EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

Can Deep Learning Methods Differentiate Temporomandibular Joint Disorders From Healthy Joints? A 3D Artificial Intelligence Algorithm Study Based on CBCT Images

İbrahim Şevki Bayrakdar iD, Sevda Kurt-Bayrakdar iD, Alican Kuran iD, Özer Çelik iD, Kaan Orhan iD, Rohan Jagtap

Journal of Oral Rehabilitation · 2026

DOI: 10.1111/joor.70247 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Oral Rehabilitation, 2026. Verfasst von İbrahim Şevki Bayrakdar und 5 weiteren Autor:innen. Beteiligte Einrichtung: Department of Oral and Maxillofacial Radiology, Faculty of Dentistry, Eskisehir Osmangazi University, Eskisehir, Turkey. Hauptthemen laut MEDLINE: Temporomandibular Joint Disorders, Deep Learning, Cone-Beam Computed Tomography, Imaging, Three-Dimensional. Die Arbeit nennt 32 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: Türkiye Bilimsel ve Teknolojik Araştırma Kurumu.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Given the high difficulty of interpreting complex bone‐related temporomandibular joint disorders (TMD) on CBCT scans, developing advanced artificial intelligence models has become essential for accurate diagnosis and detailed subcategorization.

Objectives

This study aimed to develop deep learning models capable of performing a wide range of tasks in the evaluation of the temporomandibular joint on CBCT scans.

Methods

This study followed a four‐stage design. First, an nnU‐Net v2–based model was developed for automatic segmentation of mandibular condyles on CBCT images. Second, a 3D CNN model was trained to classify healthy and pathological condyles. Third, a 3D CNN was used to differentiate five TMD types. In the fourth stage, for exploratory analysis purposes, an nnU‐Net v2–based model was implemented to grade the severity of erosion, osteophyte formation and sclerosis.

Results

The segmentation model achieved automatic segmentation of healthy and pathological condyles with 0.87 DSC and 0.77 IoU values. The classification model achieved an F1‐score of 0.65 in distinguishing healthy condyles from TMDs. For differentiating the five TMD types from healthy condyles, the F1‐scores were 0.76 for erosion, 0.88 for flattening, 0.86 for osteophyte formation, 0.88 for sclerosis and 0.85 for subchondral cysts. In grading analysis, the highest performance was observed for Grade 1 erosion, Grade 2 osteophyte formation and Grade 3 sclerosis.

Conclusion

Deep learning algorithms can achieve clinically relevant performance in the segmentation of the mandibular condyle and in differentiating TMDs from healthy condyles. However, further studies with larger and more balanced datasets are needed to improve the ability of these models to classify TMD grades accurately.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Temporomandibular Joint DisordersDeep LearningCone-Beam Computed TomographyImaging, Three-DimensionalTemporomandibular JointMandibular CondyleHumansAlgorithmsArtificial IntelligenceConvolutional Neural NetworksDiagnosis, Differential

Stichwörter der Autor:innen

artificial intelligencecone beam computed tomographydeep learningtemporomandibular disorderstemporomandibular joint

Autor:innen und Institutionen

  • Department of Oral and Maxillofacial Radiology, Faculty of Dentistry, Eskisehir Osmangazi University, Eskisehir, Turkey.
  • Division of Oral and Maxillofacial Radiology, Department of Care Planning and Restorative Sciences, University of Mississippi Medical Center School of Dentistry, Jackson, Mississippi, USA.
  • Department of Periodontology, Faculty of Dentistry, Eskisehir Osmangazi University, Eskisehir, Turkey.
  • Department of Oral and Maxillofacial Radiology, Faculty of Dentistry, Kocaeli University, Kocaeli, Turkey.
  • Artificial Intelligence Research and Application Center, Anadolu University, Eskisehir, Turkey.
  • Department of Oral and Maxillofacial Radiology, Faculty of Dentistry, Ankara University, Ankara, Turkey.

ORCID-Profile: İbrahim Şevki Bayrakdar, Sevda Kurt-Bayrakdar, Alican Kuran, Özer Çelik, Kaan Orhan

Förderung

  • Türkiye Bilimsel ve Teknolojik Araştırma Kurumu

Zitieren

APA

Bayrakdar, İ. Ş., Kurt-Bayrakdar, S., Kuran, A., Çelik, Ö., Orhan, K., & Jagtap, R. (2026). Can Deep Learning Methods Differentiate Temporomandibular Joint Disorders From Healthy Joints? A 3D Artificial Intelligence Algorithm Study Based on CBCT Images. Journal of Oral Rehabilitation. https://doi.org/10.1111/joor.70247

Vancouver

Bayrakdar İŞ, Kurt-Bayrakdar S, Kuran A, Çelik Ö, Orhan K, Jagtap R. Can Deep Learning Methods Differentiate Temporomandibular Joint Disorders From Healthy Joints? A 3D Artificial Intelligence Algorithm Study Based on CBCT Images. Journal of Oral Rehabilitation. 2026. doi:10.1111/joor.70247

BibTeX
@article{Bayrakdar2026,
  author = {Bayrakdar, İbrahim Şevki and Kurt-Bayrakdar, Sevda and Kuran, Alican and Çelik, Özer and Orhan, Kaan and Jagtap, Rohan},
  title = {Can Deep Learning Methods Differentiate Temporomandibular Joint Disorders From Healthy Joints? A 3D Artificial Intelligence Algorithm Study Based on CBCT Images},
  journal = {Journal of Oral Rehabilitation},
  year = {2026},
  doi = {10.1111/joor.70247}
}
Kostenlos starten →