EUVIMEDEUVIMED entdecken
Crossref

ComptoNet: a Compton-map guided deep learning framework for multi-scatter estimation in multi-source stationary CT

Yingxian Xia iD, Li Zhang iD, Yuxiang Xing iD, Zhiqiang Chen, Hewei Gao iD

Physics in Medicine & Biology · 20.07.2025 · 70 (14) · 145022

DOI: 10.1088/1361-6560/adebd7 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Physics in Medicine & Biology, online erschienen am 16.07.2025. Verfasst von Yingxian Xia und 4 weiteren Autor:innen. Beteiligte Einrichtung: Key Laboratory of Particle and Radiation Imaging (Tsinghua University), Ministry of Education, Beijing 100084, People's Republic of China. Hauptthemen laut MEDLINE: Deep Learning, Tomography, X-Ray Computed, Image Processing, Computer-Assisted, Scattering, Radiation. Die Arbeit wurde bisher 8-mal zitiert (Crossref) und nennt 48 Literaturangaben. Gefördert von: National Key R&D Program of China, National Natural Science Foundation of China, National Key Research and Development Program of China.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Multi-source stationary computed tomography (MSS-CT) offers significant advantages in medical and industrial applications due to its gantryless scan architecture and capability of simultaneous multi-source emission. However, the lack of anti-scatter grid deployment in MSS-CT leads to severe forward and cross scatter contamination, necessitating accurate and efficient scatter correction. In this work, we propose ComptoNet, an innovative decoupled deep learning framework that integrates Compton-scattering physics with deep learning for scatter estimation in MSS-CT. The core innovation lies in the Compton-map, a representation of large-angle Compton scatter signals outside the scan field of view. ComptoNet employs a dual-network architecture: a conditional encoder–decoder network guided by reference Compton-maps and spare detector data for cross scatter estimation, and a frequency U-Net with attention mechanisms for forward scatter correction. Experiments on Monte Carlo-simulated data demonstrate ComptoNet’s superior performance, achieving a mean absolute percentage error of 0.84% on scatter estimation. After correction, CT images show nearly artifact-free quality for all test phantoms, validating ComptoNet’s robustness in mitigating scatter-induced errors across diverse photon counts and phantoms comparing with other methods.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningTomography, X-Ray ComputedImage Processing, Computer-AssistedScattering, RadiationMonte Carlo MethodPhantoms, Imaging

Stichwörter der Autor:innen

deep scatter estimationmulti-scatter estimationmulti-source stationary CT

Autor:innen und Institutionen

  • Key Laboratory of Particle and Radiation Imaging (Tsinghua University), Ministry of Education, Beijing 100084, People's Republic of China.
  • Department of Engineering Physics, Tsinghua University, Beijing 100084, People's Republic of China.

ORCID-Profile: Yingxian Xia, Li Zhang, Yuxiang Xing, Hewei Gao

Förderung

  • National Key R&D Program of China
  • National Natural Science Foundation of China
  • National Key Research and Development Program of China

Zitieren

APA

Xia, Y., Zhang, L., Xing, Y., Chen, Z., & Gao, H. (2025). ComptoNet: a Compton-map guided deep learning framework for multi-scatter estimation in multi-source stationary CT. Physics in Medicine & Biology, 70(14), 145022. https://doi.org/10.1088/1361-6560/adebd7

Vancouver

Xia Y, Zhang L, Xing Y, Chen Z, Gao H. ComptoNet: a Compton-map guided deep learning framework for multi-scatter estimation in multi-source stationary CT. Physics in Medicine & Biology. 2025;70(14):145022. doi:10.1088/1361-6560/adebd7

BibTeX
@article{Xia2025,
  author = {Xia, Yingxian and Zhang, Li and Xing, Yuxiang and Chen, Zhiqiang and Gao, Hewei},
  title = {ComptoNet: a Compton-map guided deep learning framework for multi-scatter estimation in multi-source stationary CT},
  journal = {Physics in Medicine & Biology},
  year = {2025},
  volume = {70},
  number = {14},
  pages = {145022},
  doi = {10.1088/1361-6560/adebd7}
}
Kostenlos starten →