Comparison Of Accuracy Of Mandibular Canal Segmentation Tool Using Manual Method And Artificial Intelligence In Cone Beam Computed Tomography
DOI:
https://doi.org/10.52783/jns.v14.3252Keywords:
Mandibular canal, artificial intelligence, CBCT, segmentation, manual tracing, nerve canal assessmentAbstract
Background: Cone Beam Computed Tomography (CBCT) has become an essential imaging modality in the visualization of the maxillofacial region, gradually replacing traditional CT due to its superior imaging capabilities and reduced radiation exposure. Accurate assessment of the mandibular canal (MC) is vital for preoperative planning to avoid complications during surgical procedures. Artificial Intelligence (AI)-driven tools, such as nerve canal segmentation software, offer rapid and precise tracing of the Inferior Alveolar Nerve Canal (IANC), extending from the mandibular foramen to the mental foramen. This study aims to evaluate and compare the accuracy of AI-driven segmentation with manual tracing methods to assess their reliability and potential for clinical applications.
Materials And Methods: This study included 100 large field-of-view (16 × 17 cm) CBCT scans taken for routine screening, acquired using the CS 9600 machine with exposure parameters of 120 KVp, 5 mA, and 24 s. The CBCT scans were analysed using CS imaging software. The inclusion criteria focused on scans from individuals over 18 years of age without any pathologies, fractures, or prior surgeries affecting the mandible. Exclusion criteria included individuals under 18 years of age, patients with syndromes affecting mandibular growth, and post – mandibular surgeries.
The mandibular canal was traced using an AI-driven nerve canal segmentation tool and the results were compared with manual tracings performed by two observers with varying levels of experience (2 and 6 years). The overlap of the traced canals was categorized as complete, partial, or no overlap. Additionally, the diameter of the traced canal was assessed, with AI-driven tools defaulting to 2.5 mm. Screenshots of the traced canals in reconstructed panoramic views were used for comparison, and transparency adjustments were applied to evaluate overlaps.
Results: The AI-driven segmentation demonstrated high accuracy, with 94% of cases showing complete overlap between AI and manual tracings, 5% showing partial overlap, and only 1% displaying no overlap. Morphological assessment revealed consistency in tracing quality across most cases. The AI tool maintained the default canal diameter of 2.5 mm with no significant deviations in 98% of cases. The remaining 2% exhibited minor variations, highlighting potential limitations in specific cases with unique anatomical variations. These findings underline the reliability of AI tools in accurately identifying and tracing the mandibular canal, with results comparable to manual methods.
Conclusion: The study demonstrates that AI-driven nerve canal segmentation tools offer a highly accurate and time-efficient approach for assessing the mandibular canal, with results closely aligning with manual tracings. While AI tools showed minimal discrepancies in a small percentage of cases, they hold significant promise as a complementary tool in preoperative planning and diagnostic workflows. However, further research incorporating larger datasets and varied imaging parameters is essential to optimize the use of AI tools in clinical practice.
Downloads
Metrics
References
Shintaku WH, Venturin JS et al. Applications of cone-beam computed tomography in fractures of the maxillofacial complex. Dent Traumatol. 2009 Aug;25(4):358–366. doi: 10.1111/j.1600-9657.2009.00795.x.
Scarfe WC, Molteni R, Mozzo P. Image Processing and Visualization Techniques. In: Scarfe WC, Angelopoulos C, editors. Maxillofacial Cone Beam Computed Tomography: Principles, Techniques and Clinical Applications. Cham: Springer; 2018. p. 43–93. doi: 10.1007/978-3-319-62061-9_3
Lahoud, P., EzEldeen, M., Beznik, T., Willems, H., Leite, A., Van Gerven, A., & Jacobs, R. (2021). Artificial intelligence for fast and accurate 3-dimensional tooth segmentation on cone-beam computed tomography. Journal of Endodontics, 47(5), 827–835. https://doi.org/10.1016/j.joen.2020.12.020.
La Rosa S, Quinzi V, Palazzo G, et al. The implications of artificial intelligence in pedodontics: A scoping review of evidence-based literature. Healthcare (Basel). 2024;12(13):1311. doi: 10.3390/healthcare12131311.
Tsolakis IA, Rontogianni A, Tsolakis AI, Papadopoulos MA. Comparing CBCT to model scanner for dental model scanning: An in vitro imaging accuracy study. Int Orthod. 2024;22(1):100840. doi: 10.1016/j.ortho.2023.100840.
Issa, J., Olszewski, R., & Dyszkiewicz-Konwińska, M. (2022). The effectiveness of semi-automated and fully automatic segmentation for inferior alveolar canal localization on CBCT scans: A systematic review. International Journal of Environmental Research and Public Health, 19(1), 560. https://doi.org/10.3390/ijerph19010560.
Agbaje, J. O., de Casteele, E. V., Salem, A. S., Anumendem, D., Lambrichts, I., & Politis, C. (2017). Tracking of the inferior alveolar nerve: Its implication in surgical planning. Clinical Oral Investigations, 21(7), 2213–2220. https://doi.org/10.1007/s00784-016-2014-x.
Lahoud, P., Diels, S., Niclaes, L., Van Aelst, S., Willems, H., Van Gerven, A., Quirynen, M., & Jacobs, R. (2022). Development and validation of a novel artificial intelligence driven tool for accurate mandibular canal segmentation on CBCT. Journal of Dentistry, 116, 103891. https://doi.org/10.1016/j.jdent.2021.103891.
Jaskari, J., Sahlsten, J., Järnstedt, J., Mehtonen, H., Karhu, K., Sundqvist, O., Hietanen, A., Varjonen, V., Mattila, V., & Kaski, K. (2020). Deep learning method for mandibular canal segmentation in dental cone beam computed tomography volumes. Scientific Reports, 10(1), 5842. https://doi.org/10.1038/s41598-020-62321-3.
Jeoun, B. S., Yang, S., Lee, S. J., Kim, T. I., Kim, J. M., Kim, J. E., Huh, K. H., Lee, S. S., Heo, M. S., & Yi, W. J. (2022). Canal-net for automatic and robust 3D segmentation of mandibular canals in CBCT images using a continuity-aware contextual network. Scientific Reports, 12(1), 13460. https://doi.org/10.1038/s41598-022-17341-6.
Kwak, G. H., Kwak, E. J., Song, J. M., Park, H. R., Jung, Y. H., Cho, B. H., Hui, P., & Hwang, J. J. (2020). Automatic mandibular canal detection using a deep convolutional neural network. Scientific Reports, 10(1), 5711. https://doi.org/10.1038/s41598-020-62586-8.
Kainmueller, D., Lamecker, H., Seim, H., Zinser, M., & Zachow, S. (2009). Automatic extraction of mandibular nerve and bone from cone-beam CT data. Medical Image Computing and Computer-Assisted Intervention: MICCAI—International Conference on Medical Image Computing and Computer-Assisted Intervention, 12(Pt 2), 76–83. https://doi.org/10.1007/978-3-642-04271-3_10
Khanagar, S. B., Al-Ehaideb, A., Maganur, P. C., Vishwanathaiah, S., Patil, S., Baeshen, H. A., Sarode, S. C., & Bhandi, S. (2021). Developments, application, and performance of artificial intelligence in dentistry – A systematic review. Journal of Dental Sciences, 16(1), 508–522. https://doi.org/10.1016/j.jds.2020.06.019.
Ahmed Z, Mohamed K, Zeeshan S, Dong X. Artificial intelligence with multi-functional machine learning platform development for better healthcare and precision medicine. Database (Oxford). 2020
Downloads
Published
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution 4.0 International License.
You are free to:
- Share — copy and redistribute the material in any medium or format
- Adapt — remix, transform, and build upon the material for any purpose, even commercially.
Terms:
- Attribution — You must give appropriate credit, provide a link to the license, and indicate if changes were made. You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use.
- No additional restrictions — You may not apply legal terms or technological measures that legally restrict others from doing anything the license permits.
