ESTRO 2025 - Abstract Book

S3028

Physics - Image acquisition and processing

ESTRO 2025

Results: All linear correlations between T 1 and T 2 MRF and GS values measured on the phantom were high (R 2 >0.99, p-value <0.001). Overall, the mean absolute relative error for T 1 and T 2 decreased with the number of repetitions ( T 1 : from 5.7% to 4.4% without pause and from 6.9% to 5.5% with it; T 2 : from 13.8% to 10.2% without pause and from 14.6% to 13.4% with it). As the scheme without pause achieved better T 1 and T 2 estimation, this scheme was used for the volunteer scan. A similar trend was observed, and while T 1 error remained large after twenty repetitions (38.5% and 32.8% in the left and right parotid glands, respectively), the T 2 error decreased significantly, reaching below 10% for both parotid glands (Figure 1).

Conclusion: Consecutively repeating the MRF sequence can improve the accuracy of T 1 and T 2 mapping with MR fingerprinting on an MR-Linac. This could prove beneficial in evaluating the treatment response in the tumour and its potential side effects on organs at risk.

Keywords: MR-guided radiotherapy, quantitative imaging

References: [1] Van Houdt, P. J., Yang, Y. & Van Der Heide, U. A. Quantitative Magnetic Resonance Imaging for Biological Image-Guided Adaptive Radiotherapy. Front. Oncol. 10, 615643 (2021). 10.3389/fonc.2020.615643 [2] Ma, D. et al. Magnetic resonance fingerprinting . Nature 495, 187–192 (2013). 10.1038/nature11971 [3] Bruijnen, T. et al. Technical feasibility of magnetic resonance fingerprinting on a 1.5T MRI-linac. Phys. Med. Biol. 65, 22NT01 (2020). 10.1088/1361-6560/abbb9d [4] Goodburn, R. et al. Measuring the Gradient Impulse Response Function (GIRF) for UTE Imaging on an MR-Linac. (2021). [5] Ma, D. et al. Slice profile and B 1 corrections in 2D magnetic resonance fingerprinting. Magn. Reson. Med. 78, 1781– 1789 (2017). 10.1002/mrm.26580