Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2025 Jul 1;25(1):217.
doi: 10.1186/s12880-025-01797-9.

Automated 3D segmentation of the hyoid bone in CBCT using nnU-Net v2: a retrospective study on model performance and potential clinical utility

Ismail Gümüssoy  1 , Emre Haylaz  2 , Suayip Burak Duman  3 , Fahrettin Kalabalik  1 , Seyda Say  1 , Ozer Celik  4 , Ibrahim Sevki Bayrakdar  5
Affiliations

Automated 3D segmentation of the hyoid bone in CBCT using nnU-Net v2: a retrospective study on model performance and potential clinical utility

Ismail Gümüssoy et al. BMC Med Imaging. .

Abstract

Objective: This study aimed to identify the hyoid bone (HB) using the nnU-Net based artificial intelligence (AI) model in cone beam computed tomography (CBCT) images and assess the model's success in automatic segmentation.

Methods: CBCT images of 190 patients were randomly selected. The raw data was converted to DICOM format and transferred to the 3D Slicer Imaging Software (Version 4.10.2; MIT, Cambridge, MA, USA). HB was labeled manually using the 3D Slicer. The dataset was divided into training, validation, and test sets in a ratio of 8:1:1. The nnU-Net v2 architecture was utilized to process the training and test datasets, generating the algorithm weight factors. To assess the model's accuracy and performance, a confusion matrix was employed. F1-score, Dice coefficient (DC), 95% Hausdorff distance (95% HD), and Intersection over Union (IoU) metrics were calculated to evaluate the results.

Results: The model's performance metrics were as follows: DC = 0.9434, IoU = 0.8941, F1-score = 0.9446, and 95% HD = 1.9998. The receiver operating characteristic (ROC) curve was generated, yielding an AUC value of 0.98.

Conclusion: The results indicated that the nnU-Net v2 model achieved high precision and accuracy in HB segmentation on CBCT images. Automatic segmentation of HB can enhance clinicians' decision-making speed and accuracy in diagnosing and treating various clinical conditions.

Clinical trial number: Not applicable.

Keywords: Artificial intelligence; Cone-beam computed tomography; Convolutional neural network; Hyoid bone.

PubMed Disclaimer

Conflict of interest statement

Declarations. Ethics approval and consent to participate: This study was approved by the Non-Interventional Clinical Research Ethics Committee of Inönü University (approval no.2025/6807, approval date: 3 January 2025). We certify that the study was performed in accordance with the Helsinki Declaration of 1975 and later amendments. This study was approved by the Non-Interventional Clinical Research Ethics Committee of Inönü University, and informed consent was waived for all patients due to the retrospective nature of the study. Consent for publication: Not applicable. Competing interests: The authors declare no competing interests.

Figures

Fig. 1
Fig. 1
Computational pipeline of the nnU-Net v2 algortihm
Fig. 2
Fig. 2
Manual labeling of hyoid bone in axial (A, B, C), coronal (D, E, F), and sagittal (G, H, I) sections on CBCT images
Fig. 3
Fig. 3
The workflow of the methodology used to develop and train the deep learning model
Fig. 4
Fig. 4
Original and predictive analysis of the proposed nnU-Net v2 model
Fig. 5
Fig. 5
Quantitative evaluation of segmentation performance using standard metrics (Precision, Recall, Dice, IoU, F1-score, and 95% Hausdorff Distance)
Fig. 6
Fig. 6
Visualization of the model’s ROC curve and reflection of performance success through the AUC value
Fig. 7
Fig. 7
DC and loss function values at each number of cycles of the model
See this image and copyright information in PMC

Similar articles

See all similar articles

References

    1. Auvenshine RC, Pettit NJ. The hyoid bone: an overview. CRANIO®. 2020;38(1):6–14. - PubMed
    1. AlJulaih GH, Menezes RG. Anatomy, head and neck: hyoid bone. StatPearls [Internet]. edn.: StatPearls Publishing; 2023. - PubMed
    1. Fakhry N, Puymerail L, Michel J, Santini L, Lebreton-Chakour C, Robert D et al. Analysis of hyoid bone using 3D geometric morphometrics: an anatomical study and discussion of potential clinical implications. Dysphagia. 2013;28435–445. - PubMed
    1. Angoules A, Boutsikari E. Traumatic hyoid bone fractures: rare but potentially life threatening injuries. Emerg Med. 2013;3(1):e128.
    1. Wang X, Wang C, Zhang S, Wang W, Li X, Gao S, Li K, Chen J, Wang H, Chen L. Microstructure of the hyoid bone based on micro-computed tomography findings. Medicine. 2020;99(44):e22246. - PMC - PubMed

LinkOut - more resources