Detection Of Proximal Caries Research Articles

In vitro early proximal caries detection using trilateral short-wave infrared reflection at 1050 and 1550 nm.

This in vitro study evaluated the diagnostic potential of short-wave infrared reflection (SWIRR) at 1050 and 1550 nm for proximal caries detection from the occlusal, buccal and lingual surfaces of posterior teeth under clinically relevant conditions. Bitewing radiography (BWR) was the alternative index test and micro-computed tomography (μCT) the reference standard. 250 proximal surfaces of extracted human teeth were examined using SWIRR at 1050 and 1550 nm and BWR. SWIRR, BWR and μCT findings were evaluated twice by two trained examiners. SWIRR images were evaluated from occlusal and trilateral (occlusal, buccal and lingual combined) views. Sensitivity, specificity and AUC were calculated. Reliability assessment was performed using κ statistics. SWIRR (1050 nm) showed sensitivity of 0.44 for occlusal and 0.55 for trilateral assessment, paired with specificity of 0.96 and 0.90, whereas SWIRR (1550 nm) showed sensitivity of 0.73 and 0.85 paired with specificity of 0.76 and 0.59. Compared to occlusal view, trilateral SWIRR view revealed ≈10% higher sensitivity and lower specificity. BWR revealed lowest sensitivity (0.30) and highest specificity (0.99). Over-and underestimation of caries demonstrated opposite trends: from 1050-1550 nm, overestimation of trilateral SWIRR increased (0.08-0.29), while underestimation decreased (0.15-0.06). Trilateral SWIRR has higher sensitivity and lower specificity for proximal caries, than occlusal SWIRR. 1050 nm are more suitable for trilateral SWIRR and 1550 nm for occlusal examinations. A combination of SWIRR at 1050 and 1550 nm may exhibit a balanced sensitivity and specificity for proximal caries.

In vitro accuracy of ultra-low dose cone-beam CT for detection of proximal caries.

This study aimed to assess the accuracy of ultra-low dose (ULD) cone-beam CT (CBCT) for detection of proximal caries. This in vitro study evaluated 104 molar and premolar teeth. The teeth were mounted in dry skulls and underwent CBCT with 4 protocols of high-resolution (HR), normal (NORM), ULD-HR, and ULD-NORM; 78 CBCT images were scored by 3 observers for the presence and penetration depth of caries twice with a 2-week interval using a 5-point Likert scale. The teeth were then sectioned and observed under a stereomicroscope (gold standard). The 4 protocols were compared with each other and with the gold standard. The receiver operating characteristic curve was drawn, and the area under the curve (AUC) was calculated and compared by the Chi-square test (alpha = .05). The interobserver agreement ranged from 0.5233 to 0.6034 for ULD-NORM, 0.5380 to 0.6279 for NORM, 0.5856 to 0.6300 for ULD-HR, and 0.6614 to 0.7707 for HR images. The intra-observer agreement ranged from 0.6027 to 0.8812 for ULD-HR, 0.7083 to 0.7556 for HR, 0.6076 to 0.9452 for ULD-NORM, and 0.7012 to 0.9221 for NORM images. Comparison of AUC revealed no significant difference between NORM and ULD-NORM (P > .05), or HR and ULD-HR (P > .05). The highest AUC belonged to HR (0.8529) and the lowest to NORM (0.7774). Considering the significant reduction in radiation dose in ULD CBCT and its acceptable diagnostic accuracy for detection of proximal caries, this protocol may be used for detection of proximal carious lesions and assessment of their depth.

Evaluation of near-infrared digital imaging transillumination compared with bitewing radiography for proximal caries detection in children.

Limitations in traditional caries detection tools have driven the development of alternatives methods, focused on the early lesion detection such as near-infrared digital imaging transillumination (NIDIT). The aim of this study was to evaluate the performance of NIDIT compared with bitewing radiography (BWR) in the detection of interproximal carious lesions in children. A retrospective audit of data from children who had NIDIT, BWR and intraoral photographs was conducted. Carious lesions were scored on a tooth surface level with BWR acting as the primary reference for comparison. Accuracy was determined using multi-class area under the curve (AUC), and correlation was determined using Fleiss' Kappa. Data from 499 tooth surfaces involving 44 children were included in this study. The average age across the participants was 86 months (~7 years) with an average dmft (decayed, missing and filled teeth in primary dentition) of 5.29. Multi-class AUC comparing NIDIT to BWR was 0.70. The correlation between NIDIT and BWR was moderate (0.43), whereas the correlation between photographic examination and BWR was 0.30, which is fair. When compared to BWR, NIDIT showed a high specificity but a low sensitivity for proximal caries detection in primary teeth.

Evaluation of the effects of postprocessing settings in digital bitewing radiographs on proximal caries detection.

Radiographs are an integral part of detecting proximal caries. The aim of this study was to evaluate the effect of contrast, brightness, noise, sharpness, and γ adjustment of digital intraoral radiographs on the diagnosis of proximal caries. In this in vitro study, 40 extracted teeth including 20 premolars and 20 molars with enamel lesions (white spot or dentin discoloration seen through the enamel) were mounted together in groups of eight inside the skull. Bitewing radiographic images of each dental group were obtained by a photostimulable phosphor plate sensor with exposure conditions of 8 mA, 70 kV, and 0.2 s. The images were reconstructed by the built-in software and examined by two oral and maxillofacial radiologists in various settings of contrast, brightness, sharpness, noise, and γ. The teeth were then cut mesiodistally and the presence or absence of caries was confirmed by an oral and maxillofacial pathologist using a stereomicroscope. The data were then analyzed using the κ agreement coefficient, sensitivity, specificity, and accuracy (α = .05). Adjustment of brightness and contrast led to higher diagnostic performance with an accuracy of 82.5% and 83.8 (for observers 1 and 2, respectively) and 82.5% (for both observers), respectively. Noise adjustment was the least helpful approach for diagnosis of proximal dental caries among other adjustments, with an accuracy of 78.8% and 77.5% for observers 1 and 2, respectively. Brightness and contrast setting was more efficient in improving the diagnostic potential of bitewing radiographs compared to other adjustments.

Comparison of near-infrared imaging with cone-beam computed tomography for proximal caries detection in permanent dentition: An in vivo study

ObjectivesThis study aimed to evaluate the diagnostic performance of near-infrared imaging (NIRI) and unaided visual examination (UVE) in detecting proximal caries in permanent dentition in comparison with cone-beam computed tomography (CBCT). MethodsPatients who underwent NIRI, UVE, and CBCT imaging within 1 week were enrolled. Using CBCT as the reference test, the positive percent agreement (PPA), negative percent agreement (NPA), and overall percent agreement (OPA) of NIRI, UVE, and a combination of the two for detecting proximal caries at different depths and in different tooth locations were assessed. Additionally, the consistency of these diagnostic methods with CBCT was evaluated. ResultsWe evaluated 6,084 proximal surfaces and identified 177 CBCT-positive sites. NIRI had a PPA, NPA, and OPA of 68.93 %, 99.09 %, and 98.21 %, respectively, with a substantial agreement with CBCT. When combined with UVE, the PPA increased by approximately 50 % compared with that of UVE alone. Regarding caries at different depths, NIRI outperformed UVE in detecting initial caries (ICDAS 1–2) over moderate-to-advanced caries (ICDAS 3–6). However, the combined use of NIRI and UVE improved the detection of moderate-to-advanced caries. In the anterior teeth region, NIRI exhibited excellent agreement with CBCT, surpassing its performance in the posterior region. ConclusionsAlthough NIRI cannot fully replace radiographic methods, the substantial agreement of NIRI with CBCT in detecting proximal caries highlights its potential as a complementary tool in routine caries screening, especially when combined with UVE. Clinical SignificanceThis study highlights the potential of NIRI as a radiation-free method for detecting proximal caries in permanent teeth. Early detection through regular NIRI scanning can lead to timely intervention, improved patient outcomes, and reduced overall disease burden.

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

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. 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. 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. 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.

Accuracy of Extraoral Bitewing Compared with Histopathology in Proximal Caries Detection of Primary Molar Teeth

An accurate diagnosis of dental caries leads to a suitable treatment plan and prevents premature loss of primary teeth. Intraoral bitewing radiography helps caries determination but has limitations in pediatric patients with severe gag reflexes while the extraoral bitewing radiograph resolved the patients who refused to undergo intraoral radiographs. The research objective was to compare the accuracy of extraoral bitewing radiography and the gold standard histopathological examination for proximal caries detection in the primary molars. Twenty-four extracted primary molars with and without proximal caries were divided into three groups and arranged in the mimetic alveolar sockets of a 3D-printed skull and mandible. Two observers evaluated extraoral bitewing images separately twice at one-week intervals. The weighted kappa coefficients showed excellent intra-observer and inter-observer agreements between each session of the extraoral bitewing radiographic assessments. The Mann-Whitney U test showed no difference between the radiographic grading scores of extraoral bitewing images and the gold standard. The sensitivity, positive predictive value (PPV), negative predictive value (NPV), and the area under the receiver operating characteristic (ROC) curve in cavitated carious lesions were higher than in non-cavitated carious lesions. In conclusion, extraoral bitewing is an alternative radiography for pediatric patients who are either uncooperative or intolerant to intraoral radiography.

Accuracy of an intraoral scanner with near-infrared imaging feature in detection of interproximal caries of permanent teeth: An in vivo validation.

This study aimed to evaluate the accuracy of an intraoral scanner with near-infrared imaging (NIRI) feature in the diagnosis of interproximal caries and to compare it with the visual-tactile method (VTM), bitewing radiography (BWR), and panoramic radiography (PR). Six hundred thirty-nine interproximal surfaces (mesial-distal) of posterior teeth from 22 volunteers were examined. Results were scored by VTM, BWR, PR, and NIRI. Lesions were scored as 0 for no-caries, 1 for early-enamel lesion (EEL), and 2 for lesions involving dentino-enamel junction (DEJ). McNemar, Kappa, and Fleis Kappa tests were used to evaluate the agreement levels. Pearson's Chi-square test was used to determine the matching rates after validation. A good level of agreement was observed between examination methods (Ƙ = 0.613; p < 0.001). In pairwise comparisons, a moderate agreement was seen between all the methods for lesions with DEJ involvement, while a statistically good agreement was observed between BWR and NIRI (Ƙ = 0.675; p < 0.001). As a result of validation, the accuracy of NIRI for molars was considered 85.2% and 75.7% for premolars in EELs, 85.2% for molars, and 70% for premolars regarding the lesions involving DEJ. Intraoral scanners with the NIRI feature may be used for diagnosing interproximal caries, especially for permanent molars. Early detection of proximal caries is one of the most essential topics forming the basis of preventive dentistry. This study investigates a caries diagnostic tool integrated into intraoral scanners to diagnose interproximal caries. A caries diagnostic tool integrated into an intraoral scanner may prevent the harmful effects of ionizing radiation in early caries diagnosis and may improve the patient's oral health status.

NEAR-INFRARED TRANSILLUMINATION COMPARED TO DIGITAL BITEWING RADIOGRAPHY FOR PROXIMAL CARIES DETECTION

Proximal caries is a diagnostic and therapeutic challenge. Aim: to study the diagnostic accuracy of near-infrared transillumination (DIAGNOcam, KaVo) compared to digital bitewing radiography for initial proximal caries diagnosis. Materials and methods: 1600 proximal surfaces of premolars and molars of 50 patients were examined with DIAGNOcam and digital bitewing radiography, and the level of agreement between them was established. Results: for the diagnosis of proximal caries in dentin, a complete (100%) coincidence of the findings with the type of tooth and type of surface was established. For the diagnosis of proximal caries in enamel, a high level of agreement between the two methods was found, as transillumination with near-infrared light is more sensitive than digital bitewing radiography. Conclusion: DIAGNOcam is a promising screening device in clinical practice for early proximal caries detection.

Proximal Caries Detection Using Short-Wave Infrared Transillumination at Wavelengths of 1050, 1200 and 1300 nm in Permanent Posterior Human Teeth.

This in vitro study aimed to investigate the diagnostic potential of short-wave infrared transillumination (SWIRT) at 1050, 1200 and 1300 nm for the detection of proximal caries in molars and premolars. It was compared to the diagnostic performance of bitewing radiography (BWR) and micro-computed tomography (µCT) as the reference standard. 250 sound or decayed proximal surfaces of permanent posterior extracted teeth were examined using (1) SWIRT at 1050, 1200 and 1300 nm with two camera systems of different resolutions, (2) BWR and (3) µCT. Thresholds were defined for both test methods and the reference standard for caries in general, enamel caries and dentin caries. All images were assessed by two examiners twice, at an interval of two weeks. SWIRT at wavelengths of 1050, 1200 and 1300 nm achieved sensitivity values more than 2.5 times higher than BWR (enamel caries 3.2-4.4 times; dentin caries 3.25-4.25 times) for the detection of proximal caries. Sensitivity values of SWIRT improved with the higher wavelength. No significant difference was found in diagnostic quality between the two camera systems. SWIRT at 1300 nm imaged proximal enamel caries with the highest accuracy, while the physical optimum for transillumination in dentin was located at a lower wavelength (<1000 nm).

Assessment of the Object Detection Ability of Interproximal Caries on Primary Teeth in Periapical Radiographs Using Deep Learning Algorithms

The purpose of this study was to evaluate the performance of a model using You Only Look Once (YOLO) for object detection of proximal caries in periapical radiographs of children. A total of 2016 periapical radiographs in primary dentition were selected from the M6 database as a learning material group, of which 1143 were labeled as proximal caries by an experienced dentist using an annotation tool. After converting the annotations into a training dataset, YOLO was trained on the dataset using a single convolutional neural network (CNN) model. Accuracy, recall, specificity, precision, negative predictive value (NPV), F1-score, Precision-Recall curve, and AP (area under curve) were calculated for evaluation of the object detection model’s performance in the 187 test datasets. The results showed that the CNN-based object detection model performed well in detecting proximal caries, with a diagnostic accuracy of 0.95, a recall of 0.94, a specificity of 0.97, a precision of 0.82, a NPV of 0.96, and an F1-score of 0.81. The AP was 0.83. This model could be a valuable tool for dentists in detecting carious lesions in periapical radiographs.

Clinical validation of near-infrared imaging for early detection of proximal caries in primary molars

ObjectivesThe aim of this study was to validate the near-infrared imaging (NIRI) in comparison with visual inspection (VI) for early detection of proximal caries in primary molars. MethodsVI and intraoral scans were performed on 126 patients aged 3–12 years with at least one non-cavitied and non-restored proximal tooth surface, who were scheduled for bite wing radiography (BWR) as part of their standard care. Teeth with signs of proximal cavities, restorations or residual caries were excluded in this study. BWR, a gold standard to diagnose proximal caries in primary molars, was used to validate the findings of NIRI and VI. The accuracy, sensitivity, specificity and the area under the curve (AUC) of NIRI and VI were calculated. ResultsThe accuracy, sensitivity and specificity of NIRI were 82.89%, 74.10% and 90.97%, while those of VI were 71.64%, 43.88% and 97.14%, respectively. NIRI showed higher accuracy and sensitivity, and lower specificity (P < 0.001). The AUC of NIRI was higher than that of VI (0.826 vs 0.706; P < 0.05). ConclusionsNIRI showed higher sensitivity and lower specificity compared with VI when detecting proximal caries in primary molars. Therefore, it is recommended to use NIRI in combination with BWR to improve the detection rate of proximal caries in primary molars. Clinical significanceIn children, there is a high incidence of proximal caries in primary molars, which require high technical sensitivity for detection. NIRI shows high sensitivity in detecting proximal caries, which may improve their detection rate in primary molars.The clinical trial registration number: ChiCTR2300070916

Evaluation of diagnostic accuracy of CBCT and intraoral radiography for proximal caries detection in the presence of different dental restoration materials

PurposeThis study aimed to assess the diagnostic accuracy of cone-beam computed tomography (CBCT) and digital intraoral radiography for the detection of proximal caries adjacent to amalgam, e.max porcelain, and metal-ceramic restorations (MCRs).Materials and methodsParallel intraoral radiographs were obtained from 40 posterior teeth using PSP sensors. To obtain CBCT scans, the teeth were first radiographed alone, and were then positioned next to a tooth with an amalgam restoration, MCR, and e.max porcelain crown, and radiographed again. Two blinded observers scored radiographs using a four-point scale (0: absence of proximal caries, 1: enamel caries, 2: carious lesion extending to the outer half of dentin, 3: carious lesion extending to the inner half of dentin). Tooth sections were made, and the grade of caries was determined under a light microscope at x12 magnification. The sensitivity, specificity, and accuracy of CBCT and intraoral radiographs were then calculated.ResultsArtifact-free CBCT scans and intraoral radiographs had the highest diagnostic accuracy (0.826 and 0.657, respectively) while CBCT images of the teeth next to the amalgam restorations (0.526) had the lowest accuracy. The diagnostic accuracy of CBCT images of the teeth next to the porcelain crowns and MCRs was 0.613 and 0.601, respectively.ConclusionArtifact-free CBCT images had higher diagnostic accuracy than intraoral radiography for the detection of all grades of proximal caries. The diagnostic accuracy of CBCT images of teeth adjacent to amalgam, porcelain, and MCRs was lower compared to intraoral radiographs and artifact-free CBCT images.

Accuracy of high definition near infrared transillumination camera in detection of hidden proximal caries.

Early caries detection became mandatory in modern dentistry. However, the traditional methods in caries detection had many limitations. Hence,a novel approach based on Near Infrared technology was introduced to overcome such limitations. Proximal surfaces of 102 posterior teeth from 36 adult participants who fulfilled the eligibility criteria were assessed by two examiners using three diagnostic methods. Teeth were examined visually according to the criteria of the International Caries Detection and Assessment System (ICDAS-II) then examined by bitewing digital radiograph (BW) and near infrared light transillumination (NIRT) camera (Vista Proxi iX HD smart). Inter and intra observer agreements were assessed using Kappa test.Dignostic accuracy parameters and Area Under the ROC curve (AUROC) with 95% confidence interval (95% CI) were evaluated for the different caries assessment methods. The results of inter-observer agreement showed an excellent agreement in the different groups. There was a statistically significant difference in the score distribution between ICDASII and VistaCam modalities (P-value <0.05). While there was no statistically significant difference in the score distribution between bitewing radiography and VistaCam modalities (P-value >0.05). ROC curve analysis revealed that VistaCam when compared with ICDASII had sensitivity (99.0%), specificity (50.0 %), diagnostic accuracy (98.0%) and Area under the ROC curve (AUC) was 0.745 with 95% Confidence Interval (0.649 - 0.826).When VistaCam compared with bitewing radiography, it showed sensitivity (100.0%); specificity (40.0%), diagnostic accuracy (97.1%) and AUC (0.700) with 95% confidence interval (0.601 - 0.787). NIRT based diagnostic modality is a promising method for detection of hidden proximal lesions overcoming the hazards of radiograph. Key words:Bitewing radiography , ICDAS II , Near-infrared transillumination, proximal caries, VistaCam® iX Proxi.

Evaluation of Diagnostic Efficacy of Improved Interproximal Program and Bitewing Program in Proximal Caries Detection- An Extraoral Radiographic Comparative Study

Background: To look for an alternative to intraoral bitewing radiographs when the intraoral placement of film is not feasible or multiple teeth have to be screened. Objectives: A study to evaluate and compare the efficacy of improved interproximal and bitewing programs for detection of proximal caries Planmeca ProMax was conducted. Materials and Methods: Fifty patients who were screened for proximal caries and confirmed with intraoral bitewing radiography were subjected to improved interproximal program and bitewing program in Planmeca ProMax (Planmeca Oy, Helsinki, Finland) after obtaining informed consent. These radiographs were evaluated by three observers. Intra- and interclass correlation was done to check reliability of the observers, and ROC analysis was done to evaluate the diagnostic accuracy. Critical ratio analysis was done to know the significance. Results: Both improved interproximal program and bitewing programs have similar diagnostic efficacy. ROC curve area of improved interproximal program was 0.833 while bitewing program showed 0.878. Bitewing program showed specificity value of 76% compared to 67% in improved interproximal program while both were equally sensitive. Conclusion: Both bitewing program and improved interproximal program are highly sensitive in detecting proximal caries when compared to intraoral bitewing radiography which is still considered the gold standard in diagnosis of an interproximal caries.

Detection of Proximal Caries Lesions with Deep Learning Algorithm

This study aimed to evaluate the effectiveness of deep convolutional neural networks (CNNs) for diagnosis of interproximal caries in pediatric intraoral radiographs. A total of 500 intraoral radiographic images of first and second primary molars were used for the study. A CNN model (Resnet 50) was applied for the detection of proximal caries. The diagnostic accuracy, sensitivity, specificity, receiver operating characteristic (ROC) curve, and area under ROC curve (AUC) were calculated on the test dataset. The diagnostic accuracy was 0.84, sensitivity was 0.74, and specificity was 0.94. The trained CNN algorithm achieved AUC of 0.86. The diagnostic CNN model for pediatric intraoral radiographs showed good performance with high accuracy. Deep learning can assist dentists in diagnosis of proximal caries lesions in pediatric intraoral radiographs.

Detecting Proximal Caries on Periapical Radiographs Using Convolutional Neural Networks with Different Training Strategies on Small Datasets.

The present study aimed to evaluate the performance of convolutional neural networks (CNNs) that were trained with small datasets using different strategies in the detection of proximal caries at different levels of severity on periapical radiographs. Small datasets containing 800 periapical radiographs were randomly categorized into a training and validation dataset (n = 600) and a test dataset (n = 200). A pretrained Cifar-10Net CNN was used in the present study. Different training strategies were used to train the CNN model independently; these strategies were defined as image recognition (IR), edge extraction (EE), and image segmentation (IS). Different metrics, such as sensitivity and area under the receiver operating characteristic curve (AUC), for the trained CNN and human observers were analysed to evaluate the performance in detecting proximal caries. IR, EE, and IS recognition modes and human eyes achieved AUCs of 0.805, 0.860, 0.549, and 0.767, respectively, with the EE recognition mode having the highest values (p all < 0.05). The EE recognition mode was significantly more sensitive in detecting both enamel and dentin caries than human eyes (p all < 0.05). The CNN trained with the EE strategy, the best performer in the present study, showed potential utility in detecting proximal caries on periapical radiographs when using small datasets.

Cost-effectiveness of AI for caries detection: randomized trial

Objectives: We assessed the cost-effectiveness of AI-supported detection of proximal caries in a randomized controlled clustered cross-over superiority trial.Methods: Twenty-three dentists were sampled to assess 20 bitewings; 10 were randomly evaluated supported by an AI-based software (dentalXrai Pro 1.0.4, dentalXrai Ltd, Berlin, Germany) and the other 10 without AI support. The reference test had been established by four independent experts and an additional review. We evaluated the proportion of true and false positive and negative detections and the treatment decisions assigned to each detection (non-invasive, micro-invasive, invasive). Cost-effectiveness was assessed using a mixed public-private-payer perspective in German healthcare. Using the accuracy and treatment decision data from the trial, a Markov simulation model was populated and posterior permanent teeth in initially 31-years old individuals followed over their lifetime. The model allowed extrapolation from the initial detection and therapy to treatment success, re-treatments and, eventually, tooth loss and replacement, capturing long-term effectiveness (tooth retention) and costs (cumulative in Euro). Costs were estimated using the German public and private fee catalogues. Monte-Carlo microsimulations were used and incremental cost-effectiveness at different willingness-to-pay ceiling thresholds assessed.Results: In the trial, AI-supported detection was significantly more sensitive than detection without AI. However, in the AI group, lesions were more often treated invasively. As a result, AI and no AI showed identical effectiveness (tooth retention for a mean (2.5-97.5%) 49 (48-51) years) and nearly identical costs (AI: 330 (250-409) Euro, no AI: 330 (248-410) Euro). 41% simulations found AI and 43% no AI to be more cost-effective. The resulting cost-effectiveness remained uncertain regardless of a payer's willingness-to-pay.Conclusions: Higher accuracy of AI did not lead to higher cost-effectiveness, as more invasive treatment approaches generated costs and diminished possible effectiveness advantages.Clinical significance: The cost-effectiveness of AI could be improved by supporting not only caries detection, but also subsequent management.

Reflected near-infrared light versus bite-wing radiography for the detection of proximal caries: A multicenter prospective clinical study conducted in private practices

ObjectivesThe aim of the present prospective multicenter clinical study was to compare the detection of proximal caries with near-infrared light reflection (NILR) versus bitewing radiography (BWR). Materials and methodsIntraoral scans were performed on 100 patients in five dental clinics using an intraoral scanner (iTero Element 5D, Align Technology, Tempe, AZ, USA) that includes a near-infrared light source (850 nm) and sensor. Reflected near-infrared light images of posterior teeth were used by the individual dentists to detect proximal caries and the results were compared to the BWRs. In a total of 3499 proximal surfaces of molars and premolars which were examined, 223 carious lesions were detected by BWR, while NILR detected 549 carious lesions. Caries detection using both methods was also done by an expert team of five dentists, highly experienced in NILR image interpretation, who used the same sets of clinically-obtained data. Sensitivity, specificity, and accuracy were calculated for caries detection by both the dentists and the expert team. Fifty-nine of the detected carious lesions were clinically treated and the observations during caries excavation were compared with those done with NILR and BWR. Statistical analysis to compare between NILR and BWR diagnosis was performed using non-parametric two-sided McNemar's Chi-Square test with the significance level set at p < 0.05. Kappa coefficients were calculated to assess the level of agreement between the two caries detection methods. ResultsAccuracy of NILR detection of early enamel lesions was 88% and that of carious lesions involving the dentino-enamel junction (DEJ) was 97%. Accuracy was found to be higher at 96% and 99%, respectively, when the same data were examined by the expert team. Direct observation during caries-excavation treatment suggested that NILR detected early enamel lesions that were not detectable with BWR alone. ConclusionsWithin the limitations of the present study, NILR was more sensitive than BWR in detecting early enamel lesions and comparable to BWR in detecting lesions that involved the DEJ. Clinical relevanceReflected near-infrared light images that are generated simultaneously with 3D intra-oral scanning may be used reliably for detection, screening, and monitoring of proximal caries, thus potentially minimizing the traditional use of ionizing radiation.

Artificial intelligence for caries detection: Randomized trial

Objectives: We aimed to assess the impact of an artificial intelligence (AI)-based diagnostic-support software for proximal caries detection on bitewing radiographs.Methods: A cluster-randomized cross-over controlled trial was conducted. A commercially available software employing a fully convolutional neural network for caries detection (dentalXrai Pro, dentalXrai Ltd.) was randomly employed by 22 dentists, supporting their caries detection on 20 bitewings randomly chosen from a pool of 140 bitewings, with 10 bitewings randomly being supported by AI and 10 not. The reference test had been established by 4 + 1 independent experts in a pixelwise fashion. Caries was subgrouped as enamel, early dentin and advanced dentin caries, and accuracy and treatment decisions for each caries lesion assessed.Results: Dentists with AI showed a significantly higher mean (95% CI) area under the Receiver-Operating-Characteristics curve (0.89; 0.87–0.90) than those without AI (0.85; 0.83–0.86; p<0.05), mainly as their sensitivity was significantly higher (0.81; 0.74–0.87 compared with 0.72; 0.64–0.79; p<0.05) while the specificity was not significantly affected (p>0.05). This increase in sensitivity was found for enamel, but not early or advanced dentin lesions. Higher sensitivity came with an increase in non-invasive, but also invasive treatment decisions (p<0.05).Conclusion: AI can increase dentists’ diagnostic accuracy but may also increase invasive treatment decisions.Clinical significance: AI can increase dentists’ diagnostic accuracy, mainly via increasing their sensitivity for detecting enamel lesions, but may also increase invasive therapy decisions. Differences in the effects of AI for different dentists should be explored, and dentists should be guided as to which therapy to choose when detecting caries lesions using AI support.