secondary analysis on the internal validation dataset, the algorithm presented an AUC of 0·995 (0·988–0·999), sensitivity of 97·4% (0·932–1·000) and specificity of 93·5% (0·882–0·979) in detecting early-stage OCSCC. On the clinical validation dataset (n = 666), our algorithm achieved comparable performance to that of the average oral cancer expert in terms of accuracy (92·3% [0·902–0·943] vs 92.4% [0·912–0·936]), sensitivity (91·0% [0·879–0·941] vs 91·7% [0·898–0·934]), and specificity (93·5% [0·909–0·960] vs 93·1% [0·914–0·948]). The algorithm also achieved significantly better performance than that of the average medical student (accuracy of 87·0% [0·855–0·885], sensitivity of 83·1% [0·807–0·854], and specificity of 90·7% [0·889–0·924]) and the average non-medical student (accuracy of 77·2% [0·757–0·787], sensitivity of 76·6% [0·743–0·788], and specificity of 77·9% [0·759–0·797]).InterpretationAutomated detection of OCSCC by deep-learning-powered algorithm is a rapid, non-invasive, low-cost, and convenient method, which yielded comparable performance to that of human specialists and has the potential to be used as a clinical tool for fast screening, earlier detection, and therapeutic efficacy assessment of the cancer.

Maxillary Incisors of the Horse before and at the Beginning of the Teeth Shedding: Radiographic and CT Study.

Simple SummaryAlthough much is known about equine dentistry, there is a period of the horse’s life, prior to teeth shedding, in which there is lack of knowledge related to the development of deciduous incisors and dental germs of permanent incisors. To gain insight into the radiographic appearance of maxillary deciduous incisors and dental germs of maxillary permanent incisors during this period, a radiographic and computed tomography study of 25 horse skulls was made. Data regarding morphology and development were obtained. The results of the present study indicate that radiographic intraoral images are suitable to identify the grade of development of the dental germs of permanent incisors in horses before dental change. A detailed description of the radiographic appearance of deciduous incisors and dental germs of permanent incisors will help clinicians to expand their knowledge for diagnostic and treatment purposes.To gain insight into the radiographic appearance of maxillary deciduous incisors and dental germs of maxillary permanent incisors in the period prior to teeth shedding, radiographs and computed tomography (CT) of 25 horse skulls, with an estimated age of between 12 and 42 months, were studied. Data regarding morphology and development were obtained. Dental germs of first maxillary permanent incisors were identified radiographically as rounded radiolucent areas at the level of the apical parts of the first deciduous incisors, in skulls with an estimated age of twelve months. The first sign of crown mineralization of these dental germs appeared in skulls supposedly a few months older. Before teeth shedding, the unerupted, mineralized crowns of the first permanent incisor could be identified radiographically relatively caudal to the corresponding first deciduous incisors. The results of the present study indicate that radiographic intraoral images are suitable to identify the grade of development of the dental germs of maxillary permanent incisors. A detailed description of the radiographic appearance of deciduous incisors and dental germs of permanent incisors will help clinicians to expand their knowledge for diagnostic or treatment purposes.