ArticleBMC oral health2026
Deep learning-based detection of the second mesiobuccal canal in maxillary first molars using cone-beam computed tomography.
Article in BMC oral health, 2026. The graph could read no effect estimate from its abstract, so it casts no vote on the map. Not yet cited in PubMed.
What it found
Each row is one number read from the abstract, on the scale the paper reported it, with its interval. Left of the dashed line favours the treatment, right favours the comparator. Under each row is the sentence it came from. New to these charts? A ten-minute tutorial.
The abstract states no effect estimate the extractor could read, or names no intervention and outcome on the map, so this paper lights no cell and moves no belief. It is still indexed, cited and linked below.
The trial behind it
Trials whose registry record cites this paper, or whose number appears in the abstract. A trial that started after this paper was published is citing it as background, not reporting it.
Neither the registry nor the abstract names a trial number. If this is a trial report, that itself is worth knowing.
Who cites it
0 citing papers in PubMed.
No citing paper in PubMed yet.
Corrections and comments
PubMed lists nothing against this paper. Absence here is not a guarantee, only a check that was made.
Authors and funding
7 authors.
Funding
Abstract
backgroundMissed detection of the second mesiobuccal canal (MB2) in maxillary first molars is a recognized contributor to endodontic failure. Although cone-beam computed tomography (CBCT) enhances three-dimensional visualization of root canal anatomy, image interpretation remains observer-dependent and variable.
objectiveTo develop and validate a deep learning (DL) model for automated detection of MB2 canals on CBCT images and to evaluate its impact on the diagnostic performance of junior dentists.
methodsCBCT scans from 1,022 patients were annotated by expert consensus using comprehensive volumetric evaluation. A two-stage DL framework—region-of-interest localization using a modified VGG16 network followed by classification with a ConvNeXt–Transformer architecture-was trained on 1,944 ROIs and evaluated on an independent test set of 100 ROIs (50 MB2-positive, 50 MB2-negative). Three junior dentists assessed the test set with and without DL assistance.
resultsThe proposed model achieved a sensitivity of 0.84, specificity of 0.86, positive predictive value of 0.86, negative predictive value of 0.84, F1-score of 0.85, and an area under the receiver operating characteristic curve (AUC) of 0.85, outperforming benchmark CNN architectures. DL assistance increased junior dentists’ F1-scores (0.78–0.82 manually vs. 0.82–0.88 with assistance) and improved inter-reader agreement (Fleiss’ κ: 0.730 to 0.880).
conclusionsThe proposed DL framework demonstrated balanced diagnostic performance for MB2 detection and enhanced diagnostic consistency among junior clinicians. Deep learning may serve as an adjunctive decision-support tool in endodontic imaging and education. Further multicenter validation is warranted to establish generalizability and clinical applicability.
Indexed as
Identifiers
What OpenQuestion holds
Registered trials
Read under generation 80e0d062 · epoch 390. Bibliography from PubMed, PubMed Central and OpenAlex; grants from NIH RePORTER; trial links from ClinicalTrials.gov; estimates, votes and beliefs from the OpenQuestion graph.