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.
- MeSH
- interproximální technika MeSH
- lidé MeSH
- náchylnost k zubnímu kazu * MeSH
- zubní kaz * diagnostické zobrazování MeSH
- Check Tag
- lidé MeSH
- Publikační typ
- časopisecké články MeSH
OBJECTIVE: The aim of this work was to assemble a large annotated dataset of bitewing radiographs and to use convolutional neural networks to automate the detection of dental caries in bitewing radiographs with human-level performance. MATERIALS AND METHODS: A dataset of 3989 bitewing radiographs was created, and 7257 carious lesions were annotated using minimal bounding boxes. The dataset was then divided into 3 parts for the training (70%), validation (15%), and testing (15%) of multiple object detection convolutional neural networks (CNN). The tested CNN architectures included YOLOv5, Faster R-CNN, RetinaNet, and EfficientDet. To further improve the detection performance, model ensembling was used, and nested predictions were removed during post-processing. The models were compared in terms of the [Formula: see text] score and average precision (AP) with various thresholds of the intersection over union (IoU). RESULTS: The twelve tested architectures had [Formula: see text] scores of 0.72-0.76. Their performance was improved by ensembling which increased the [Formula: see text] score to 0.79-0.80. The best-performing ensemble detected caries with the precision of 0.83, recall of 0.77, [Formula: see text], and AP of 0.86 at IoU=0.5. Small carious lesions were predicted with slightly lower accuracy (AP 0.82) than medium or large lesions (AP 0.88). CONCLUSIONS: The trained ensemble of object detection CNNs detected caries with satisfactory accuracy and performed at least as well as experienced dentists (see companion paper, Part II). The performance on small lesions was likely limited by inconsistencies in the training dataset. CLINICAL SIGNIFICANCE: Caries can be automatically detected using convolutional neural networks. However, detecting incipient carious lesions remains challenging.
- MeSH
- deep learning * MeSH
- lidé MeSH
- náchylnost k zubnímu kazu MeSH
- neuronové sítě (počítačové) MeSH
- zubní kaz * diagnostické zobrazování MeSH
- Check Tag
- lidé MeSH
- Publikační typ
- časopisecké články MeSH
- MeSH
- čištění zubů metody MeSH
- lidé MeSH
- zubní kaz * diagnostické zobrazování prevence a kontrola terapie MeSH
- Check Tag
- lidé MeSH
- MeSH
- dospělí MeSH
- fluorescence MeSH
- impedanční spektroskopie přístrojové vybavení MeSH
- interproximální technika přístrojové vybavení MeSH
- lidé středního věku MeSH
- lidé MeSH
- mladý dospělý MeSH
- optické zobrazování metody přístrojové vybavení MeSH
- senioři MeSH
- statistika jako téma MeSH
- zubní kaz diagnostické zobrazování MeSH
- Check Tag
- dospělí MeSH
- lidé středního věku MeSH
- lidé MeSH
- mladý dospělý MeSH
- senioři MeSH
- Publikační typ
- srovnávací studie MeSH
- Geografické názvy
- Dánsko MeSH
- MeSH
- dospělí MeSH
- fluorescence MeSH
- fyzikální vyšetření statistika a číselné údaje MeSH
- interproximální technika statistika a číselné údaje MeSH
- lidé středního věku MeSH
- lidé MeSH
- rentgendiagnostika zubní metody statistika a číselné údaje MeSH
- senzitivita a specificita MeSH
- transiluminace statistika a číselné údaje MeSH
- trvalá zubní náhrada metody ošetřování MeSH
- zubní kaz * diagnostické zobrazování MeSH
- Check Tag
- dospělí MeSH
- lidé středního věku MeSH
- lidé MeSH
- mužské pohlaví MeSH
- ženské pohlaví MeSH
- Publikační typ
- práce podpořená grantem MeSH
x
The aim of the study was to compare different methods used for orientation during cavity preparation in dentin. We verified the presence of carious dentin, which should be completely removed during cavity preparation by a) visual inspection b) palpation, c) staining solutions, and d) using a blue light with a color filter. According to the authors' opinion, no softened dentin should be inadvertently left in the cavity before applying the filling. The results show that none of the methods is absolutely reliable on it’s own. Palpation during cavity preparation of carious dentin is considered the standard method, yet is not always complete. Both organic dye and blue light with a color filter showed false positive and false negative results when compared with the tactile examination. On the other hand, these methods helped identify additional areas of softened dentin that went undetected by palpation in less accessible areas. We can therefore declare, that despite the possibility of using several diagnostic methods, preparation of the carious dentin is still a subjective therapeutic intervention. For its greater succes we recommend combining several detection methods.
- MeSH
- barvicí látky MeSH
- dentin patologie MeSH
- falešně pozitivní reakce MeSH
- fluorescence MeSH
- fyzikální vyšetření MeSH
- lidé MeSH
- optické zobrazování MeSH
- porfyriny MeSH
- preparace zubní kavity MeSH
- stomatologické polymerizační lampy využití MeSH
- zubní kaz * diagnostické zobrazování diagnóza MeSH
- Check Tag
- lidé MeSH
- Publikační typ
- srovnávací studie MeSH
