Improving accuracy for inferior alveolar nerve segmentation with multi-label of anatomical adjacent structures using active learning in cone-beam computed tomography

Abstract Recent advancements in deep learning have revolutionized digital dentistry, highlighting the importance of precise dental segmentation. This study leverages active learning with the three-dimensional (3D) nnU-net and multi-labels to improve segmentation accuracy of dental anatomies, includi...

Full description

Saved in:
Bibliographic Details
Main Authors: Sungchul On, Junhyeok Ock, Myungsoo Bae, Jae-Woo Park, Seung-Hak Baek, Sungwon Ham, Namkug Kim
Format: Article
Language:English
Published: Nature Portfolio 2025-03-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-91725-2
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850029989784715264
author Sungchul On
Junhyeok Ock
Myungsoo Bae
Jae-Woo Park
Seung-Hak Baek
Sungwon Ham
Namkug Kim
author_facet Sungchul On
Junhyeok Ock
Myungsoo Bae
Jae-Woo Park
Seung-Hak Baek
Sungwon Ham
Namkug Kim
author_sort Sungchul On
collection DOAJ
description Abstract Recent advancements in deep learning have revolutionized digital dentistry, highlighting the importance of precise dental segmentation. This study leverages active learning with the three-dimensional (3D) nnU-net and multi-labels to improve segmentation accuracy of dental anatomies, including the maxillary sinuses, maxilla, mandible, and inferior alveolar nerves (IAN), which are important for implant planning, in 3D cone-beam computed tomography (CBCT) scans. Segmentation accuracy was compared using single-label, adjacent pair-label, and multi-label relevant anatomic structures with 60 CBCT scans from Kooalldam Dental Hospital and externally validated using data from Seoul National University Dental Hospital. The dataset was divided into three training stages for active learning. The evaluation metrics were assessed through the Dice similarity coefficient (DSC) and mean absolute difference. The overall internal test set DSCs from the multi-label, single-label, and pair-label models were 95%, 91% (paired t-test; p = 0.01), and 93% (p = 0.03), respectively. The DSC of the IAN in the internal and external datasets increased from 83% to 79%, 87% and 81%, to 90% and 86% for the single-label, pair-label, and multi-label models, respectively (all p = 0.01). Prediction accuracy improved over time, significantly reducing the manual correction time. Our active learning and multi-label strategies facilitated accurate automatic segmentation.
format Article
id doaj-art-150bd5ec739b43cfbe6aea2e2cab8aa6
institution DOAJ
issn 2045-2322
language English
publishDate 2025-03-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-150bd5ec739b43cfbe6aea2e2cab8aa62025-08-20T02:59:20ZengNature PortfolioScientific Reports2045-23222025-03-0115111010.1038/s41598-025-91725-2Improving accuracy for inferior alveolar nerve segmentation with multi-label of anatomical adjacent structures using active learning in cone-beam computed tomographySungchul On0Junhyeok Ock1Myungsoo Bae2Jae-Woo Park3Seung-Hak Baek4Sungwon Ham5Namkug Kim6Department of Convergence Medicine, Asan Medical Institute of Convergence Science and Technology, Asan Medical Center, University of Ulsan College of MedicineDepartment of Convergence Medicine, Asan Medical Institute of Convergence Science and Technology, Asan Medical Center, University of Ulsan College of MedicineDepartment of Convergence Medicine, Asan Medical Institute of Convergence Science and Technology, Asan Medical Center, University of Ulsan College of MedicineDepartment of Convergence Medicine, Asan Medical Institute of Convergence Science and Technology, Asan Medical Center, University of Ulsan College of MedicineDepartment of Orthodontics, School of Dentistry, Seoul National UniversityHealthcare Readiness Institute for Unified Korea, Korea Univerisity Ansan Hospital, Korea University College of MedicineDepartment of Convergence Medicine, Asan Medical Institute of Convergence Science and Technology, Asan Medical Center, University of Ulsan College of MedicineAbstract Recent advancements in deep learning have revolutionized digital dentistry, highlighting the importance of precise dental segmentation. This study leverages active learning with the three-dimensional (3D) nnU-net and multi-labels to improve segmentation accuracy of dental anatomies, including the maxillary sinuses, maxilla, mandible, and inferior alveolar nerves (IAN), which are important for implant planning, in 3D cone-beam computed tomography (CBCT) scans. Segmentation accuracy was compared using single-label, adjacent pair-label, and multi-label relevant anatomic structures with 60 CBCT scans from Kooalldam Dental Hospital and externally validated using data from Seoul National University Dental Hospital. The dataset was divided into three training stages for active learning. The evaluation metrics were assessed through the Dice similarity coefficient (DSC) and mean absolute difference. The overall internal test set DSCs from the multi-label, single-label, and pair-label models were 95%, 91% (paired t-test; p = 0.01), and 93% (p = 0.03), respectively. The DSC of the IAN in the internal and external datasets increased from 83% to 79%, 87% and 81%, to 90% and 86% for the single-label, pair-label, and multi-label models, respectively (all p = 0.01). Prediction accuracy improved over time, significantly reducing the manual correction time. Our active learning and multi-label strategies facilitated accurate automatic segmentation.https://doi.org/10.1038/s41598-025-91725-2Active learningCone-beam computed tomographyDental segmentationInferior alveolar nerveMulti-label segmentation
spellingShingle Sungchul On
Junhyeok Ock
Myungsoo Bae
Jae-Woo Park
Seung-Hak Baek
Sungwon Ham
Namkug Kim
Improving accuracy for inferior alveolar nerve segmentation with multi-label of anatomical adjacent structures using active learning in cone-beam computed tomography
Scientific Reports
Active learning
Cone-beam computed tomography
Dental segmentation
Inferior alveolar nerve
Multi-label segmentation
title Improving accuracy for inferior alveolar nerve segmentation with multi-label of anatomical adjacent structures using active learning in cone-beam computed tomography
title_full Improving accuracy for inferior alveolar nerve segmentation with multi-label of anatomical adjacent structures using active learning in cone-beam computed tomography
title_fullStr Improving accuracy for inferior alveolar nerve segmentation with multi-label of anatomical adjacent structures using active learning in cone-beam computed tomography
title_full_unstemmed Improving accuracy for inferior alveolar nerve segmentation with multi-label of anatomical adjacent structures using active learning in cone-beam computed tomography
title_short Improving accuracy for inferior alveolar nerve segmentation with multi-label of anatomical adjacent structures using active learning in cone-beam computed tomography
title_sort improving accuracy for inferior alveolar nerve segmentation with multi label of anatomical adjacent structures using active learning in cone beam computed tomography
topic Active learning
Cone-beam computed tomography
Dental segmentation
Inferior alveolar nerve
Multi-label segmentation
url https://doi.org/10.1038/s41598-025-91725-2
work_keys_str_mv AT sungchulon improvingaccuracyforinferioralveolarnervesegmentationwithmultilabelofanatomicaladjacentstructuresusingactivelearninginconebeamcomputedtomography
AT junhyeokock improvingaccuracyforinferioralveolarnervesegmentationwithmultilabelofanatomicaladjacentstructuresusingactivelearninginconebeamcomputedtomography
AT myungsoobae improvingaccuracyforinferioralveolarnervesegmentationwithmultilabelofanatomicaladjacentstructuresusingactivelearninginconebeamcomputedtomography
AT jaewoopark improvingaccuracyforinferioralveolarnervesegmentationwithmultilabelofanatomicaladjacentstructuresusingactivelearninginconebeamcomputedtomography
AT seunghakbaek improvingaccuracyforinferioralveolarnervesegmentationwithmultilabelofanatomicaladjacentstructuresusingactivelearninginconebeamcomputedtomography
AT sungwonham improvingaccuracyforinferioralveolarnervesegmentationwithmultilabelofanatomicaladjacentstructuresusingactivelearninginconebeamcomputedtomography
AT namkugkim improvingaccuracyforinferioralveolarnervesegmentationwithmultilabelofanatomicaladjacentstructuresusingactivelearninginconebeamcomputedtomography