Unsupervised domain adaptation based on feature and edge alignment for femur X-ray image segmentation
Computerized Medical Imaging and Graphics · 2024-09 · 116 · 102407
DOI: 10.1016/j.compmedimag.2024.102407 ↗Kurz gefasst
Fachartikel in der Zeitschrift Computerized Medical Imaging and Graphics, 2024-09. Verfasst von Xiaoming Jiang und 9 weiteren Autor:innen. Beteiligte Einrichtung: Chongqing Engineering Research Center of Medical Electronics and Information Technology, Chongqing University of Post and Telecommunications, Chongqi…. Hauptthemen laut MEDLINE: Absorptiometry, Photon, Femur, Bone Density. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 46 Literaturangaben. Gefördert von: Peking University, National Natural Science Foundation of China, Peking University Third Hospital.
Abstract (Originalsprache)
The gold standard for diagnosing osteoporosis is bone mineral density (BMD) measurement by dual-energy X-ray absorptiometry (DXA). However, various factors during the imaging process cause domain shifts in DXA images, which lead to incorrect bone segmentation. Research shows that poor bone segmentation is one of the prime reasons of inaccurate BMD measurement, severely affecting the diagnosis and treatment plans for osteoporosis. In this paper, we propose a Multi-feature Joint Discriminative Domain Adaptation (MDDA) framework to improve segmentation performance and the generalization of the network in domain-shifted images. The proposed method learns domain-invariant features between the source and target domains from the perspectives of multi-scale features and edges, and is evaluated on real data from multi-center datasets. Compared to other state-of-the-art methods, the feature prior from the source domain and edge prior enable the proposed MDDA to achieve the optimal domain adaptation performance and generalization. It also demonstrates superior performance in domain adaptation tasks on small amount datasets, even using only 5 or 10 images. In this study, MDDA provides an accurate bone segmentation tool for BMD measurement based on DXA imaging.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Chongqing Engineering Research Center of Medical Electronics and Information Technology, Chongqing University of Post and Telecommunications, Chongqing, China.
- Department of Sports Medicine, Peking University Third Hospital, Institute of Sports Medicine of Peking University, Beijing Key Laboratory of Sports Injuries, Engineering Research Center of Sports Trauma Treatment Techn…
- Department of Foot and Ankle Surgery, Wuhan Fourth Hospital, Wuhan, Hubei, China.
- National Health Commission Capacity Building and Continuing Education Center, Beijing, China.
- Chinese Academy of Sciences Key Laboratory of Molecular Imaging, Institute of Automation, Beijing 100190, China. Electronic address: lizhi.shao@ia.ac.cn.
- School of Data Science and Artificial Intelligence, Wenzhou University of Technology, Wenzhou 325000, China. Electronic address: wangmingjing.style@gmail.com.
Förderung
- Peking University
- National Natural Science Foundation of China
- Peking University Third Hospital
- Beijing Natural Science Foundation
- China Postdoctoral Science Foundation
Zitieren
APA
Jiang, X., Yang, Y., Su, T., Xiao, K., Lu, L., Wang, W., Guo, C., Shao, L., Wang, M., & Jiang, D. (2024). Unsupervised domain adaptation based on feature and edge alignment for femur X-ray image segmentation. Computerized Medical Imaging and Graphics, 116, 102407. https://doi.org/10.1016/j.compmedimag.2024.102407
Vancouver
Jiang X, Yang Y, Su T, Xiao K, Lu L, Wang W, et al. Unsupervised domain adaptation based on feature and edge alignment for femur X-ray image segmentation. Computerized Medical Imaging and Graphics. 2024;116:102407. doi:10.1016/j.compmedimag.2024.102407
BibTeX
@article{Jiang2024,
author = {Jiang, Xiaoming and Yang, Yongxin and Su, Tong and Xiao, Kai and Lu, LiDan and Wang, Wei and Guo, Changsong and Shao, Lizhi and Wang, Mingjing and Jiang, Dong},
title = {Unsupervised domain adaptation based on feature and edge alignment for femur X-ray image segmentation},
journal = {Computerized Medical Imaging and Graphics},
year = {2024},
volume = {116},
pages = {102407},
doi = {10.1016/j.compmedimag.2024.102407}
}
EUVIMED