ESTRO 2022 - Abstract Book

S1396

Abstract book

ESTRO 2022

Results All BioXmark ® fiducials were recognized by Cyberknife ® tracking system. Longitudinal, vertical and lateral offsets presented by 50µL, 100µL BioXmark ® fiducials compared to standard metal markers showed minor, insignificant differences. However 200 µL BioXmark ® fiducials longitudinal offset was 0.57 mm which is considered as significant (Table 1, Graph 1). Differences in rotational offsets showed uncertain results, that might be affected by uncertainty error which was higher for BioXmark ® compared to reference fiducials.

Graph 1

Conclusion The BioXmark ® can be recognized and followed by CyberKnife ® tracking system, however significant differences were observed compared to standard fiducials. Smaller volume of the fiducial tend to show smaller errors compared to reference fiducials, however more measurements are needed to show statistically significant results. Additional research with smaller marker volumes and dosimetric films is required for more precise evaluation of tracking abilities of BioXmark ® fiducials prior to clinical application.

PO-1609 Generating synthetic hypoxia images from FDG-PET using Generative Adversarial Networks (GANs)

A. Traverso 1 , C. Rao 1 , A. Briassouli 2 , A. Dekker 1 , D. De Ruysscher 1 , W. van Elmpt 1

1 Maastro Clinic, Department of Radiotherapy, Maastricht, The Netherlands; 2 Maastricht University, Department of Knowledge Engineering , Maastricht, The Netherlands

Purpose or Objective

Hypoxia imaging can provide important information on tumour biology such as radio resistance. However, hypoxia imaging based on HX4-PET (Positron Emission Tomography) is not routinely acquired because of its higher costs and acquisition times compared to standard FDG-PET. We used deep learning models based on GANs to synthesise HX4-PET images from FDG-PET and Compute Tomography (CT) scans. Materials and Methods We applied image translation GANs to synthesise HX4-PET images from FDG-PET and panning CT scans of 34 non small cell lung cancer patients from two clinical trials: the PET-Boost trial (registration number NCT01024829, n = 15) and