TY - JOUR
T1 - Deep Learning for Caries Detection
T2 - A Systematic Review
AU - Mohammad-rahimi, Hossein
AU - Motamedian, Saeed Reza
AU - Rohban, Mohammad Hossein
AU - Krois, Joachim
AU - Uribe, Sergio
AU - Nia, Erfan Mahmoudi
AU - Rokhshad, Rata
AU - Nadimi, Mohadeseh
AU - Schwendicke, Falk
N1 - Publisher Copyright:
© 2022 Elsevier Ltd
PY - 2022/7
Y1 - 2022/7
N2 - ObjectivesDetecting caries lesions is challenging for dentists, and deep learning models may help practitioners to increase accuracy and reliability. We aimed to systematically review deep learning studies on caries detection.DataWe selected diagnostic accuracy studies that used deep learning models on dental imagery (including radiographs, photographs, optical coherence tomography images, near-infrared light transillumination images). The latest version of the quality assessment tool for diagnostic accuracy studies (QUADAS-2) tool was used for risk of bias assessment. Meta-analysis was not performed due to heterogeneity in the studies methods and their performance measurements.SourcesDatabases (Medline via PubMed, Google Scholar, Scopus, Embase) and a repository (ArXiv) were screened for publications published after 2010, without any limitation on language.Study selectionFrom 252 potentially eligible references, 48 studies were assessed full-text and 42 included, using classification (n=26), object detection (n=6), or segmentation models (n=10). A wide range of performance metrics was used; image, object or pixel accuracy ranged between 68%-99%. The minority of studies (n=11) showed a low risk of biases in all domains, and 13 studies (31.0%) low risk for concerns regarding applicability. The accuracy of caries classification models varied, i.e. 71% to 96% on intra-oral photographs, 82% to 99.2% on peri-apical radiographs, 87.6% to 95.4% on bitewing radiographs, 68.0% to 78.0% on near-infrared transillumination images, 88.7% to 95.2% on optical coherence tomography images, and 86.1% to 96.1% on panoramic radiographs. Pooled diagnostic odds ratios varied from 2.27 to 32767. For detection and segmentation models, heterogeneity in reporting did not allow useful pooling.ConclusionAn increasing number of studies investigated caries detection using deep learning, with a diverse types of architectures being employed. Reported accuracy seems promising, while study and reporting quality are currently low.Clinical significanceDeep learning models can be considered as an assistant for decisions regarding the presence or absence of carious lesions.
AB - ObjectivesDetecting caries lesions is challenging for dentists, and deep learning models may help practitioners to increase accuracy and reliability. We aimed to systematically review deep learning studies on caries detection.DataWe selected diagnostic accuracy studies that used deep learning models on dental imagery (including radiographs, photographs, optical coherence tomography images, near-infrared light transillumination images). The latest version of the quality assessment tool for diagnostic accuracy studies (QUADAS-2) tool was used for risk of bias assessment. Meta-analysis was not performed due to heterogeneity in the studies methods and their performance measurements.SourcesDatabases (Medline via PubMed, Google Scholar, Scopus, Embase) and a repository (ArXiv) were screened for publications published after 2010, without any limitation on language.Study selectionFrom 252 potentially eligible references, 48 studies were assessed full-text and 42 included, using classification (n=26), object detection (n=6), or segmentation models (n=10). A wide range of performance metrics was used; image, object or pixel accuracy ranged between 68%-99%. The minority of studies (n=11) showed a low risk of biases in all domains, and 13 studies (31.0%) low risk for concerns regarding applicability. The accuracy of caries classification models varied, i.e. 71% to 96% on intra-oral photographs, 82% to 99.2% on peri-apical radiographs, 87.6% to 95.4% on bitewing radiographs, 68.0% to 78.0% on near-infrared transillumination images, 88.7% to 95.2% on optical coherence tomography images, and 86.1% to 96.1% on panoramic radiographs. Pooled diagnostic odds ratios varied from 2.27 to 32767. For detection and segmentation models, heterogeneity in reporting did not allow useful pooling.ConclusionAn increasing number of studies investigated caries detection using deep learning, with a diverse types of architectures being employed. Reported accuracy seems promising, while study and reporting quality are currently low.Clinical significanceDeep learning models can be considered as an assistant for decisions regarding the presence or absence of carious lesions.
KW - Artificial intelligence
KW - Machine learning
KW - Neural Networks
KW - Dental Caries
KW - Dentistry
KW - Systematic Review
UR - http://www.scopus.com/inward/record.url?scp=85131701962&partnerID=8YFLogxK
U2 - 10.1016/j.jdent.2022.104115
DO - 10.1016/j.jdent.2022.104115
M3 - Review article
SN - 0300-5712
VL - 122
JO - Journal of Dentistry
JF - Journal of Dentistry
M1 - 104115
ER -