Deep ensemble learning-driven fully automated multi-structure segmentation for precision craniomaxillofacial surgery

ObjectivesAccurate segmentation of craniomaxillofacial (CMF) structures and individual teeth is essential for advancing computer-assisted CMF surgery. This study developed CMF-ELSeg, a novel fully automatic multi-structure segmentation model based on deep ensemble learning.MethodsA total of 143 CMF...

Full description

Saved in:
Bibliographic Details
Main Authors: Jiahao Bao, Zongcai Tan, Yifeng Sun, Xinyu Xu, Huazhen Liu, Weiyi Cui, Yang Yang, Mengjia Cheng, Yiming Wang, Congshuang Ku, Yuen Ka Ho, Jiayi Zhu, Linfeng Fan, Dahong Qian, Shunyao Shen, Yaofeng Wen, Hongbo Yu
Format: Article
Language:English
Published: Frontiers Media S.A. 2025-05-01
Series:Frontiers in Bioengineering and Biotechnology
Subjects:
Online Access:https://www.frontiersin.org/articles/10.3389/fbioe.2025.1580502/full
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850035627432607744
author Jiahao Bao
Zongcai Tan
Yifeng Sun
Xinyu Xu
Huazhen Liu
Weiyi Cui
Yang Yang
Mengjia Cheng
Yiming Wang
Congshuang Ku
Yuen Ka Ho
Jiayi Zhu
Linfeng Fan
Dahong Qian
Shunyao Shen
Yaofeng Wen
Hongbo Yu
author_facet Jiahao Bao
Zongcai Tan
Yifeng Sun
Xinyu Xu
Huazhen Liu
Weiyi Cui
Yang Yang
Mengjia Cheng
Yiming Wang
Congshuang Ku
Yuen Ka Ho
Jiayi Zhu
Linfeng Fan
Dahong Qian
Shunyao Shen
Yaofeng Wen
Hongbo Yu
author_sort Jiahao Bao
collection DOAJ
description ObjectivesAccurate segmentation of craniomaxillofacial (CMF) structures and individual teeth is essential for advancing computer-assisted CMF surgery. This study developed CMF-ELSeg, a novel fully automatic multi-structure segmentation model based on deep ensemble learning.MethodsA total of 143 CMF computed tomography (CT) scans were retrospectively collected and manually annotated by experts for model training and validation. Three 3D U-Net–based deep learning models (V-Net, nnU-Net, and 3D UX-Net) were benchmarked. CMF-ELSeg employed a coarse-to-fine cascaded architecture and an ensemble approach to integrate the strengths of these models. Segmentation performance was evaluated using Dice score and Intersection over Union (IoU) by comparing model predictions to ground truth annotations. Clinical feasibility was assessed through qualitative and quantitative analyses.ResultsIn coarse segmentation of the upper skull, mandible, cervical vertebra, and pharyngeal cavity, 3D UX-Net and nnU-Net achieved Dice scores above 0.96 and IoU above 0.93. For fine segmentation and classification of individual teeth, the cascaded 3D UX-Net performed best. CMF-ELSeg improved Dice scores by 3%–5% over individual models for facial soft tissue, upper skull, mandible, cervical vertebra, and pharyngeal cavity segmentation, and maintained high accuracy Dice > 0.94 for most teeth. Clinical evaluation confirmed that CMF-ELSeg performed reliably in patients with skeletal malocclusion, fractures, and fibrous dysplasia.ConclusionCMF-ELSeg provides high-precision segmentation of CMF structures and teeth by leveraging multiple models, serving as a practical tool for clinical applications and enhancing patient-specific treatment planning in CMF surgery.
format Article
id doaj-art-727d087834794b028c8309ca27b4272d
institution DOAJ
issn 2296-4185
language English
publishDate 2025-05-01
publisher Frontiers Media S.A.
record_format Article
series Frontiers in Bioengineering and Biotechnology
spelling doaj-art-727d087834794b028c8309ca27b4272d2025-08-20T02:57:25ZengFrontiers Media S.A.Frontiers in Bioengineering and Biotechnology2296-41852025-05-011310.3389/fbioe.2025.15805021580502Deep ensemble learning-driven fully automated multi-structure segmentation for precision craniomaxillofacial surgeryJiahao Bao0Zongcai Tan1Yifeng Sun2Xinyu Xu3Huazhen Liu4Weiyi Cui5Yang Yang6Mengjia Cheng7Yiming Wang8Congshuang Ku9Yuen Ka Ho10Jiayi Zhu11Linfeng Fan12Dahong Qian13Shunyao Shen14Yaofeng Wen15Hongbo Yu16Department of Oral and Craniomaxillofacial Surgery, Shanghai Ninth People’s Hospital, Shanghai Jiao Tong University School of Medicine, College of Stomatology, Shanghai Jiao Tong University, National Center for Stomatology, National Clinical Research Center for Oral Diseases, Shanghai Research Institute of Stomatology, Shanghai Key Laboratory of Stomatology, Shanghai, ChinaHamlyn Centre for Robotic Surgery, Institute of Global Health Innovation, Imperial College London, London, United KingdomSchool of Mechanical Engineering, Shanghai Dianji University, Shanghai, ChinaSchool of Electronic Information and Electrical Engineering, Shanghai Jiao Tong University, Shanghai, ChinaSchool of Electronic Information and Electrical Engineering, Shanghai Jiao Tong University, Shanghai, ChinaDepartment of Oral and Craniomaxillofacial Surgery, Shanghai Ninth People’s Hospital, Shanghai Jiao Tong University School of Medicine, College of Stomatology, Shanghai Jiao Tong University, National Center for Stomatology, National Clinical Research Center for Oral Diseases, Shanghai Research Institute of Stomatology, Shanghai Key Laboratory of Stomatology, Shanghai, ChinaShanghai Lanhui Medical Technology Co., Ltd., Shanghai, ChinaFaculty of Dentistry, The University of Hong Kong, Hong Kong, Hong Kong SAR, ChinaDepartment of Oral and Craniomaxillofacial Surgery, Shanghai Ninth People’s Hospital, Shanghai Jiao Tong University School of Medicine, College of Stomatology, Shanghai Jiao Tong University, National Center for Stomatology, National Clinical Research Center for Oral Diseases, Shanghai Research Institute of Stomatology, Shanghai Key Laboratory of Stomatology, Shanghai, ChinaDepartment of Oral and Craniomaxillofacial Surgery, Shanghai Ninth People’s Hospital, Shanghai Jiao Tong University School of Medicine, College of Stomatology, Shanghai Jiao Tong University, National Center for Stomatology, National Clinical Research Center for Oral Diseases, Shanghai Research Institute of Stomatology, Shanghai Key Laboratory of Stomatology, Shanghai, ChinaDepartment of Oral and Craniomaxillofacial Surgery, Shanghai Ninth People’s Hospital, Shanghai Jiao Tong University School of Medicine, College of Stomatology, Shanghai Jiao Tong University, National Center for Stomatology, National Clinical Research Center for Oral Diseases, Shanghai Research Institute of Stomatology, Shanghai Key Laboratory of Stomatology, Shanghai, ChinaDepartment of Oral and Craniomaxillofacial Surgery, Shanghai Ninth People’s Hospital, Shanghai Jiao Tong University School of Medicine, College of Stomatology, Shanghai Jiao Tong University, National Center for Stomatology, National Clinical Research Center for Oral Diseases, Shanghai Research Institute of Stomatology, Shanghai Key Laboratory of Stomatology, Shanghai, ChinaDepartment of Radiology, Shanghai Ninth People’s Hospital, College of Stomatology, Shanghai Jiao Tong University School of Medicine, Shanghai, ChinaSchool of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai, ChinaDepartment of Oral and Craniomaxillofacial Surgery, Shanghai Ninth People’s Hospital, Shanghai Jiao Tong University School of Medicine, College of Stomatology, Shanghai Jiao Tong University, National Center for Stomatology, National Clinical Research Center for Oral Diseases, Shanghai Research Institute of Stomatology, Shanghai Key Laboratory of Stomatology, Shanghai, ChinaSchool of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai, ChinaDepartment of Oral and Craniomaxillofacial Surgery, Shanghai Ninth People’s Hospital, Shanghai Jiao Tong University School of Medicine, College of Stomatology, Shanghai Jiao Tong University, National Center for Stomatology, National Clinical Research Center for Oral Diseases, Shanghai Research Institute of Stomatology, Shanghai Key Laboratory of Stomatology, Shanghai, ChinaObjectivesAccurate segmentation of craniomaxillofacial (CMF) structures and individual teeth is essential for advancing computer-assisted CMF surgery. This study developed CMF-ELSeg, a novel fully automatic multi-structure segmentation model based on deep ensemble learning.MethodsA total of 143 CMF computed tomography (CT) scans were retrospectively collected and manually annotated by experts for model training and validation. Three 3D U-Net–based deep learning models (V-Net, nnU-Net, and 3D UX-Net) were benchmarked. CMF-ELSeg employed a coarse-to-fine cascaded architecture and an ensemble approach to integrate the strengths of these models. Segmentation performance was evaluated using Dice score and Intersection over Union (IoU) by comparing model predictions to ground truth annotations. Clinical feasibility was assessed through qualitative and quantitative analyses.ResultsIn coarse segmentation of the upper skull, mandible, cervical vertebra, and pharyngeal cavity, 3D UX-Net and nnU-Net achieved Dice scores above 0.96 and IoU above 0.93. For fine segmentation and classification of individual teeth, the cascaded 3D UX-Net performed best. CMF-ELSeg improved Dice scores by 3%–5% over individual models for facial soft tissue, upper skull, mandible, cervical vertebra, and pharyngeal cavity segmentation, and maintained high accuracy Dice > 0.94 for most teeth. Clinical evaluation confirmed that CMF-ELSeg performed reliably in patients with skeletal malocclusion, fractures, and fibrous dysplasia.ConclusionCMF-ELSeg provides high-precision segmentation of CMF structures and teeth by leveraging multiple models, serving as a practical tool for clinical applications and enhancing patient-specific treatment planning in CMF surgery.https://www.frontiersin.org/articles/10.3389/fbioe.2025.1580502/fulldeep learningcraniomaxillofacial surgeryvirtual surgical planningcomputed tomographysegmentation
spellingShingle Jiahao Bao
Zongcai Tan
Yifeng Sun
Xinyu Xu
Huazhen Liu
Weiyi Cui
Yang Yang
Mengjia Cheng
Yiming Wang
Congshuang Ku
Yuen Ka Ho
Jiayi Zhu
Linfeng Fan
Dahong Qian
Shunyao Shen
Yaofeng Wen
Hongbo Yu
Deep ensemble learning-driven fully automated multi-structure segmentation for precision craniomaxillofacial surgery
Frontiers in Bioengineering and Biotechnology
deep learning
craniomaxillofacial surgery
virtual surgical planning
computed tomography
segmentation
title Deep ensemble learning-driven fully automated multi-structure segmentation for precision craniomaxillofacial surgery
title_full Deep ensemble learning-driven fully automated multi-structure segmentation for precision craniomaxillofacial surgery
title_fullStr Deep ensemble learning-driven fully automated multi-structure segmentation for precision craniomaxillofacial surgery
title_full_unstemmed Deep ensemble learning-driven fully automated multi-structure segmentation for precision craniomaxillofacial surgery
title_short Deep ensemble learning-driven fully automated multi-structure segmentation for precision craniomaxillofacial surgery
title_sort deep ensemble learning driven fully automated multi structure segmentation for precision craniomaxillofacial surgery
topic deep learning
craniomaxillofacial surgery
virtual surgical planning
computed tomography
segmentation
url https://www.frontiersin.org/articles/10.3389/fbioe.2025.1580502/full
work_keys_str_mv AT jiahaobao deepensemblelearningdrivenfullyautomatedmultistructuresegmentationforprecisioncraniomaxillofacialsurgery
AT zongcaitan deepensemblelearningdrivenfullyautomatedmultistructuresegmentationforprecisioncraniomaxillofacialsurgery
AT yifengsun deepensemblelearningdrivenfullyautomatedmultistructuresegmentationforprecisioncraniomaxillofacialsurgery
AT xinyuxu deepensemblelearningdrivenfullyautomatedmultistructuresegmentationforprecisioncraniomaxillofacialsurgery
AT huazhenliu deepensemblelearningdrivenfullyautomatedmultistructuresegmentationforprecisioncraniomaxillofacialsurgery
AT weiyicui deepensemblelearningdrivenfullyautomatedmultistructuresegmentationforprecisioncraniomaxillofacialsurgery
AT yangyang deepensemblelearningdrivenfullyautomatedmultistructuresegmentationforprecisioncraniomaxillofacialsurgery
AT mengjiacheng deepensemblelearningdrivenfullyautomatedmultistructuresegmentationforprecisioncraniomaxillofacialsurgery
AT yimingwang deepensemblelearningdrivenfullyautomatedmultistructuresegmentationforprecisioncraniomaxillofacialsurgery
AT congshuangku deepensemblelearningdrivenfullyautomatedmultistructuresegmentationforprecisioncraniomaxillofacialsurgery
AT yuenkaho deepensemblelearningdrivenfullyautomatedmultistructuresegmentationforprecisioncraniomaxillofacialsurgery
AT jiayizhu deepensemblelearningdrivenfullyautomatedmultistructuresegmentationforprecisioncraniomaxillofacialsurgery
AT linfengfan deepensemblelearningdrivenfullyautomatedmultistructuresegmentationforprecisioncraniomaxillofacialsurgery
AT dahongqian deepensemblelearningdrivenfullyautomatedmultistructuresegmentationforprecisioncraniomaxillofacialsurgery
AT shunyaoshen deepensemblelearningdrivenfullyautomatedmultistructuresegmentationforprecisioncraniomaxillofacialsurgery
AT yaofengwen deepensemblelearningdrivenfullyautomatedmultistructuresegmentationforprecisioncraniomaxillofacialsurgery
AT hongboyu deepensemblelearningdrivenfullyautomatedmultistructuresegmentationforprecisioncraniomaxillofacialsurgery