Cross-center Model Adaptive Tooth segmentation

Ruizhe Chen; Jianfei Yang; Huimin Xiong; Ruiling Xu; Yang Feng; Jian Wu; Zuozhu Liu

doi:10.1016/j.media.2024.103443

Cross-center Model Adaptive Tooth segmentation

Med Image Anal. 2024 Dec 27:101:103443. doi: 10.1016/j.media.2024.103443. Online ahead of print.

Authors

Ruizhe Chen¹, Jianfei Yang², Huimin Xiong³, Ruiling Xu⁴, Yang Feng⁵, Jian Wu⁶, Zuozhu Liu⁷

Affiliations

¹ Stomatology Hospital Affliated to Zhejiang University of Medicine, Zhejiang University, Hangzhou, 310016, China; ZJU-Angelalign R&D Center for Intelligence Healthcare, ZJU-UIUC Institute, Zhejiang University, Haining, 314400, China; Zhejiang Key Laboratory of Medical Imaging Artificial Intelligence, Zhejiang University, Hangzhou, 310058, China.
² School of Mechanical and Aerospace Engineering, Nanyang Technological University, Singapore, 639798, Singapore.
³ ZJU-Angelalign R&D Center for Intelligence Healthcare, ZJU-UIUC Institute, Zhejiang University, Haining, 314400, China; Zhejiang Key Laboratory of Medical Imaging Artificial Intelligence, Zhejiang University, Hangzhou, 310058, China.
⁴ ZJU-Angelalign R&D Center for Intelligence Healthcare, ZJU-UIUC Institute, Zhejiang University, Haining, 314400, China.
⁵ Angelalign Technology Inc., Shanghai, 200433, China.
⁶ Zhejiang Key Laboratory of Medical Imaging Artificial Intelligence, Zhejiang University, Hangzhou, 310058, China; State Key Laboratory of Transvascular Implantation Devices of The Second Affiliated Hospital, School of Medicine and School of Public Health, Zhejiang University, Hangzhou, 310058, China.
⁷ Stomatology Hospital Affliated to Zhejiang University of Medicine, Zhejiang University, Hangzhou, 310016, China; ZJU-Angelalign R&D Center for Intelligence Healthcare, ZJU-UIUC Institute, Zhejiang University, Haining, 314400, China; Zhejiang Key Laboratory of Medical Imaging Artificial Intelligence, Zhejiang University, Hangzhou, 310058, China. Electronic address: zuozhuliu@intl.zju.edu.cn.

PMID: 39778266
DOI: 10.1016/j.media.2024.103443

Abstract

Automatic 3-dimensional tooth segmentation on intraoral scans (IOS) plays a pivotal role in computer-aided orthodontic treatments. In practice, deploying existing well-trained models to different medical centers suffers from two main problems: (1) the data distribution shifts between existing and new centers, which causes significant performance degradation. (2) The data in the existing center(s) is usually not permitted to be shared, and annotating additional data in the new center(s) is time-consuming and expensive, thus making re-training or fine-tuning unfeasible. In this paper, we propose a framework for Cross-center Model Adaptive Tooth segmentation (CMAT) to alleviate these issues. CMAT takes the trained model(s) from the source center(s) as input and adapts them to different target centers, without data transmission or additional annotations. CMAT is applicable to three cross-center scenarios: source-data-free, multi-source-data-free, and test-time. The model adaptation in CMAT is realized by a tooth-level prototype alignment module, a progressive pseudo-labeling transfer module, and a tooth-prior regularized information maximization module. Experiments under three cross-center scenarios on two datasets show that CMAT can consistently surpass existing baselines. The effectiveness is further verified with extensive ablation studies and statistical analysis, demonstrating its applicability for privacy-preserving model adaptive tooth segmentation in real-world digital dentistry.

Keywords: Cross-center; Source-free Domain Adaptation; Tooth segmentation.