What question did this study set out to answer?

The aim is to develop and assess MultiDentNet, a deep learning framework for identifying various dental conditions and evaluating cancer risk of oral lesions.

May 15, 2026Scientific Reports0 citationsOpen Access

MultiDentNet: a unified deep learning framework for multi-class dental condition screening and preliminary oral lesion triage

AAAli Zeydi Abdian MJMohammad Masoud Javidi NMNajme Mansouri

Puntos clave

The aim is to develop and assess MultiDentNet, a deep learning framework for identifying various dental conditions and evaluating cancer risk of oral lesions.
Evaluated on 10,235 intraoral images for dental conditions and 940 labeled images for lesion risk.
Utilized four CNN backbones combined with attention and graph networks for enhanced performance.
Assessed using metrics like accuracy, Cohen's kappa, and false negative rates.
Achieved 99.70% accuracy (95% CI 99.32–100.00%) for dental classification, with a kappa of 0.996.
For lesion risk stratification, accuracy was 95.71% (95% CI 92.20–98.58%), kappa of 0.913, and recall for cancer-suspicious lesions reached 97.31%.
Significantly reduced false negative rates for Hypodontia to 1.65% through architectural diversity.

Resumen

Abstract To develop and evaluate MultiDentNet, a unified deep learning framework for multi-class dental condition screening and preliminary risk stratification of cancer-suspicious oral lesions, utilizing backbone-diverse ensembling and inter-class relational modeling. Four complementary CNN backbones (DenseNet-121, EfficientNetV2-S, ResNet-50, Inception-V3) integrated with Squeeze-and-Excitation (SE) attention, graph convolutional networks (GCNs), and multi-task learning were fused via validation-optimized weighting. The framework was evaluated on 10, 235 clinical intraoral images (five conditions) and 940 clinically labeled oral lesion images (cancer-suspicious vs. non-cancer-suspicious; histopathological confirmation unavailable). Performance was assessed using accuracy, Cohen’s, Matthews correlation coefficient (MCC), false negative rate (FNR), and bootstrapped 95% confidence intervals (CIs), alongside simulated domain-shift robustness testing. The ensemble achieved 99. 70% accuracy (95% CI 99. 32–100. 00%) for dental classification (= 0. 996) and 95. 71% (95% CI 92. 20–98. 58%) for oral lesion risk stratification (= 0. 913). For cancer-suspicious lesions, recall reached 97. 31% (FNR: 2. 69%, 95% CI 0. 00–7. 15%). Architectural diversity successfully mitigated class imbalance, significantly reducing the Hypodontia FNR to 1. 65% (95% CI 0. 00–4. 24%) compared to single-model baselines. Performance demonstrated moderate resilience to acquisition variability (accuracy 8–12% degradation under brightness and contrast perturbations) but degraded substantially under high-frequency noise. Grad-CAM visualizations localized attention to clinically relevant morphological features. MultiDentNet provides an interpretable, efficient baseline for dental screening and lesion triage. Serving as an adjunctive proof-of-concept, its metrics reflect an upper bound due to reliance on single-center, clinically labeled data. Prospective multi-center validation with histopathological standards remains necessary. Code: https: //github. com/Aliyar4061/MultiDentNetV3. Positioned as an adjunctive triage tool for resource-constrained settings, high-volume workflows, and tele-dentistry, designed to augment clinician-in-the-loop oversight.

Preguntar a la IA

Me gusta

Guardar

Ver artículo completo