Saved in:
Bibliographic Details
Main Authors: Ji, Zongrui, Cui, Zhiming, Li, Na, Zheng, Qianhan, Shi, Miaojing, Deng, Ke, Zhang, Jingyang, Li, Chaoyuan, Chen, Xuepeng, Dong, Yi, Ma, Lei
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2511.16936
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914166141353984
author Ji, Zongrui
Cui, Zhiming
Li, Na
Zheng, Qianhan
Shi, Miaojing
Deng, Ke
Zhang, Jingyang
Li, Chaoyuan
Chen, Xuepeng
Dong, Yi
Ma, Lei
author_facet Ji, Zongrui
Cui, Zhiming
Li, Na
Zheng, Qianhan
Shi, Miaojing
Deng, Ke
Zhang, Jingyang
Li, Chaoyuan
Chen, Xuepeng
Dong, Yi
Ma, Lei
contents Background:Accurate tooth segmentation from cone beam computed tomography (CBCT) images is crucial for digital dentistry but remains challenging in cases of interdental adhesions, which cause severe anatomical shape distortion. Methods: To address this, we propose a deep learning framework that integrates semantic and shape awareness for shape-preserving segmentation. Our method introduces a target-tooth-centroid prompted multi-label learning strategy to model semantic relationships between teeth, reducing shape ambiguity. Additionally, a tooth-shape-aware learning mechanism explicitly enforces morphological constraints to preserve boundary integrity. These components are unified via multi-task learning, jointly optimizing segmentation and shape preservation. Results: Extensive evaluations on internal and external datasets demonstrate that our approach significantly outperforms existing methods. Conclusions: Our approach effectively mitigates shape distortions and providing anatomically faithful tooth boundaries.
format Preprint
id arxiv_https___arxiv_org_abs_2511_16936
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Shape-preserving Tooth Segmentation from CBCT Images Using Deep Learning with Semantic and Shape Awareness
Ji, Zongrui
Cui, Zhiming
Li, Na
Zheng, Qianhan
Shi, Miaojing
Deng, Ke
Zhang, Jingyang
Li, Chaoyuan
Chen, Xuepeng
Dong, Yi
Ma, Lei
Computer Vision and Pattern Recognition
Background:Accurate tooth segmentation from cone beam computed tomography (CBCT) images is crucial for digital dentistry but remains challenging in cases of interdental adhesions, which cause severe anatomical shape distortion. Methods: To address this, we propose a deep learning framework that integrates semantic and shape awareness for shape-preserving segmentation. Our method introduces a target-tooth-centroid prompted multi-label learning strategy to model semantic relationships between teeth, reducing shape ambiguity. Additionally, a tooth-shape-aware learning mechanism explicitly enforces morphological constraints to preserve boundary integrity. These components are unified via multi-task learning, jointly optimizing segmentation and shape preservation. Results: Extensive evaluations on internal and external datasets demonstrate that our approach significantly outperforms existing methods. Conclusions: Our approach effectively mitigates shape distortions and providing anatomically faithful tooth boundaries.
title Shape-preserving Tooth Segmentation from CBCT Images Using Deep Learning with Semantic and Shape Awareness
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2511.16936