CT Synthesis from MRI Using GAN Architecture

In radiotherapy treatment, a CT volume with a spatially corresponding MR volume is required for dose planning as well as the sectioning of the tumour mass and organs at risk. Although MR images frequently offer better anatomical and functional information than CT images, CT images are typically util...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Navneet, Saumya, Naqvi, Najme Zehra
Format: Tagungsbericht
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:In radiotherapy treatment, a CT volume with a spatially corresponding MR volume is required for dose planning as well as the sectioning of the tumour mass and organs at risk. Although MR images frequently offer better anatomical and functional information than CT images, CT images are typically utilized in conjunction with MR images for image guidance and radiation treatment planning. However, the CT radiation used for imaging can raise cancer risks in the patients who are exposed to it. Therefore, for radiotherapy planning, researchers have recently been highly driven to use generative models or cross-modality adaption models to generate CT images from an equivalent MR image. Our study primarily focuses on the comparison of different cross-modality image synthesis techniques using generative adversarial networks, which create synthetic images in target imaging modality from actual source imaging modality images.
ISSN:0094-243X
1551-7616
DOI:10.1063/5.0183090