Deploying a Novel Deep Learning Framework for Segmentation of Specific Anatomical Structures on Cone-Beam CT

dc.authorid Bayrakdar, Ibrahim Sevki/0000-0001-5036-9867
dc.authorscopusid 57867839600
dc.authorscopusid 56593197800
dc.authorscopusid 57194858125
dc.authorscopusid 57215643433
dc.authorscopusid 55751747900
dc.authorwosid Bayrakdar, Ibrahim Sevki/B-2411-2015
dc.contributor.author Yuce, Fatma
dc.contributor.author Buyuk, Cansu
dc.contributor.author Bilgir, Elif
dc.contributor.author Celik, Ozer
dc.contributor.author Bayrakdar, Ibrahim Sevki
dc.date.accessioned 2025-06-15T22:09:06Z
dc.date.available 2025-06-15T22:09:06Z
dc.date.issued 2025
dc.department Okan University en_US
dc.department-temp [Yuce, Fatma] Istanbul Kent Univ, Dent Fac, Dentomaxillofacial Radiol, Siraselviler St 71, Beyoglu 34433, Istanbul, Turkiye; [Buyuk, Cansu] Istanbul Okan Univ, Dent Fac, Dentomaxillofacial Radiol, Istanbul, Turkiye; [Bilgir, Elif; Bayrakdar, Ibrahim Sevki] Eskisehir Osmangazi Univ, Dent Fac, Dentomaxillofacial Radiol, Eskisehir, Turkiye; [Celik, Ozer] Eskisehir Osmangazi Univ, Dept Math & Comp, Dentomaxillofacial Radiol, Eskisehir, Turkiye en_US
dc.description Bayrakdar, Ibrahim Sevki/0000-0001-5036-9867 en_US
dc.description.abstract AimCone-beam computed tomography (CBCT) imaging plays a crucial role in dentistry, with automatic prediction of anatomical structures on CBCT images potentially enhancing diagnostic and planning procedures. This study aims to predict anatomical structures automatically on CBCT images using a deep learning algorithm.Materials and methodsCBCT images from 70 patients were analyzed. Anatomical structures were annotated using a regional segmentation tool within an annotation software by two dentomaxillofacial radiologists. Each volumetric dataset comprised 405 slices, with relevant anatomical structures marked in each slice. Seventy DICOM images were converted to Nifti format, with seven reserved for testing and the remaining sixty-three used for training. The training utilized nnUNetv2 with an initial learning rate of 0.01, decreasing by 0.00001 at each epoch, and was conducted for 1000 epochs. Statistical analysis included accuracy, Dice score, precision, and recall results.ResultsThe segmentation model achieved an accuracy of 0.99 for nasal fossa, maxillary sinus, nasopalatine canal, mandibular canal, foramen mentale, and foramen mandible, with corresponding Dice scores of 0.85, 0.98, 0.79, 0.73, 0.78, and 0.74, respectively. Precision values ranged from 0.73 to 0.98. Maxillary sinus segmentation exhibited the highest performance, while mandibular canal segmentation showed the lowest performance.ConclusionThe results demonstrate high accuracy and precision across most structures, with varying Dice scores indicating the consistency of segmentation. Overall, our segmentation model exhibits robust performance in delineating anatomical features in CBCT images, promising potential applications in dental diagnostics and treatment planning. en_US
dc.description.woscitationindex Science Citation Index Expanded
dc.identifier.doi 10.1007/s11282-025-00831-4
dc.identifier.issn 0911-6028
dc.identifier.issn 1613-9674
dc.identifier.pmid 40445488
dc.identifier.scopus 2-s2.0-105006925924
dc.identifier.scopusquality Q2
dc.identifier.uri https://doi.org/10.1007/s11282-025-00831-4
dc.identifier.uri https://hdl.handle.net/20.500.14517/8008
dc.identifier.wos WOS:001499179400001
dc.identifier.wosquality Q3
dc.language.iso en en_US
dc.publisher Springer en_US
dc.relation.publicationcategory Makale - Uluslararası Hakemli Dergi - Kurum Öğretim Elemanı en_US
dc.rights info:eu-repo/semantics/closedAccess en_US
dc.subject Tomographic Anatomy en_US
dc.subject Cbct en_US
dc.subject Deep Learning en_US
dc.subject Segmentation en_US
dc.subject Head And Neck Anatomy en_US
dc.title Deploying a Novel Deep Learning Framework for Segmentation of Specific Anatomical Structures on Cone-Beam CT en_US
dc.type Article en_US

Files