A Novel Method Combining Global and Local Assessments to Evaluate CBCT-Based Synthetic CTs

Chelsea Sargeant, Andrew Green, Jane Shortall, Robert Chuter, Jiaofeng Xu, Daniel Thill, Nicolette O’Connell, Alan Mcwilliam

Research output: Chapter in Book/Report/Conference proceedingChapterpeer-review


Deep learning models are increasingly used to generate synthetic images. Synthetic CTs (sCTs) generated from on-treatment cone-beam CTs (CBCTs) hold potential for adaptive radiotherapy, promising a high-quality representation of daily anatomy without requiring additional imaging or dose to the patient. However, validating sCT is very challenging as an accurate and appropriate ground truth is hard to come by in medical imaging. Current global metrics in the literature fail to provide a complete picture of how accurate synthetic images are. We introduce a novel method to evaluate sCTs utilising global error assessment and a local, voxel-wise statistical assessment of the sCT and the current ground truth, a deformably registered CT (dCT). Our methodology allows for the identification of individual cases where the sCT might offer an improved representation of the daily anatomy due to changes that occur over time, as well as showing regions where either the model or image registration under-performs. Our methodology can be used to guide future model development to improve the mapping between modalities, and also assist in deciphering when it is most appropriate to choose a sCT for image guided radiotherapy over the existing standard, the dCT.
Original languageEnglish
Title of host publicationSimulation and Synthesis in Medical Imaging
Subtitle of host publication7th International Workshop, SASHIMI 2022, Held in Conjunction with MICCAI 2022, Singapore, September 18, 2022, Proceedings
EditorsCan Zhao, David Svoboda, Jelmer M. Wolterink, Maria Escobar
Place of PublicationCham
PublisherSpringer Cham
Number of pages10
ISBN (Electronic)9783031169809
ISBN (Print)9783031169793
Publication statusPublished - 22 Sept 2022

Publication series

NameLecture Notes in Computer Science
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


  • Synthetic CT evaluation method
  • Adaptive radiotherapy
  • Image synthesis validation


Dive into the research topics of 'A Novel Method Combining Global and Local Assessments to Evaluate CBCT-Based Synthetic CTs'. Together they form a unique fingerprint.

Cite this