Systematic review and meta-analysis of diagnostic methods for occlusal surface caries

Kapor, Svetlana; Ranković, Mila Janjić; Khazaei, Yeganeh; Crispin, Alexander; Schüler, Ina Manuela; Krause, Felix; Lussi, Adrian; Neuhaus, Klaus W.; Eggmann, Florin; Michou, Stavroula; Ekstrand, Kim R.; Huysmans, M.C.D.N.J.M.; Kühnisch, Jan

doi:10.1007/s00784-021-04024-1

Cited by 23 publications

(22 citation statements)

References 83 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…All the images were labeled by an experienced examiner (JK, >20 y of clinical practice and scientific experience), who was also aware of the patients’ history and overall dental status, into the following categories: 0, surfaces with no caries; 1, surfaces with signsof a noncavitated caries lesion (first signs, established lesion, localized enamel breakdown); and 2, surfaces with caries-related cavitation (dentin exposure, large cavity). Both caries thresholds are of clinical relevance and also commonly used in caries diagnostic studies ( Schwendicke, Splieth, et al 2019 ; Kapor et al 2021 ). Each diagnostic decision—1 per image—served as a reference standard for cyclic training and repeated evaluation of the deep learning–based CNN.…”

Section: Methodsmentioning

confidence: 99%

Caries Detection on Intraoral Images Using Artificial Intelligence

et al. 2021

Self Cite

View full text Add to dashboard Cite

Although visual examination (VE) is the preferred method for caries detection, the analysis of intraoral digital photographs in machine-readable form can be considered equivalent to VE. While photographic images are rarely used in clinical practice for diagnostic purposes, they are the fundamental requirement for automated image analysis when using artificial intelligence (AI) methods. Considering that AI has not been used for automatic caries detection on intraoral images so far, this diagnostic study aimed to develop a deep learning approach with convolutional neural networks (CNNs) for caries detection and categorization (test method) and to compare the diagnostic performance with respect to expert standards. The study material consisted of 2,417 anonymized photographs from permanent teeth with 1,317 occlusal and 1,100 smooth surfaces. All the images were evaluated into the following categories: caries free, noncavitated caries lesion, or caries-related cavitation. Each expert diagnosis served as a reference standard for cyclic training and repeated evaluation of the AI methods. The CNN was trained using image augmentation and transfer learning. Before training, the entire image set was divided into a training and test set. Validation was conducted by selecting 25%, 50%, 75%, and 100% of the available images from the training set. The statistical analysis included calculations of the sensitivity (SE), specificity (SP), and area under the receiver operating characteristic (ROC) curve (AUC). The CNN was able to correctly detect caries in 92.5% of cases when all test images were considered (SE, 89.6; SP, 94.3; AUC, 0.964). If the threshold of caries-related cavitation was chosen, 93.3% of all tooth surfaces were correctly classified (SE, 95.7; SP, 81.5; AUC, 0.955). It can be concluded that it was possible to achieve more than 90% agreement in caries detection using the AI method with standardized, single-tooth photographs. Nevertheless, the current approach needs further improvement.

show abstract

Section: Methodsmentioning

confidence: 99%

Caries Detection on Intraoral Images Using Artificial Intelligence

et al. 2021

Self Cite

View full text Add to dashboard Cite

show abstract

“…Most of the previous systematic reviews and meta-analyses about technologies in caries detection have focused on traditional and commonly used non-traditional technologies [25][26][27]29,30]. They did not address or study new methods, such as the Midwest Caries I.D.…”

Section: Discussionmentioning

confidence: 99%

“…The emerging technologies in this review had few identified studies that met the inclusion criteria. Although recent systematic reviews and meta-analyses overlooked these technologies probably due to the low number of available studies [29,30], their diagnostic potential should be acknowledged. We systematically identified these studies, collected the performance data, synthesized them and then put them in context, irrespective of the reports' number; this contribution might help future investigators to build up on existing body of evidence.…”

Section: Newly Emerging Technologiesmentioning

confidence: 99%

“…Previous studies have evaluated the performance of emerging technologies in detecting and quantifying carious lesions [25][26][27][28][29][30], but results from in vivo and in vitro studies have been heterogeneous and there is ambiguity regarding their overall effectiveness as diagnostic methods in clinical practice. Recent systematic reviews have investigated the most commonly used caries detection methods for occlusal and proximal studies, but they either overlooked less studied or reported methods [29,30], such as near-infrared light transillumination and alternating current impedance spectroscopy, or included only in vivo studies [28].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Emerging Technologies for Dentin Caries Detection—A Systematic Review and Meta-Analysis

Serban

Lungeanu

Bota

et al. 2022

JCM

View full text Add to dashboard Cite

This systematic review and meta-analysis aimed at assessing the diagnostic accuracy of emerging technologies, such as laser fluorescence (LF), transillumination, light-emitting diode devices, optical coherence tomography (OCT), alternating current impedance spectroscopy, fluorescence cameras (FC), photo-thermal radiometry, and modulated luminescence technology. In vivo and in vitro results of such non-ionizing, non-invasive, and non-destructive methods’ effectiveness in non-cavitated dentin caries detection are sometimes ambiguous. Following the PRISMA guidelines, 34 relevant research articles published between 2011‒2021 were selected. The risk of bias was assessed with a tool tailored for caries diagnostic studies, and subsequent quantitative uni- and bi-variate meta-analysis was carried out in separate sub-groups according to the investigated surface (occlusal/proximal) and study setting (in vivo/in vitro). In spite of the high heterogeneity across the review groups, in vitro studies on LF and FC proved a good diagnostic ability for the occlusal surface, with area under the curve (AUC) of 0.803 (11 studies) and 0.845 (five studies), respectively. OCT studies reported an outstanding performance with an overall AUC = 0.945 (four studies). Promising technologies, such as OCT or FC VistaProof, still need well-designed and well-powered studies to accrue experimental and clinical data for conclusive medical evidence, especially for the proximal surface. Registration: INPLASY202210097.

show abstract

“…A number of diagnostic approaches are available to assist clinicians in caries detection and diagnosis [ 7 , 8 ]. Studies evaluating the accuracy of these approaches, however, primarily focus on coronally located caries lesions.…”

Section: Introductionmentioning

confidence: 99%

Accuracy of different approaches for detecting proximal root caries lesions in vitro

et al. 2022

View full text Add to dashboard Cite

Objectives The objective was to evaluate the diagnostic accuracy of radiographic evaluation (XR), visual-tactile assessment (VT), laser-fluorescence (LF) (DIAGNOdent Pen/KaVo), and near-infrared-light transillumination (NILT) (DIAGNOcam/KaVo) on proximal root caries lesions in vitro. Methods Two-hundred extracted permanent premolars and molars with and without proximal root caries lesions were allocated to 50 diagnostic models simulating the proximal contacts between teeth and mounted in a phantom dummy head. Two independent examiners used the diagnostic approaches to detect any or advanced root caries lesions, with histologic evaluation of the lesions serving as reference. Receiver operating characteristic (ROC) curves were employed, and sensitivity, specificity, and the area under the ROC curve (AUC) are calculated. Significant differences in mean AUCs between approaches were assumed if p < 0.05 (two-sample t-test). Results NILT was not applicable for proximal root caries detection. The sensitivity/specificity to detect any lesions was 0.81/0.63 for XR, 0.76/0.88 for VT and 0.81/0.95 for LF, and the sensitivity/specificity to detect advanced lesions was 0.43/0.94 for XR, 0.66/0.99 for VT, and 0.83/0.78 for LF, respectively. For both, any and advanced root caries lesions, mean AUCs for LF and VT were significantly higher compared to XR (p < 0.05). For any root caries lesions, LF was significantly more accurate than VT (p = 0.01), but not for advanced root caries lesions (p = 0.59). Conclusions Under the in vitro conditions chosen, LF and VT were more accurate than XR to detect proximal root caries lesions, with LF being particularly useful for initial lesion stages. Clinical relevance LF might be a useful diagnostic aid for proximal root caries diagnosis. Clinical studies are necessary to corroborate the findings.

show abstract

Systematic review and meta-analysis of diagnostic methods for occlusal surface caries

Cited by 23 publications

References 83 publications

Caries Detection on Intraoral Images Using Artificial Intelligence

Caries Detection on Intraoral Images Using Artificial Intelligence

Emerging Technologies for Dentin Caries Detection—A Systematic Review and Meta-Analysis

Accuracy of different approaches for detecting proximal root caries lesions in vitro

Contact Info

Product

Resources

About