OralHybridNet: A Deep Learning Framework for Multi-Label Classification of Dental Restorations and Prostheses in Panoramic Radiographs
INQUIRY: The Journal of Health Care Organization, Provision, and Financing · 2026-03 · 63
DOI: 10.1177/00469580261439986 ↗Kurz gefasst
Fachartikel in der Zeitschrift INQUIRY: The Journal of Health Care Organization, Provision, and Financing, online erschienen am 13.04.2026. Verfasst von Zohaib Khurshid und 5 weiteren Autor:innen. Beteiligte Einrichtung: Department of Prosthodontics and Dental Implantology, College of Dentistry, King Faisal University, Al-Ahsa, Saudi Arabia. Hauptthemen laut MEDLINE: Deep Learning, Radiography, Panoramic, Dental Restoration, Permanent, Dental Prosthesis. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 33 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: Deanship of Scientific Research, King Faisal University.
Abstract (Originalsprache)
Automated prediction of dental conditions in Orthopantomogram (OPG) panoramic radiographs faces significant challenges due to class imbalance, rare pathologies, and complex anatomical structures. This study proposes OralHybridNet, a novel hybrid deep learning framework integrating hierarchical convolutional neural networks, such as CustomDentalNet integrate dual-attention mechanisms and OralNetXPlus. A multinational dataset comprising 2047 clinician-annotated panoramic radiographs spanning 7 diagnostic labels was used. An adaptive augmentation protocol combining Elastic Transformations and gamma correction mitigated class imbalance. A Hybrid Feature Selection (HFS) algorithm condensed 1208-dimensional embeddings into a discriminative 300-feature subset. Evaluated against ResNet50 baselines, OralHybridNet achieved 96.0% accuracy, 97.6% precision, and 0.993 AUC-ROC. The KNN Fine classifier on fused features yielded the highest performance, with real-time capability (9 ms inference time) using a neural network classifier. The proposed framework demonstrates a promising proof-of-concept framework for automated multi-label dental restoration classification.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Prosthodontics and Dental Implantology, College of Dentistry, King Faisal University, Al-Ahsa, Saudi Arabia.
- Department of Anatomy, Faculty of Dentistry, Center of Artificial Intelligence and Innovation (CAII), Center of Excellence for Dental Stem Cell Biology, Chulalongkorn University, Bangkok, Thailand.
- Center of Excellence in Precision Medicine and Digital Health, Department of Physiology, Faculty of Dentistry, Chulalongkorn University, Bangkok, Thailand.
- University of Zurich, Switzerland.
ORCID-Profile: Zohaib Khurshid, Ramy Moustafa Moustafa Ali, Ali Sulaiman Alharbi
Förderung
- Deanship of Scientific Research, King Faisal University
Zitieren
APA
Khurshid, Z., Moustafa Ali, R. M., Alharbi, A. S., Osathanon, T. N., Alfaraj, A., & Porntaveetus, T. (2026). OralHybridNet: A Deep Learning Framework for Multi-Label Classification of Dental Restorations and Prostheses in Panoramic Radiographs. INQUIRY: The Journal of Health Care Organization, Provision, and Financing, 63. https://doi.org/10.1177/00469580261439986
Vancouver
Khurshid Z, Moustafa Ali RM, Alharbi AS, Osathanon TN, Alfaraj A, Porntaveetus T. OralHybridNet: A Deep Learning Framework for Multi-Label Classification of Dental Restorations and Prostheses in Panoramic Radiographs. INQUIRY: The Journal of Health Care Organization, Provision, and Financing. 2026;63. doi:10.1177/00469580261439986
BibTeX
@article{Khurshid2026,
author = {Khurshid, Zohaib and Moustafa Ali, Ramy Moustafa and Alharbi, Ali Sulaiman and Osathanon, Thanaphum Noom and Alfaraj, Amal and Porntaveetus, Thantrira},
title = {OralHybridNet: A Deep Learning Framework for Multi-Label Classification of Dental Restorations and Prostheses in Panoramic Radiographs},
journal = {INQUIRY: The Journal of Health Care Organization, Provision, and Financing},
year = {2026},
volume = {63},
doi = {10.1177/00469580261439986}
}
EUVIMED