Automatic caries detection in bitewing radiographs-Part II: experimental comparison.

OBJECTIVE: The objective of this study was to compare the detection of caries in bitewing radiographs by multiple dentists with an automatic method and to evaluate the detection performance in the absence of a reliable ground truth. MATERIALS AND METHODS: Four experts and three novices marked caries using bounding boxes in 100 bitewing radiographs. The same dataset was processed by an automatic object detection deep learning method. All annotators were compared in terms of the number of errors and intersection over union (IoU) using pairwise comparisons, with respect to the consensus standard, and with respect to the annotator of the training dataset of the automatic method. RESULTS: The number of lesions marked by experts in 100 images varied between 241 and 425. Pairwise comparisons showed that the automatic method outperformed all dentists except the original annotator in the mean number of errors, while being among the best in terms of IoU. With respect to a consensus standard, the performance of the automatic method was best in terms of the number of errors and slightly below average in terms of IoU. Compared with the original annotator, the automatic method had the highest IoU and only one expert made fewer errors. CONCLUSIONS: The automatic method consistently outperformed novices and performed as well as highly experienced dentists. CLINICAL SIGNIFICANCE: The consensus in caries detection between experts is low. An automatic method based on deep learning can improve both the accuracy and repeatability of caries detection, providing a useful second opinion even for very experienced dentists.

ORCA-EFCD consensus report on clinical recommendation for caries diagnosis. Paper I: caries lesion detection and depth assessment.

OBJECTIVES: The aim of the present consensus paper was to provide recommendations for clinical practice considering the use of visual examination, dental radiography and adjunct methods for primary caries detection. MATERIALS AND METHODS: The executive councils of the European Organisation for Caries Research (ORCA) and the European Federation of Conservative Dentistry (EFCD) nominated ten experts each to join the expert panel. The steering committee formed three work groups that were asked to provide recommendations on (1) caries detection and diagnostic methods, (2) caries activity assessment and (3) forming individualised caries diagnoses. The experts responsible for "caries detection and diagnostic methods" searched and evaluated the relevant literature, drafted this manuscript and made provisional consensus recommendations. These recommendations were discussed and refined during the structured process in the whole work group. Finally, the agreement for each recommendation was determined using an anonymous Delphi survey. RESULTS: Recommendations (N = 8) were approved and agreed upon by the whole expert panel: visual examination (N = 3), dental radiography (N = 3) and additional diagnostic methods (N = 2). While the quality of evidence was found to be heterogeneous, all recommendations were agreed upon by the expert panel. CONCLUSION: Visual examination is recommended as the first-choice method for the detection and assessment of caries lesions on accessible surfaces. Intraoral radiography, preferably bitewing, is recommended as an additional method. Adjunct, non-ionising radiation methods might also be useful in certain clinical situations. CLINICAL RELEVANCE: The expert panel merged evidence from the scientific literature with practical considerations and provided recommendations for their use in daily dental practice.

Automatic deep learning detection of overhanging restorations in bitewing radiographs.

OBJECTIVES: This study aimed to assess the effectiveness of deep convolutional neural network (CNN) algorithms for the detecting and segmentation of overhanging dental restorations in bitewing radiographs. METHODS: A total of 1160 anonymized bitewing radiographs were used to progress the artificial intelligence (AI) system for the detection and segmentation of overhanging restorations. The data were then divided into three groups: 80% for training (930 images, 2399 labels), 10% for validation (115 images, 273 labels), and 10% for testing (115 images, 306 labels). A CNN model known as You Only Look Once (YOLOv5) was trained to detect overhanging restorations in bitewing radiographs. After utilizing the remaining 115 radiographs to evaluate the efficacy of the proposed CNN model, the accuracy, sensitivity, precision, F1 score, and area under the receiver operating characteristic curve (AUC) were computed. RESULTS: The model demonstrated a precision of 90.9%, a sensitivity of 85.3%, and an F1 score of 88.0%. Furthermore, the model achieved an AUC of 0.859 on the receiver operating characteristic (ROC) curve. The mean average precision (mAP) at an intersection over a union (IoU) threshold of 0.5 was notably high at 0.87. CONCLUSIONS: The findings suggest that deep CNN algorithms are highly effective in the detection and diagnosis of overhanging dental restorations in bitewing radiographs. The high levels of precision, sensitivity, and F1 score, along with the significant AUC and mAP values, underscore the potential of these advanced deep learning techniques in revolutionizing dental diagnostic procedures.

Caries detection and characterization in pediatric patients using iTero 5D near-infrared technology.

INTRODUCTION: Near-infrared imaging (NIRI) has been proposed as an alternative to radiographs and uses nonionizing radiation in the near-infrared spectrum to differentially scatter light off tooth surfaces and generate images allowing interproximal caries detection. The new iTero 5D Element Scanner (Align Technology, Santa Clara, Calif) has integrated NIRI capture and viewing technology but has not been specifically studied in a pediatric population. Therefore, this study aimed to assess clinicians' abilities to detect and characterize caries in pediatric patients using this instrument. METHODS: Bitewing (BW) radiographs and an intraoral scan were captured on 17 pediatric patients (344 surfaces were analyzed). Data were randomized and graded by 5 calibrated clinicians individually with 2 different rounds of grading. RESULTS: The reliability of lesion characterization (ie, grade) among examiners was poor to fair in both systems, whereas the reliability of caries detection was moderate. Both systems had a high specificity and low sensitivity. The reliability of the characterization of the combined dataset was moderate to substantial, whereas, for detection, it was substantial. CONCLUSIONS: When using either BW or NIRI analysis, reliability is relatively poor, and clinicians are more likely to correctly identify a healthy tooth surface when compared with a carious surface. There is a small difference in error rate between BW and NIRI systems that is not likely to be clinically significant. When NIRI and BW data are combined, clinician agreement for both lesion characterization and detection increases significantly.

Near-infrared imaging in orthodontic intraoral scanners for early interproximal caries detection.

INTRODUCTION: Intraoral scanners commonly used in orthodontic offices now offer near-infrared imaging (NIRI) technology, advertised as a screening tool to identify interproximal caries. This study aimed to evaluate the reliability and validity of NIRI detection of interproximal carious lesions in a common intraoral scanner (iTero Element 5D; Align Technology, San Jose, Calif) with and without bitewing radiograph complement, compared with a microcomputed tomography (micro-CT) reference standard. METHODS: Extracted human posterior teeth (premolars and molars) were selected for early (noncavitated) interproximal carious lesions (n = 39) and sound control surfaces (n = 47). The teeth were scanned via micro-CT for evaluation by 2 blinded evaluators using consensus scoring. The teeth were mounted to simulate anatomic interproximal contacts and underwent a NIRI scan using iTero Element 5D and bitewing radiographs. Two trained, calibrated examiners independently evaluated (1) near-infrared images alone with clinical photograph, (2) bitewing radiograph alone with clinical photograph, and (3) near-infrared images with bitewing radiograph and clinical photograph in combination, after at least a 10-day washout period between each evaluation. RESULTS: Interrater reliability was highest for NIRI alone (k = 0.533) compared with bitewing radiograph alone (k = 0.176) or in combination (k = 0.256). NIRI alone showed high specificity (0.83-0.96) and moderate sensitivity (0.42-0.63) compared with a micro-CT reference standard. Dentin lesions were significantly more reliably detected than enamel lesions. CONCLUSIONS: After rigorous training and calibration, NIRI can be used with moderate reliability, high specificity, and moderate sensitivity to detect noncavitated interproximal carious lesions.

AI-Dentify: deep learning for proximal caries detection on bitewing x-ray - HUNT4 Oral Health Study.

BACKGROUND: Dental caries diagnosis requires the manual inspection of diagnostic bitewing images of the patient, followed by a visual inspection and probing of the identified dental pieces with potential lesions. Yet the use of artificial intelligence, and in particular deep-learning, has the potential to aid in the diagnosis by providing a quick and informative analysis of the bitewing images. METHODS: A dataset of 13,887 bitewings from the HUNT4 Oral Health Study were annotated individually by six different experts, and used to train three different object detection deep-learning architectures: RetinaNet (ResNet50), YOLOv5 (M size), and EfficientDet (D0 and D1 sizes). A consensus dataset of 197 images, annotated jointly by the same six dental clinicians, was used for evaluation. A five-fold cross validation scheme was used to evaluate the performance of the AI models. RESULTS: The trained models show an increase in average precision and F1-score, and decrease of false negative rate, with respect to the dental clinicians. When compared against the dental clinicians, the YOLOv5 model shows the largest improvement, reporting 0.647 mean average precision, 0.548 mean F1-score, and 0.149 mean false negative rate. Whereas the best annotators on each of these metrics reported 0.299, 0.495, and 0.164 respectively. CONCLUSION: Deep-learning models have shown the potential to assist dental professionals in the diagnosis of caries. Yet, the task remains challenging due to the artifacts natural to the bitewing images.

Clinical, radiographic and biochemical assessment of proximal cavities restored with composite resin using incremental vs. bulk packing techniques: One-year randomized clinical trial.

BACKGROUND: Bulk-fill resin composites may suffer from recurrent caries around compound proximal restorations in posterior teeth, especially at the proximo-gingival interface.Over 12 months, will the bulk fill technique affect the caries recurrence rate at gingival margins when compared to the conventional incremental packing technique? How early will the first clinical, radiographical, and biochemical evidence of caries recurrence occur? METHODS: After randomization, in 30 patients with two compound (OM or OD) supragingival lesions, one tooth was restored using the bulk fill technique on one side (group 1) (n = 15). In contrast, the other tooth on the other side was restored utilizing the incremental layering technique (group 2) (n = 15). Both teeth received restorative material (X-tra fil, Voco, Cuxhaven, Germany). The FDI criteria were used to evaluate restorations. As for the periodontal assessment, the gingival index, plaque index, papillary bleeding scoring index and periodontal pocket depth were evaluated. The gingival crevicular fluid (GCF) specimens were gathered, and MMP-9 was extracted and quantitated by ELISA. A customized radiographic template was designed, and 3D printed digital bitewing radiographs were taken. Assessments were done clinically, radiographically and biochemically at baseline (1 week) and after 3, 6 and 12 months. Data was statistically analyzed. RESULTS: The null hypothesis was accepted clinically; no statistically significant differences appeared between bulk and incrementally filled posterior restorations. As for the radiographic assessment, the null hypothesis was accepted except for increased periodontal ligament width at 3 months. The null hypothesis for the biochemical evaluation was rejected as there were significant changes in levels of MMP-9 at different testing times. CONCLUSIONS: 1. With similar results but less sensitivity and significant time saving, the bulk fill technique can be considered an efficient alternative to the incremental fill technique in restoring proximal cavities. 2. Early evidence of caries recurrence can be correlated to an increase in the MMP-9 level in gingival crevicular fluid, followed by an increase in radiographic periodontal ligament width measurement. TRIAL REGISTRATION: An ethical approval from the Research Ethics Committee at the Faculty of Dentistry, October 6 University, (Approval No. RECO6U/5-2022). The study was registered at the Pan African Clinical Trials Registry on 24/07/2023 with an identification number (PACTR202307573531455).

Vertical root fracture detection with cone-beam computed tomography in Biodentine™ filled teeth.

PURPOSE: This study aimed to evaluate the accuracy of detecting vertical root fractures in Biodentine™-filled teeth using the Promax 3Dmax cone-beam computed tomography (CBCT) unit compared to periapical radiographs. It tested hypotheses regarding CBCT's diagnostic superiority in non-root-filled and Biodentine™-root-filled maxillary central incisors and assessed the impact of smaller field of view and lower intensity settings on detection accuracy. MATERIALS AND METHODS: Extracted maxillary incisors were divided into groups based on fracture status and root filling material, then placed in a Thiel-embalmed skull to simulate clinical conditions. The teeth were imaged using periapical radiographs and the CBCT unit under different settings. Fracture thickness was measured with microcomputed tomography for accuracy benchmarking. Multiple observers assessed the images, and statistical analyses were conducted to evaluate diagnostic performance. RESULTS: Intra-rater reliabilities of consensus scores ranged from good to very good. Specificities were generally higher than sensitivities across all imaging modalities, but sensitivities remained constantly low. None of the Area Under the Curve scores exceeded 0.6, indicating poor overall accuracy for all imaging modalities. Paired comparisons of the area differences under Receiver Operator Characteristic curves revealed no significant differences between the CBCT and periapical radiograph techniques for detecting vertical root fractures in either Biodentine™-filled or non-root-filled teeth. CONCLUSIONS: There was no significant accuracy improvement of the current CBCT device (Promax 3Dmax, Planmeca, Finland) over periapical radiographs in detecting small vertical root fractures in both non-root-filled and Biodentine™-root-filled maxillary central incisors. A smaller field of view with lower intensity did not enhance detection accuracy. These results highlight the challenges in accurately detecting small VRFs, emphasizing the need for further research and technological advancements in this domain.

Use of artificial intelligence software in dental education: A study on assisted proximal caries assessment in bitewing radiographs.

INTRODUCTION: Teaching of dental caries diagnostics is an essential part of dental education. Diagnosing proximal caries is a challenging task, and automated systems applying artificial intelligence (AI) have been introduced to assist in this respect. Thus, the implementation of AI for teaching purposes may be considered. The aim of this study was to assess the impact of an AI software on students' ability to detect enamel-only proximal caries in bitewing radiographs (BWs) and to assess whether proximal tooth overlap interferes with caries detection. MATERIALS AND METHODS: The study included 74 dental students randomly allocated to either a test or control group. At two sessions, both groups assessed proximal enamel caries in BWs. At the first session, the test group registered caries in 25 BWs using AI software (AssistDent®) and the control group without using AI. One month later, both groups detected caries in another 25 BWs in a clinical setup without using the software. The student's registrations were compared with a reference standard. Positive agreement (caries) and negative agreement (no caries) were calculated, and t-tests were applied to assess whether the test and control groups performed differently. Moreover, t-tests were applied to test whether proximal overlap interfered with caries registration. RESULTS: At the first and second sessions, 56 and 52 tooth surfaces, respectively, were detected with enamel-only caries according to the reference standard. At session 1, no significant difference between the control (48%) and the test (42%) group was found for positive agreement (p = .08), whereas the negative agreement was higher for the test group (86% vs. 80%; p = .02). At session 2, there was no significant difference between the groups. The test group improved for positive agreement from session 1 to session 2 (p < .001), while the control group improved for negative agreement (p < .001). Thirty-eight per cent of the tooth surfaces overlapped, and the mean positive agreement and negative agreement were significantly lower for overlapping surfaces than non-overlapping surfaces (p < .001) in both groups. CONCLUSION: Training with the AI software did not impact on dental students' ability to detect proximal enamel caries in bitewing radiographs although the positive agreement improved over time. It was revealed that proximal tooth overlap interfered with caries detection.

Underlying Dentin Shadows (ICDAS 4) in Occlusal Surface of Permanent Teeth Have Low Progression Rate after 1-2 Years.

The aim of this prospective cohort study was to assess the radiographic progression of underlying dentin shadows (UDS) on the occlusal surfaces of permanent posterior teeth of adolescents and young adults over 1-2 years and to identify possible risk factors. A total of 149 UDS lesions (from 101 individuals) were included at baseline. Each participant had to present at least one UDS to be considered eligible for the study. Data collection included the application of a questionnaire, clinical examination, and bilateral bitewing radiographs, performed at baseline and after 1-2 years. The association between possible predictors and UDS progression (defined radiographically as an increase in the radiographic score from baseline to follow-up) was assessed using Weibull regression models. Hazard ratios (HRs) and their 95% confidence intervals (CIs) were estimated. A total of 81 individuals (mean age: 24.0, standard deviation: 8.03) were reexamined after 1-2 years (742 occlusal surfaces, of which 118 were UDS). The overall progression rate was 8.6% after 1-2 years, being 12.6% for UDS without baseline radiolucency and 20% for UDS with baseline radiolucency. The risk analysis showed that UDS without radiolucency at baseline had a similar likelihood of progression (adjusted HR = 1.71, 95% CI = 0.68-4.32, p = 0.26) while UDS with radiolucency at baseline were more likely to progress (adjusted HR = 2.96, 95% CI = 1.06-8.26, p = 0.04) than the reference category (sound occlusal surfaces without radiolucency). These estimates were adjusted for caries prevalence, tooth type, and arch. This study showed low progression rates of UDS after 1-2 years. The presence of radiolucency at baseline was found to predict UDS progression.

Assessment of YOLOv3 for caries detection in bitewing radiographs based on the ICCMS™ radiographic scoring system.

OBJECTIVES: To assess the feasibility of the YOLOv3 model under the intersection over union (IoU) thresholds of 0.5 (IoU50) and 0.75 (IoU75) for caries detection in bitewing radiographs based on the International Caries Classification and Management System (ICCMS™). MATERIALS AND METHODS: We trained the YOLOv3 model by feeding 994 annotated radiographs with the IoU50 and IoU75 thresholds. The testing procedure (n = 175) was subsequently conducted to evaluate the model's prediction metrics on caries classification based on the ICCMS™ radiographic scoring system. RESULTS: Regarding the 4-class classification representing caries severity, YOLOv3 could accurately detect and classify enamel caries and initial dentin caries (class RA) (IoU50 vs IoU75: precision, 0.75 vs 0.71; recall, 0.67 vs 0.64). Concerning the 7-class classification signifying specific caries depth (class 0, healthy tooth; classes RA1-3, initial caries affecting outer half, inner half of enamel, and the outer 1/3 of dentin; class RB4, caries extending to the middle 1/3 of dentin; classes RC5-6, extensively cavitated caries affecting the inner 1/3 of dentin and involving the pulp chamber), YOLOv3 could accurately detect and classify caries with pulpal exposure (class RC6) (IoU50 vs IoU75: precision, 0.77 vs 0.73; recall, 0.61 vs 0.57) but it failed to predict the outer half of enamel caries (class RA1) (IoU50 vs IoU75: precision, 0.35 vs 0.32; recall, 0.23 vs 0.21). CONCLUSIONS: YOLOv3 yielded acceptable performances in both IoU50 and IoU75. Although the performance metrics decreased in the 7-class detection, the two thresholds revealed comparable results. However, the model could not consistently detect initial-stage caries affecting the outermost surface of the enamel. CLINICAL RELEVANCE: YOLOv3 could be implemented to detect and classify dental caries according to the ICCMS™ classification with acceptable performances to assist dentists in making treatment decisions.

Enhancing Caries Detection in Bitewing Radiographs Using YOLOv7.

The study aimed to evaluate the impact of image size, area of detection (IoU) thresholds and confidence thresholds on the performance of the YOLO models in the detection of dental caries in bitewing radiographs. A total of 2575 bitewing radiographs were annotated with seven classes according to the ICCMS™ radiographic scoring system. YOLOv3 and YOLOv7 models were employed with different configurations, and their performances were evaluated based on precision, recall, F1-score and mean average precision (mAP). Results showed that YOLOv7 with 640 × 640 pixel images exhibited significantly superior performance compared to YOLOv3 in terms of precision (0.557 vs. 0.268), F1-score (0.555 vs. 0.375) and mAP (0.562 vs. 0.458), while the recall was significantly lower (0.552 vs. 0.697). The following experiment found that the overall mAPs did not significantly differ between 640 × 640 pixel and 1280 × 1280 pixel images, for YOLOv7 with an IoU of 50% and a confidence threshold of 0.001 (p = 0.866). The last experiment revealed that the precision significantly increased from 0.570 to 0.593 for YOLOv7 with an IoU of 75% and a confidence threshold of 0.5, but the mean-recall significantly decreased and led to lower mAPs in both IoUs. In conclusion, YOLOv7 outperformed YOLOv3 in caries detection and increasing the image size did not enhance the model's performance. Elevating the IoU from 50% to 75% and confidence threshold from 0.001 to 0.5 led to a reduction of the model's performance, while simultaneously improving precision and reducing recall (minimizing false positives and negatives) for carious lesion detection in bitewing radiographs.

Performance of the caries diagnosis feature of intraoral scanners and near-infrared imaging technology-A narrative review.

PURPOSE: To describe and discuss the benefits and drawbacks of various dental caries diagnostic techniques, including the use of intraoral scanners for caries diagnosis based on near-infrared imaging (NIR) technology. MATERIAL AND METHODS: A MEDLINE search from 1980-2023 focused on dental caries diagnostic techniques, emphasizing intraoral scanners using NIR technology. Alternative caries detection methods were also evaluated for their advantages and limitations, enabling a comparison with NIR. The review included traditional caries tools, the latest detection methods, and NIR's role in intraoral scanners, drawing from case reports and both in vivo and in vitro studies. Keywords like "caries detection," "intraoral scanners," and "Near Infrared Imaging (NIRI)" guided the search. After screening titles and abstracts for relevance, full texts with valuable insights were thoroughly analyzed. The data was grouped into three: traditional diagnostics, advanced digital methods, and intraoral scanner-based detection. RESULTS: This comprehensive narrative review described and discussed the current state of dental caries diagnostic methods, given the insufficient number of clinical investigations suitable for a systematic review. Traditional caries diagnosis techniques have shown variable accuracy dependent on a dentist's experience and the potential over-removal of healthy tooth structures. Intraoral scanners have emerged as a novel caries detection method, because of their integration of NIR technology. Various studies have confirmed the efficacy of NIR in detecting interproximal caries and in the early diagnosis of non-cavitated caries. Specifically, intraoral scanners have demonstrated promising results, proving comparable to established diagnostic methods like bitewing radiography. Nevertheless, while the integration of NIR into intraoral scanners seems promising, the technology still faces challenges, notably its accuracy in detecting secondary and subgingival cavities. However, with anticipated integrations of AI, NIR in intraoral scanners could revolutionize early caries detection. CONCLUSIONS: Intraoral scanners with NIR technology offer non-destructive imaging, real-time lesion visuals, and enhanced patient communication. Although comparable to bitewing radiography in some studies, a universally accepted diagnostic tool is lacking. Future research should compare them with existing methods, focusing on clinical outcomes, cost-effectiveness, and patient acceptance.

Detection of Proximal Caries Lesions on Bitewing Radiographs Using Deep Learning Method.

This study aimed to evaluate the validity of a deep learning-based convolutional neural network (CNN) for detecting proximal caries lesions on bitewing radiographs. A total of 978 bitewing radiographs, 10,899 proximal surfaces, were evaluated by two endodontists and a radiologist, of which 2,719 surfaces were diagnosed and annotated with proximal caries and 8,180 surfaces were sound. The data were randomly divided into two datasets, with 818 bitewings in the training and validation dataset and 160 bitewings in the test dataset. Each annotation in the test set was then classified into 5 stages according to the extent of the lesion (E1, E2, D1, D2, D3). Faster R-CNN, a deep learning-based object detection method, was trained to detect proximal caries in the training and validation dataset and then was assessed on the test dataset. The diagnostic accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and receiver operating characteristic curve were calculated. The performance of the network in the overall and different stages of lesions was compared with that of postgraduate students on the test dataset. A total of 388 carious lesions and 1,435 sound surfaces were correctly identified by the neural network; hence, the accuracy was 0.87. Furthermore, 27.6% of lesions went undetected, and 7% of sound surfaces were misdiagnosed by the neural network. The sensitivity, specificity, PPV, and NPV of the neural network were 0.72, 0.93, 0.77, and 0.91, respectively. In contrast with the network, 52.8% of lesions went undetected by the students, yielding a sensitivity of only 0.47. The F1-score of the students was 0.57, while the F1-score of the network was 0.74 despite the accuracy of 0.82. A significant difference in the sensitivity was found between the model and the postgraduate students when detecting different stages of lesions (p < 0.05). For early lesions which limited in enamel and the outer third of dentin, the neural network had sensitivities all above or at 0.65, while students showed sensitivities below 0.40. From our results, we conclude that the CNN may be an assistant in detecting proximal caries on bitewings.

Augmented Vision for Dental Students' Education in Detecting Proximal Carious Lesions on Bitewing Radiographs: A Randomized Controlled Trial.

This two-arm, parallel, randomized controlled trial aimed to assess the effect of augmented vision (AV, using interactive color overlays) on the education of dental students in detecting proximal carious lesions on bitewing radiographs compared to black-and-white textbook-like illustrations. Forty-eight preclinical third-year dental students were randomized using a random number generator into two learning groups: test (AV, allowing interaction with color-highlighted carious lesions, n = 24) and control (showing the native radiograph and a black-and-white illustration displaying the carious lesion, n = 24). First, students had 2 weeks to assess 50 bitewings (lesion prevalence on the tooth level: 54.5%) in the test or control. Due to the nature of the intervention, participants could not be blinded toward the intervention. After that, they were asked to detect lesions on 10 independent bitewings and to assess lesion extent (outer/inner enamel; outer/middle/inner dentin). The reference test was constituted by two experienced dentists. No significant differences in accuracy (test 0.84 [95% CI: 0.79, 0.88]; control 0.83 [0.78, 0.87]), AUC (test 0.82 [0.81, 0.84]; control 0.81 [0.80, 0.83]) and F1 score (test 0.79 [0.75, 0.82]; control 0.77 [0.72, 0.81]) were observed between groups. Students of both groups showed difficulties in differentiating enamel from dentin carious lesions. While AV was reported to be motivating by students, it did not increase their accuracy.

Diagnosis of Interproximal Caries Lesions in Bitewing Radiographs Using a Deep Convolutional Neural Network-Based Software.

The aim of this study was to evaluate the diagnostic reliability of a web-based artificial intelligence program for the detection of interproximal caries in bitewing radiographs. Three hundred bitewing radiographs of patients were subjected to the evaluation of a convolutional neural network. First, the images were visually evaluated by a previously trained and calibrated operator with radiodiagnosis experience. Then, ground truth was established and was clinically validated. For enamel caries, clinical assessment included a combination of clinical-visual and radiography evaluations. For dentin caries, clinical validation was performed by instrumentally accessing the cavity. Second, the images were uploaded and analyzed by the web-based software. Four different models were established to analyze its evaluations according to the confidence threshold (0-100%) offered by the program: model 1 (values >0% were considered positive and values of 0% were considered negative), model 2 (values ≥25% were considered positive and values <25% were considered negative), model 3 (values ≥50% were considered positive and values <50% were considered negative), and model 4 (values ≥75% were considered positive and values <75% were considered negative). The accuracy rate (A), sensitivity (S), specificity (E), positive predictive value (PPV), negative predictive value (NPV), positive likelihood ratio (PLR), negative likelihood ratio (NLR), and areas under receiver operating characteristic curves (AUC) were calculated for the four models of agreement with the software. Models showed the following results respectively: A = 70.8%, 82%, 85.6%, 86.1%; S = 87%, 69.8%, 57%, 41.6%; E = 66.3%, 85.4%, 93.7%, 98.5%; PPV = 42%, 57.2%, 71.6%, 88.6%; NPV = 94.8%, 91%, 88.6%, 85.8%; PLR = 2.58, 4.78, 9.05, 27.73; NLR = 0.2, 0.35, 0.46, 0.59; AUC = 0.767, 0.777, 0.753, 0.701. Findings in the present study suggest that the artificial intelligence web-based software provides a good diagnostic reliability on the detection of dental caries. Our study highlighted model 2 for showing the best results to differentiate between healthy teeth and decayed teeth.

Diagnostic validity of early proximal caries detection using near-infrared imaging technology on 3D range data of posterior teeth.

OBJECTIVES: This in vitro study analysed potential of early proximal caries detection using 3D range data of teeth consisting of near-infrared reflection images at 850 nm (NIRR). MATERIALS AND METHODS: Two hundred fifty healthy and carious permanent human teeth were arranged pairwise, examined with bitewing radiography (BWR) and NIRR and validated with micro-computed tomography. NIRR findings were evaluated from buccal, lingual and occlusal (trilateral) views according to yes/no decisions about presence of caries. Reliability assessments included kappa statistics and revealed high agreement for both methods. Statistical analysis included cross tabulation and calculation of sensitivity, specificity and AUC. RESULTS: Underestimation of caries was 24.8% for NIRR and 26.4% for BWR. Overestimation was 10.4% for occlusal NIRR and 0% for BWR. Trilateral NIRR had overall accuracy of 64.8%, overestimation of 15.6% and underestimation of 19.6%. NIRR and BWR showed high specificity and low sensitivity for proximal caries detection. CONCLUSIONS: NIRR achieved diagnostic results comparable to BWR. Trilateral NIRR assessments overestimated presence of proximal caries, revealing stronger sensitivity for initial caries detection than BWR. CLINICAL RELEVANCE: NIRR provided valid complement to BWR as diagnostic instrument. Investigation from multiple angles did not substantially improve proximal caries detection with NIRR.

Diagnosis of interproximal caries lesions with deep convolutional neural network in digital bitewing radiographs.

OBJECTIVES: This study aimed to investigate the effectiveness of deep convolutional neural network (CNN) in the diagnosis of interproximal caries lesions in digital bitewing radiographs. METHODS AND MATERIALS: A total of 1,000 digital bitewing radiographs were randomly selected from the database. Of these, 800 were augmented and annotated as "decay" by two experienced dentists using a labeling tool developed in Python programming language. The 800 radiographs were consisted of 11,521 approximal surfaces of which 1,847 were decayed (lesion prevalence for train data was 16.03%). A CNN model known as you only look once (YOLO) was modified and trained to detect caries lesions in bitewing radiographs. After using the other 200 radiographs to test the effectiveness of the proposed CNN model, the accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and area under the receiver operating characteristic curve (AUC) were calculated. RESULTS: The lesion prevalence for test data was 13.89%. The overall accuracy of the CNN model was 94.59% (94.19% for premolars, 94.97% for molars), sensitivity was 72.26% (75.51% for premolars, 68.71% for molars), specificity was 98.19% (97.43% for premolars, 98.91% for molars), PPV was 86.58% (83.61% for premolars, 90.44% for molars), and NPV was 95.64% (95.82% for premolars, 95.47% for molars). The overall AUC was measured as 87.19%. CONCLUSIONS: The proposed CNN model showed good performance with high accuracy scores demonstrating that it could be used in the diagnosis of caries lesions in bitewing radiographs. CLINICAL SIGNIFICANCE: Correct diagnosis of dental caries is essential for a correct treatment procedure. CNNs can assist dentists in diagnosing approximal caries lesions in bitewing radiographs.

New Caries Diagnostic Tools in Intraoral Scanners: A Comparative In Vitro Study to Established Methods in Permanent and Primary Teeth.

The aim of this in vitro study was to systematically investigate new caries diagnostic tools, including three intraoral scanners, and compare them to established diagnostic methods. For a standardized analysis of occlusal and proximal caries lesions, human permanent and primary teeth (n = 64) were embedded in models and investigated in a phantom head using six different caries diagnostic methods: visual examination, bitewing radiography, Diagnocam (KaVo, Biberach, Germany), Trios 4 (3Shape, Copenhagen, Denmark), iTero Element 5D (Align Technology, San José, CA, USA), and Planmeca Emerald S (Planmeca, Helsinki, Finland). The diagnostic methods were investigated and compared to reference µ-CT for permanent and primary teeth separately. For occlusal caries diagnostics in permanent teeth, the best agreement to the reference (reliability) was obtained for Planmeca Emerald S (ĸ = 0.700), whereas in primary teeth, for visual examination (ĸ = 0.927), followed by Trios 4 (ĸ = 0.579). Regarding proximal caries diagnostics, bitewing radiography, as the gold standard, exhibited the highest agreement for permanent (ĸ = 0.643) and primary teeth (ĸ = 0.871). Concerning the analysis of the diagnostic quality (sensitivity and specificity) using receiver operating characteristic (ROC) curve analysis, comparable findings were obtained for area under curve (AUC) values as for reliability. No diagnostic method could be identified that is generally suitable for occlusal and proximal lesions in both dentitions. Overall, caries diagnostics with intraoral scanners seem to be interesting tools that should be further investigated in clinical studies.

Detection Methods for Early Caries Diagnosis: A Systematic Review and Meta-Analysis.

The aim was to appraise the evidence on the performance of various means for the detection of incipient caries in vivo. Five databases of published and unpublished research were searched for studies from January 2000 to October 2019. Search terms included "early caries" and "caries detection." Inclusion criteria involved diagnostic test accuracy studies for early caries detection in permanent and primary teeth. A risk-of-bias assessment was performed using the QUADAS-2 tool. We performed the study selection, data extraction, and risk-of-bias assessment in duplicate. The review protocol was a priori registered in the Open Science Framework. Of the initially 22,964 search results, 51 articles were included. For permanent teeth, when histologic examination was considered as the reference for occlusal surfaces, the sensitivity (Se) range appeared high for the DIAGNOdent Pen (DD Pen) at 0.81-0.89, followed by ICDAS-II at 0.62-1, DIAGNOdent (DD) at 0.48-1, and bitewing radiography (BW) at 0-0.29. The corresponding specificity (Sp) range was: DD Pen 0.71-0.8, ICDAS-II 0.5-0.84, DD 0.54-1, and BW 0.96-1. When operative intervention served as the reference for occlusal surfaces, again, the DD means valued the most promising results on Se: DD 0.7-0.96 and DD Pen 0.55-0.90, followed by ICDAS-II 0.25-0.93, and BW 0-0.83. The Sp range was: DD 0.54-1, DD Pen 0.71-1, ICDAS-II 0.44-1, and BW 0.6-1. For approximal surfaces, the Se was: BW 0.75-0.83, DD Pen 0.6, and ICDAS-II 0.54; the Sp was: BW 0.6-0.9, DD Pen 0.2, and ICDAS-II 1. For primary teeth, under the reference of histologic assessment, the Se range for occlusal surfaces was: DD 0.55-1, DD Pen 0.63-1, ICDAS-II 0.42-1, and BW 0.31-0.96; the respective Sp was: DD 0.5-1, DD Pen 0.44-1, ICDAS-II 0.61-1, and BW 0.79-0.98. For approximal surfaces, the Se range was: DD Pen 0.58-0.63, ICDAS-II 0.42-0.55, and BW 0.14-0.71. The corresponding Sp range was: DD Pen 0.85-0.87, ICDAS-II 0.73-0.93, and BW 0.79-0.98. Se and Sp values varied, due to the heterogeneity regarding the setting of individual studies. Evidently, robust conclusions cannot be drawn, and different diagnostic means should be used as adjuncts to clinical examination. In permanent teeth, visual examination may be enhanced by DD on occlusal surfaces and BW on approximal surfaces. In primary teeth, DD Pen may serve as a supplementary tool across all surfaces.