Deep learning for automated segmentation of the temporomandibular joint

Shankeeth Vinayahalingam; Bo Berends; Frank Baan; David Anssari Moin; Rik van Luijn; Stefaan Bergé; Tong Xi

doi:10.1016/j.jdent.2023.104475

Deep learning for automated segmentation of the temporomandibular joint

J Dent. 2023 May:132:104475. doi: 10.1016/j.jdent.2023.104475. Epub 2023 Mar 2.

Authors

Shankeeth Vinayahalingam¹, Bo Berends², Frank Baan², David Anssari Moin³, Rik van Luijn⁴, Stefaan Bergé⁴, Tong Xi⁴

Affiliations

¹ Department of Oral and Maxillofacial Surgery, Radboud University Nijmegen Medical Centre, P.O. Box 9101, Postal number 590, Nijmegen, HB 6500, The Netherlands. Electronic address: Shankeeth.Vinayahalingam@radboudumc.nl.
² Department of Oral and Maxillofacial Surgery, Radboud University Nijmegen Medical Centre, P.O. Box 9101, Postal number 590, Nijmegen, HB 6500, The Netherlands; Radboudumc 3DLab, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands.
³ Promaton Co. Ltd., Amsterdam, GR 1076, The Netherlands.
⁴ Department of Oral and Maxillofacial Surgery, Radboud University Nijmegen Medical Centre, P.O. Box 9101, Postal number 590, Nijmegen, HB 6500, The Netherlands.

PMID: 36870441
DOI: 10.1016/j.jdent.2023.104475

Abstract

Objective: Quantitative analysis of the volume and shape of the temporomandibular joint (TMJ) using cone-beam computed tomography (CBCT) requires accurate segmentation of the mandibular condyles and the glenoid fossae. This study aimed to develop and validate an automated segmentation tool based on a deep learning algorithm for accurate 3D reconstruction of the TMJ.

Materials and methods: A three-step deep-learning approach based on a 3D U-net was developed to segment the condyles and glenoid fossae on CBCT datasets. Three 3D U-Nets were utilized for region of interest (ROI) determination, bone segmentation, and TMJ classification. The AI-based algorithm was trained and validated on 154 manually segmented CBCT images. Two independent observers and the AI algorithm segmented the TMJs of a test set of 8 CBCTs. The time required for the segmentation and accuracy metrics (intersection of union, DICE, etc.) was calculated to quantify the degree of similarity between the manual segmentations (ground truth) and the performances of the AI models.

Results: The AI segmentation achieved an intersection over union (IoU) of 0.955 and 0.935 for the condyles and glenoid fossa, respectively. The IoU of the two independent observers for manual condyle segmentation were 0.895 and 0.928, respectively (p<0.05). The mean time required for the AI segmentation was 3.6 s (SD 0.9), whereas the two observers needed 378.9 s (SD 204.9) and 571.6 s (SD 257.4), respectively (p<0.001).

Conclusion: The AI-based automated segmentation tool segmented the mandibular condyles and glenoid fossae with high accuracy, speed, and consistency. Potential limited robustness and generalizability are risks that cannot be ruled out, as the algorithms were trained on scans from orthognathic surgery patients derived from just one type of CBCT scanner.

Clinical significance: The incorporation of the AI-based segmentation tool into diagnostic software could facilitate 3D qualitative and quantitative analysis of TMJs in a clinical setting, particularly for the diagnosis of TMJ disorders and longitudinal follow-up.

Keywords: Artificial intelligence; Computer-assisted planning; Cone-beam computed tomography; Deep learning; Digital imaging.

Publication types

Research Support, Non-U.S. Gov't

MeSH terms

Cone-Beam Computed Tomography / methods
Deep Learning*
Humans
Image Processing, Computer-Assisted / methods
Mandibular Condyle / diagnostic imaging
Mandibular Condyle / surgery
Temporomandibular Joint / diagnostic imaging
Temporomandibular Joint Disorders* / diagnostic imaging