ESTRO 2025 - Abstract Book

S3017

Physics - Image acquisition and processing

ESTRO 2025

Conclusion: The present study displayed clinically important tumor ADC differences between DS and MRL, probably owning to the signal-to-noise optimization and the intrinsic variation in the diffusion and echo times between protocols. Such optimization may introduce variations in the tumor ADC values, attributed to the presence of multiple tissue compartments with different diffusivities and T2-relaxation within the image voxels [1-3]. This highlights the importance of fixating the imaging parameters when building DWI protocols for transitioning of candidates from DS to MRL or when conducting multicenter studies utilizing ADC values as a quantitative biomarker for treatment response assessment.

Keywords: MRI, DWI, ADC

References: [1] H.H. Lee, E. Fieremans, D.S. Novikov, What dominates the time dependence of diffusion transverse to axons: Intra- or extra-axonal water?, Neuroimage, 182 (2018) 500-510. [2] W. Qin, C.S. Yu, F. Zhang, X.Y. Du, H. Jiang, Y.X. Yan, K.C. Li, Effects of echo time on diffusion quantification of brain white matter at 1.5 T and 3.0 T, Magn Reson Med, 61 (2009) 755-760. [3] J. Veraart, D.S. Novikov, E. Fieremans, TE dependent Diffusion Imaging (TEdDI) distinguishes between compartmental T(2) relaxation times, Neuroimage, 182 (2018) 360-369.

2420

Poster Discussion Standardizing MR intensities for quantitative longitudinal analyses of salivary glands during MRI-guided radiotherapy of head & neck cancer patients Philipp Wallimann, Riccardo Dal Bello, Matthias Guckenberger, Stephanie Tanadini-Lang, Panagiotis Balermpas Department of Radiation Oncology, University Hospital Zurich and University of Zurich, Zurich, Switzerland Purpose/Objective: Radiotherapy at hybrid Magnetic Resonance Linear Accelerators (MR-Linacs) allows for assessment of functional and anatomical changes during treatment. We present an image standardization procedure for the quantitative analysis of anatomical MR images acquired with an MR-Linac during radiotherapy for head-and-neck cancer patients. We report changes over time in the salivary glands and relate them to the delivered dose. Material/Methods: The developed image standardization procedure consisted of an N4 bias field correction 1 and an N-Peaks intensity normalization technique 2 , which determined the intensity level of background, neck muscles and fat, and aligned those intensity values across all images. For 42 patients with head & neck cancer treated with MR guided radiotherapy in 35 fractions on a 0.35T MR-Linac (ViewRay, Denver, CO, USA), we analyzed one balanced steady-state free precession (bSSFP) image per week, on which the salivary glands were manually recontoured and the treatment plan was adapted. For the submandibular and parotid glands, we determined the average change per fraction of the structure volume and median image intensity with and without the standardization, assuming a linear change with the fraction number. Spearman correlations were investigated between the volume changes and intensity changes and between each of them and the median accumulated dose to the structure. We performed eight tests, thus chose a Bonferroni-corrected significance threshold of 0.05/8=0.00625. Results: The mean (±SD) change in median standardized intensity per fraction across all patients was 0.37±0.32 for the submandibular glands and 0.17±0.52 for the parotid glands. There was a significant correlation between the standardized intensity change in the submandibular glands over time and the dose delivered to them, but only after image standardization (figure 1).