EUVIMEDEUVIMED entdecken
Crossref

Deep learning-based segmentation of the mandibular canals in cone-beam CT reaches human-level performance

Wiebke Semper-Hogg iD, Alexander Rau iD, Marc Anton Fuessinger, Sabrina Zimmermann, Fabian Bamberg iD, Marc Christian Metzger, Rainer Schmelzeisen, Stephan Rau, Marco Reisert, Maximilian Frederik Russe iD

Dentomaxillofacial Radiology · 2025

DOI: 10.1093/dmfr/twae069 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Dentomaxillofacial Radiology, 2025. Verfasst von Wiebke Semper-Hogg und 9 weiteren Autor:innen. Beteiligte Einrichtung: Department of Oral and Maxillofacial Surgery, University Medical Center Freiburg, Faculty of Medicine, University of Freiburg, 79106 Freiburg, German…. Hauptthemen laut MEDLINE: Cone-Beam Computed Tomography, Deep Learning, Mandible. Die Arbeit wurde bisher 8-mal zitiert (Crossref) und nennt 24 Literaturangaben. Gefördert von: Berta-Ottenstein-Programme for Clinician Scientists, Faculty of Medicine, University of Freiburg.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Objectives

This study evaluated the accuracy and reliability of deep learning-based segmentation techniques for mandibular canal identification in cone-beam CT (CBCT) data to provide a reliable and efficient support tool for dental implant treatment planning.

Methods

A dataset of 90 CBCT scans was annotated as ground truth for mandibular canal segmentation. The dataset was split into training (n = 69), validation (n = 1), and testing (n = 20) subsets. A deep learning model based on a hierarchical convolutional neural network architecture was developed and trained. The model’s performance was evaluated using dice similarity coefficient (DSC), 95% Hausdorff distance (HD), and average symmetric surface distance (ASSD). Qualitative assessment was performed by 2 experienced dental imaging practitioners who evaluated the segmentation quality in terms of trust and safety on a 5-point Likert scale at 3 mandibular locations per side.

Results

The trained model achieved a mean DSC of 0.77 ± 0.09, HD of 1.66 ± 0.86 mm, and ASSD of 0.31 ± 0.15 mm on the testing subset. Qualitative assessment showed no significant difference between the deep learning-based segmentation and ground truth in terms of trust and safety across all investigated locations (P > 0.05).

Conclusions

The proposed deep learning-based segmentation technique exhibits sufficient accuracy for the reliable identification of mandibular canals in CBCT scans. This automated approach could streamline the pre-operative planning process for dental implant placement, reducing the risk of neurovascular complications and enhancing patient safety.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Cone-Beam Computed TomographyDeep LearningMandibleHumansReproducibility of ResultsPatient Care PlanningNeural Networks, Computer

Stichwörter der Autor:innen

cone-beam CTdeep learningmandibular canalsegmentation

Autor:innen und Institutionen

  • Department of Oral and Maxillofacial Surgery, University Medical Center Freiburg, Faculty of Medicine, University of Freiburg, 79106 Freiburg, Germany.
  • Department of Neuroradiology, University Medical Center Freiburg, Faculty of Medicine, University of Freiburg, 79106 Freiburg, Germany.
  • Department of Diagnostic and Interventional Radiology, University Medical Center Freiburg, Faculty of Medicine, University of Freiburg, 79106 Freiburg, Germany.
  • Medical Physics, Department of Diagnostic and Interventional Radiology, University Medical Center Freiburg, Faculty of Medicine, University of Freiburg, 79106 Freiburg, Germany.
  • Department of Stereotactic and Functional Neurosurgery, University Medical Center Freiburg, Faculty of Medicine, University of Freiburg, 79106 Freiburg, Germany.

ORCID-Profile: Wiebke Semper-Hogg, Alexander Rau, Fabian Bamberg, Maximilian Frederik Russe

Förderung

  • Berta-Ottenstein-Programme for Clinician Scientists, Faculty of Medicine
  • University of Freiburg

Zitieren

APA

Semper-Hogg, W., Rau, A., Fuessinger, M. A., Zimmermann, S., Bamberg, F., Metzger, M. C., Schmelzeisen, R., Rau, S., Reisert, M., & Russe, M. F. (2025). Deep learning-based segmentation of the mandibular canals in cone-beam CT reaches human-level performance. Dentomaxillofacial Radiology. https://doi.org/10.1093/dmfr/twae069

Vancouver

Semper-Hogg W, Rau A, Fuessinger MA, Zimmermann S, Bamberg F, Metzger MC, et al. Deep learning-based segmentation of the mandibular canals in cone-beam CT reaches human-level performance. Dentomaxillofacial Radiology. 2025. doi:10.1093/dmfr/twae069

BibTeX
@article{SemperHogg2025,
  author = {Semper-Hogg, Wiebke and Rau, Alexander and Fuessinger, Marc Anton and Zimmermann, Sabrina and Bamberg, Fabian and Metzger, Marc Christian and Schmelzeisen, Rainer and Rau, Stephan and Reisert, Marco and Russe, Maximilian Frederik},
  title = {Deep learning-based segmentation of the mandibular canals in cone-beam CT reaches human-level performance},
  journal = {Dentomaxillofacial Radiology},
  year = {2025},
  doi = {10.1093/dmfr/twae069}
}
Kostenlos starten →