A novel image‐domain denoising method for triple‐material decomposition in dual‐energy CT
Medical Physics · 2026-11 · 53 (11)
DOI: 10.1002/mp.70723 ↗Kurz gefasst
Fachartikel in der Zeitschrift Medical Physics, online erschienen am 14.11.2026. Verfasst von Jiexiong Wu und 2 weiteren Autor:innen. Beteiligte Einrichtung: Department of Engineering and Applied Physics, University of Science and Technology of China, Hefei, Anhui, People's Republic of China. Hauptthemen laut MEDLINE: Tomography, X-Ray Computed, Image Processing, Computer-Assisted, Signal-To-Noise Ratio. Die Arbeit nennt 42 Literaturangaben. Gefördert von: Ministry of Science and Technology of China Key Research and Development Projects, Fundamental Research Funds for the Central Universities, Hefei Comprehensive National Science Center.
Abstract (Originalsprache)
Background
Triple‐material decomposition (TMD) holds significant clinical value in dual‐energy computed tomography (DECT), as it enables simultaneous differentiation of bone, soft tissue, and contrast agents—three key components commonly targeted in clinical imaging assessments. However, the inherent noise amplification artifact associated with material decomposition remains a critical technical bottleneck that restricts its widespread clinical translation.
Purpose
To address this limitation, this study proposes an image‐domain denoising strategy to mitigate noise amplification, thereby improving the quantitative accuracy and qualitative quality of TMD in DECT.
Method
First, DECT images were transformed into amplitude and argument components via vectorization. Second, a spectral joint bilateral filtering algorithm was developed to specifically denoise the argument images, leveraging inter‐spectral correlation to preserve structural details while suppressing noise. Subsequently, denoised DECT images were reconstructed using the original amplitude and denoised argument components. TMD was performed with soft tissue, bone, and contrast agent as basis materials, utilizing two decomposition approaches: direct matrix inversion (DMI) and the penalized weighted least squares‐expectation maximization‐loop (PWLS‐EP‐LOOP) iterative method. The proposed denoising strategy was validated on a benchtop micro‐CT system, with evaluations conducted on both a customized anthropomorphic phantom and an in vivo mouse model injected with ExiTron nano 6000 contrast agent.
Results
For DMI‐based decomposition, the proposed method achieved noise reduction rates (NRRs) of 68.9% in soft tissue images, 74.4% in contrast‐enhanced images, and 78.7% in bone images. For PWLS‐EP‐LOOP‐based iterative decomposition, the corresponding NRRs were 22.4% (soft tissue), 48.6% (contrast agent), and 37.5% (bone). Furthermore, the number of iteration cycles required for convergence in the PWLS‐EP‐LOOP algorithm was reduced by 75.4%, substantially improving computational efficiency.
Conclusion
Experimental results demonstrate that the proposed image‐domain denoising method, tailored for DECT TMD, effectively suppresses noise in both DMI and iterative workflows while preserving structural fidelity. This work thus provides a promising solution to enhance the clinical utility of DECT material decomposition.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Engineering and Applied Physics, University of Science and Technology of China, Hefei, Anhui, People's Republic of China.
- Department of Radiation Oncology, Division of Life Sciences and Medicine, the First Affiliated Hospital of USTC, University of Science and Technology of China, Hefei, Anhui, People's Republic of China.
- Ion Medical Research Institute, University of Science and Technology of China, Hefei, Anhui, People's Republic of China.
Förderung
- Ministry of Science and Technology of China Key Research and Development Projects
- Fundamental Research Funds for the Central Universities
- Hefei Comprehensive National Science Center
Zitieren
APA
Wu, J., Yuan, X., & Yang, Y. (2026). A novel image‐domain denoising method for triple‐material decomposition in dual‐energy CT. Medical Physics, 53(11). https://doi.org/10.1002/mp.70723
Vancouver
Wu J, Yuan X, Yang Y. A novel image‐domain denoising method for triple‐material decomposition in dual‐energy CT. Medical Physics. 2026;53(11). doi:10.1002/mp.70723
BibTeX
@article{Wu2026,
author = {Wu, Jiexiong and Yuan, Xiaogang and Yang, Yidong},
title = {A novel image‐domain denoising method for triple‐material decomposition in dual‐energy CT},
journal = {Medical Physics},
year = {2026},
volume = {53},
number = {11},
doi = {10.1002/mp.70723}
}
EUVIMED