ESTRO 2025 - Abstract Book

S3049

Physics - Image acquisition and processing

ESTRO 2025

ventilated lung tissue in cancer patients, and could ultimately improve clinical outcomes such as incidence, severity and persistence of radiation induced lung injury.

Keywords: MRI; lung cancer; ventilation

References: 1. Marks, L.B., et al., Radiation Dose-Volume Effects in the Lung. IJROBP, 2010. 76 (3 SUPPL.): p. S70-S76. 2. Bucknell, N., et al., Single-arm prospective interventional study assessing feasibility of using gallium-68 ventilation and perfusion PET/CT to avoid functional lung in patients with stage III non-small cell lung cancer. BMJ Open, 2020. 10 (12): p. e042465. 3. Rankine, L.J., et al., Hyperpolarized (129)Xe Magnetic Resonance Imaging for Functional Avoidance Treatment Planning in Thoracic Radiation Therapy. IJROBP, 2021. 111 (4): p. 1044-1057. 4. Neal, M.A. et al. Dynamic 19F-MRI of pulmonary ventilation in lung transplant recipients with and without chronic lung allograft dysfunction. JHLT Open, Volume 7, 100167

3436

Mini-Oral Assessing a novel photon counting computed tomography for quantitative imaging in radiotherapy Didier Lustermans 1 , Guillaume Landry 2 , Cécile Jeukens 3 , Frank Verhaegen 1 , Gabriel Paiva Fonseca 1 1 Department of Radiation Oncology (Maastro), GROW Research Institute for Oncology and Reproduction, Maastricht University Medical Centre+, Maastricht, Netherlands. 2 Department of Radiation Oncology, LMU University Hospital, Munich, Germany. 3 Department of Radiology, Maastricht University Medical Centre+, Maastricht, Netherlands Purpose/Objective: Photon counting computed tomography (PCCT) represents a significant advancement in medical imaging, offering increased resolution and counting individual photons with different energies. While PCCT currently demonstrates its clinical utility in diagnostic imaging, its potential in radiotherapy for quantitative imaging is mostly unexplored. It is expected to enhance delineation and detection, provide multi-energy spectral imaging, and extract precise quantitative data, potentially improving radiotherapy workflows. Specifically, reducing proton relative stopping power (RSP) estimation uncertainties. This study assessed the relative electron density (RED), effective atomic number (Z eff ) and RSP estimation errors in a clinical PCCT and evaluated various spectra, calibration methods, and phantom sizes, comparing it to dual-energy CT (DECT), to highlight the importance in radiotherapy. Material/Methods: Images were acquired of the Multi-Energy CT phantom (Model 1472; Sun Nuclear), with tissue-equivalent materials. PCCT scans were performed on the NAEOTOM Alpha (Siemens Healthineers) at 120 and 140kVp spectra (40mGy), with virtual monoenergetic images (VMIs) reconstructed in the ranges of 40-70keV and 150-190keV (10keV increments). Performance was compared to DECT acquisition using the SOMATOM Confidence (Siemens Healthineers) with similar dose-levels, where spectral separation was achieved by scanning at 80kVp and 140kVp and, additionally, VMIs were reconstructed. AMIGOpy, an open-source software, used and compared two established formalisms; Saito and Hunemohr. Both use fits of known materials to estimate RED, Z eff and excitation energy from low- and high-energy pairs, which is used for RSP estimation by the Bethe-Bloch formula. Optimal VMI pair selection was determined by quantifying the residual root-mean-squared-error (RMSE) averaged over the inserts for each individual property. Reference values were based on insert composition. Tissue differentiation, essential for radiotherapy, was assessed by plotting RED/Z eff of calcium and iodine concentration inserts for both scanners.

Results: VMI-pairs with low-energy keV of 50-70 demonstrated the most accurate estimations (Figure 1). The Saito formalism