Automatic identification of hard and soft tissue landmarks in cone-beam computed tomography via deep learning with diversity datasets: a methodological study

Abstract Background Manual landmark detection in cone beam computed tomography (CBCT) for evaluating craniofacial structures relies on medical expertise and is time-consuming. This study aimed to apply a new deep learning method to predict and locate soft and hard tissue craniofacial landmarks on CB...

Full description

Saved in:
Bibliographic Details
Main Authors: Yan Jiang, Canyang Jiang, Bin Shi, You Wu, Shuli Xing, Hao Liang, Jianping Huang, Xiaohong Huang, Li Huang, Lisong Lin
Format: Article
Language:English
Published: BMC 2025-04-01
Series:BMC Oral Health
Subjects:
Online Access:https://doi.org/10.1186/s12903-025-05831-8
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850201069591724032
author Yan Jiang
Canyang Jiang
Bin Shi
You Wu
Shuli Xing
Hao Liang
Jianping Huang
Xiaohong Huang
Li Huang
Lisong Lin
author_facet Yan Jiang
Canyang Jiang
Bin Shi
You Wu
Shuli Xing
Hao Liang
Jianping Huang
Xiaohong Huang
Li Huang
Lisong Lin
author_sort Yan Jiang
collection DOAJ
description Abstract Background Manual landmark detection in cone beam computed tomography (CBCT) for evaluating craniofacial structures relies on medical expertise and is time-consuming. This study aimed to apply a new deep learning method to predict and locate soft and hard tissue craniofacial landmarks on CBCT in patients with various types of malocclusion. Methods A total of 498 CBCT images were collected. Following the calibration procedure, two experienced clinicians identified 43 landmarks in the x-, y-, and z-coordinate planes on the CBCT images using Checkpoint Software, creating the ground truth by averaging the landmark coordinates. To evaluate the accuracy of our algorithm, we determined the mean absolute error along the x-, y-, and z-axes and calculated the mean radial error (MRE) between the reference landmark and predicted landmark, as well as the successful detection rate (SDR). Results Each landmark prediction took approximately 4.2 s on a conventional graphics processing unit. The mean absolute error across all coordinates was 0.74 mm. The overall MRE for the 43 landmarks was 1.76 ± 1.13 mm, and the SDR was 60.16%, 91.05%, and 97.58% within 2-, 3-, and 4-mm error ranges of manual marking, respectively. The average MRE of the hard tissue landmarks (32/43) was 1.73 mm, while that for soft tissue landmarks (11/43) was 1.84 mm. Conclusions Our proposed algorithm demonstrates a clinically acceptable level of accuracy and robustness for automatic detection of CBCT soft- and hard-tissue landmarks across all types of malformations. The potential for artificial intelligence to assist in identifying three dimensional-CT landmarks in routine clinical practice and analysing large datasets for future research is promising.
format Article
id doaj-art-e728dd891ad04e64871e18e9868ec0ac
institution OA Journals
issn 1472-6831
language English
publishDate 2025-04-01
publisher BMC
record_format Article
series BMC Oral Health
spelling doaj-art-e728dd891ad04e64871e18e9868ec0ac2025-08-20T02:12:07ZengBMCBMC Oral Health1472-68312025-04-0125111610.1186/s12903-025-05831-8Automatic identification of hard and soft tissue landmarks in cone-beam computed tomography via deep learning with diversity datasets: a methodological studyYan Jiang0Canyang Jiang1Bin Shi2You Wu3Shuli Xing4Hao Liang5Jianping Huang6Xiaohong Huang7Li Huang8Lisong Lin9Department of Stomatology, The First Affiliated Hospital of Fujian Medical University, Tai-Jiang DistrictDepartment of Stomatology, National Regional Medical Center, Binhai Campus of the First Affiliated Hospital, Fujian Medical UniversityDepartment of Stomatology, The First Affiliated Hospital of Fujian Medical University, Tai-Jiang DistrictSchool of Stomatology, Fujian Medical UniversityCollege of Computer Science and Mathematics, Fujian University of TechnologyCollege of Computer Science and Mathematics, Fujian University of TechnologyDepartment of Stomatology, National Regional Medical Center, Binhai Campus of the First Affiliated Hospital, Fujian Medical UniversityDepartment of Stomatology, The First Affiliated Hospital of Fujian Medical University, Tai-Jiang DistrictDepartment of Stomatology, The First Affiliated Hospital of Fujian Medical University, Tai-Jiang DistrictDepartment of Stomatology, The First Affiliated Hospital of Fujian Medical University, Tai-Jiang DistrictAbstract Background Manual landmark detection in cone beam computed tomography (CBCT) for evaluating craniofacial structures relies on medical expertise and is time-consuming. This study aimed to apply a new deep learning method to predict and locate soft and hard tissue craniofacial landmarks on CBCT in patients with various types of malocclusion. Methods A total of 498 CBCT images were collected. Following the calibration procedure, two experienced clinicians identified 43 landmarks in the x-, y-, and z-coordinate planes on the CBCT images using Checkpoint Software, creating the ground truth by averaging the landmark coordinates. To evaluate the accuracy of our algorithm, we determined the mean absolute error along the x-, y-, and z-axes and calculated the mean radial error (MRE) between the reference landmark and predicted landmark, as well as the successful detection rate (SDR). Results Each landmark prediction took approximately 4.2 s on a conventional graphics processing unit. The mean absolute error across all coordinates was 0.74 mm. The overall MRE for the 43 landmarks was 1.76 ± 1.13 mm, and the SDR was 60.16%, 91.05%, and 97.58% within 2-, 3-, and 4-mm error ranges of manual marking, respectively. The average MRE of the hard tissue landmarks (32/43) was 1.73 mm, while that for soft tissue landmarks (11/43) was 1.84 mm. Conclusions Our proposed algorithm demonstrates a clinically acceptable level of accuracy and robustness for automatic detection of CBCT soft- and hard-tissue landmarks across all types of malformations. The potential for artificial intelligence to assist in identifying three dimensional-CT landmarks in routine clinical practice and analysing large datasets for future research is promising.https://doi.org/10.1186/s12903-025-05831-83D landmark identificationAccuracyCBCTDeep learning
spellingShingle Yan Jiang
Canyang Jiang
Bin Shi
You Wu
Shuli Xing
Hao Liang
Jianping Huang
Xiaohong Huang
Li Huang
Lisong Lin
Automatic identification of hard and soft tissue landmarks in cone-beam computed tomography via deep learning with diversity datasets: a methodological study
BMC Oral Health
3D landmark identification
Accuracy
CBCT
Deep learning
title Automatic identification of hard and soft tissue landmarks in cone-beam computed tomography via deep learning with diversity datasets: a methodological study
title_full Automatic identification of hard and soft tissue landmarks in cone-beam computed tomography via deep learning with diversity datasets: a methodological study
title_fullStr Automatic identification of hard and soft tissue landmarks in cone-beam computed tomography via deep learning with diversity datasets: a methodological study
title_full_unstemmed Automatic identification of hard and soft tissue landmarks in cone-beam computed tomography via deep learning with diversity datasets: a methodological study
title_short Automatic identification of hard and soft tissue landmarks in cone-beam computed tomography via deep learning with diversity datasets: a methodological study
title_sort automatic identification of hard and soft tissue landmarks in cone beam computed tomography via deep learning with diversity datasets a methodological study
topic 3D landmark identification
Accuracy
CBCT
Deep learning
url https://doi.org/10.1186/s12903-025-05831-8
work_keys_str_mv AT yanjiang automaticidentificationofhardandsofttissuelandmarksinconebeamcomputedtomographyviadeeplearningwithdiversitydatasetsamethodologicalstudy
AT canyangjiang automaticidentificationofhardandsofttissuelandmarksinconebeamcomputedtomographyviadeeplearningwithdiversitydatasetsamethodologicalstudy
AT binshi automaticidentificationofhardandsofttissuelandmarksinconebeamcomputedtomographyviadeeplearningwithdiversitydatasetsamethodologicalstudy
AT youwu automaticidentificationofhardandsofttissuelandmarksinconebeamcomputedtomographyviadeeplearningwithdiversitydatasetsamethodologicalstudy
AT shulixing automaticidentificationofhardandsofttissuelandmarksinconebeamcomputedtomographyviadeeplearningwithdiversitydatasetsamethodologicalstudy
AT haoliang automaticidentificationofhardandsofttissuelandmarksinconebeamcomputedtomographyviadeeplearningwithdiversitydatasetsamethodologicalstudy
AT jianpinghuang automaticidentificationofhardandsofttissuelandmarksinconebeamcomputedtomographyviadeeplearningwithdiversitydatasetsamethodologicalstudy
AT xiaohonghuang automaticidentificationofhardandsofttissuelandmarksinconebeamcomputedtomographyviadeeplearningwithdiversitydatasetsamethodologicalstudy
AT lihuang automaticidentificationofhardandsofttissuelandmarksinconebeamcomputedtomographyviadeeplearningwithdiversitydatasetsamethodologicalstudy
AT lisonglin automaticidentificationofhardandsofttissuelandmarksinconebeamcomputedtomographyviadeeplearningwithdiversitydatasetsamethodologicalstudy