EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Polychromatic neural CBCT reconstruction through density-attenuation modeling

Lukas Birklein iD, Elmar Schömer iD, Ulrich Schwanecke iD, Ralf Schulze iD

Physics in Medicine & Biology · 2025

DOI: 10.1088/1361-6560/ae28af ↗

Kurz gefasst

Fachartikel in der Zeitschrift Physics in Medicine & Biology, 2025. Verfasst von Lukas Birklein und 3 weiteren Autor:innen. Beteiligte Einrichtung: Johannes Gutenberg University, Mainz, Germany. Hauptthemen laut MEDLINE: Cone-Beam Computed Tomography, Neural Networks, Computer, Image Processing, Computer-Assisted. Die Arbeit nennt 54 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Monochromatic cone beam computed tomography reconstruction algorithms are still most prominent in practice. Since the x-ray detectors of today’s machines are mostly energy integrating detectors and thus not able to resolve photon energy levels, reconstruction artifacts, often termed as beam-hardening artifacts, are a frequent observation. We propose a polychromatic 3D reconstruction technique, using a coordinate based neural representation, that requires zero additional prior information, in which we model attenuation related to a fixed energy level E 0 together with its derivative. We concurrently optimize an intermediate density value for each spatial position together with a composite attenuation function. It assigns each intermediate density value its attenuation coefficient and computes the derivative of the attenuation coefficient w.r.t. the energy. We implement this function as a neural network that models a monotonic relationship between intermediate density and attenuation. In our results we can show that the proposed method is able to considerably improve reconstruction quality in various scenarios, both quantitatively using a synthetic numerical phantom as well as visual quality in real-world clinical examples.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Cone-Beam Computed TomographyNeural Networks, ComputerImage Processing, Computer-AssistedPhantoms, ImagingHumansImaging, Three-DimensionalAlgorithms

Stichwörter der Autor:innen

CBCTartificial neural networkbeam hardeningcone beam CTpolychromaticreconstruction

Autor:innen und Institutionen

  • Johannes Gutenberg University, Mainz, Germany.
  • Rhein-Main University of Applied Sciences, Wiesbaden, Germany.
  • University of Bern, Bern, Switzerland.

ORCID-Profile: Lukas Birklein, Elmar Schömer, Ulrich Schwanecke, Ralf Schulze

Zitieren

APA

Birklein, L., Schömer, E., Schwanecke, U., & Schulze, R. (2025). Polychromatic neural CBCT reconstruction through density-attenuation modeling. Physics in Medicine & Biology. https://doi.org/10.1088/1361-6560/ae28af

Vancouver

Birklein L, Schömer E, Schwanecke U, Schulze R. Polychromatic neural CBCT reconstruction through density-attenuation modeling. Physics in Medicine & Biology. 2025. doi:10.1088/1361-6560/ae28af

BibTeX
@article{Birklein2025,
  author = {Birklein, Lukas and Schömer, Elmar and Schwanecke, Ulrich and Schulze, Ralf},
  title = {Polychromatic neural CBCT reconstruction through density-attenuation modeling},
  journal = {Physics in Medicine & Biology},
  year = {2025},
  doi = {10.1088/1361-6560/ae28af}
}
Kostenlos starten →