View the CAT printer-friendly / share this CAT
Title Artificial Intelligence Software Improves Accuracy of caries Detection by Dentists
Clinical Question For patients with interproximal caries, is artificial intelligence software effective at diagnosing interproximal caries as compared to pediatric dentists using bitewing radiographs?
Clinical Bottom Line Artificial intelligence software can be used effectively as an adjunct tool for diagnosis of interproximal caries by dentists. However, there is no evidence supporting use of artificial intelligence compared to pediatric dentists for interproximal caries detection on bitewing radiographs.
Best Evidence (you may view more info by clicking on the PubMed ID link)
PubMed ID Author / Year Patient Group Study type
(level of evidence)
#1) 34656656Mertens/ 202120 bitewing radiographs for adult teeth taken in hospital based dental clinic in GermanyRandomized Controlled Trial
Key resultsArtificial intelligence (AI) increases dentists' sensitivity to enamel caries but does not markedly increase sensitivity for dentin and deep caries nor the specificity. This study also found that dentists treat enamel lesions more aggressively under AI assistance.
#2) 35626239Khanagar/ 2022Adult teeth images used in 34 selected studies for caries detection in regular dental visits, clinical studies, or extracted teethSystematic review of non-randomized trials
Key resultsArtificial intelligence can be a supportive tool for dentists in caries diagnosis, detection, and prediction. AI can aid in caries risk prediction by increased accuracy and sensitivity to dental caries on radiographic images, intra-oral photos, CBCT, and transillumination. Heterogeneity in study designs and reported outcomes makes meta-analysis impossible.
#3) 35367318Rahimi/ 2022Adult teeth in 42 selected studies for caries detection in regular dental visits, clinical studies, or extracted teethSystematic review of non-randomized trials
Key resultsThere is an uptick in investigations of Deep Learning (DL), a form of artificial intelligence, for caries detection which show increased accuracy of caries diagnosis, with most data available for enamel and dentin caries. However, heterogeneity in study designs, lack of standard reference tests and variety in reported findings make quality of data low.
Evidence Search (artificial intelligence OR machine learning OR deep learning OR neural networks) AND (caries lesion OR dental caries)
Comments on
The Evidence
No evidence is available comparing the diagnostic accuracy of artificial intelligence and pediatric dentists. Available studies do not specify dental specialties of participating dentists. Lack of standardized reference tests, reporting outcomes, data sets and study designs make existing data of low quality.
Applicability AI can enhance caries detection for dentists and non-dental professionals in non-dental or rural settings for early intervention.
Specialty/Discipline (Oral Medicine/Pathology/Radiology) (General Dentistry) (Pediatric Dentistry) (Restorative Dentistry)
Keywords artificial intelligence, deep learning, machine learning, caries diagnosis, proximal caries
ID# 3494
Date of submission: 11/19/2022spacer
E-mail khans7@livemail.uthscsa.edu
Author Shireen Khan
Co-author(s) Brian Ross
Co-author(s) e-mail rossbj@livemail.uthscsa.edu
Faculty mentor/Co-author Dr. Jungyi Alexis Liu
Faculty mentor/Co-author e-mail liuja@uthscsa.edu
Basic Science Rationale
(Mechanisms that may account for and/or explain the clinical question, i.e. is the answer to the clinical question consistent with basic biological, physical and/or behavioral science principles, laws and research?)
post a rationale
None available
Comments and Evidence-Based Updates on the CAT
post a comment
None available

Return to Found CATs list