Click here to


Are you sure ?

Yes, do it No, cancel

Towards Time Resolved 3D Dosimetry in MRI-Linacs: Scintillation Images Surrogate Dose Distribution with High Accuracy

P Bruza1*, J Andreozzi1 , O Green2 , J Cammin2 , D Gladstone1,3 , B Pogue1 , (1) Dartmouth College, Hanover, NH, (2) Washington University, School of Medicine, St. Louis, MO, (3) Dartmouth-Hitchcock Medical Center, Lebanon, NH


(Tuesday, 7/31/2018) 4:30 PM - 6:00 PM

Room: Room 207

Purpose: Sharp dose gradients achieved with MRI-Linac systems enable highly confined radiation delivery. Millimeter resolution, MRI compatible dosimeters are therefore needed for pre-treatment quality assurance. The purpose of this study was to assess the performance and accuracy of a gated scintillation camera as a part of a high resolution time resolved 3D dosimetric system.

Methods: A TG-119 C-shape plan was delivered by ViewRay MR-linac (0.34T primary magnetic field, 6MV FFF beams) to a cylindrical scintillator phantom (Bicron BC412, 127mm(d)x57mm(l)). Intensified gated complementary metal-oxide semiconductor (CMOS) camera (DoseOptics LLC., Hanover, NH) was coupled with a 135mm lens, recording laterally-projected distributions of scintillation through the base of the cylindrical phantom. The camera gating was synchronized to linac X-ray pulses using a novel scatter X-ray trigger module. Projected scintillation images were recorded at a rate of 20 frames per second. An algorithm to merge the laterally-projected scintillation images with 3D geometrical beam volume, based on perspective-projected MLC shape, was developed in Matlab (MathWorks Inc., Natick, MA).

Results: Spatial resolution of 1 mm was achieved across the whole interaction area. Gamma analysis of 2D dose projections at 3%/3mm criteria yielded 98% pass rate between measured and calculated cumulative dose distributions. High gradient regions were reproduced with a precision of < 3% relative dose. Gamma analysis revealed a mismatch in the central low-dose region, indicating an overdose by 35%. This may be attributed to the luminescence re-absorption effect, and will be investigated further. Electron return effect was observed at the phantom-air boundary, and it contributed to the cumulative dose profile.

Conclusion: Measured laterally-projected scintillation images agreed with TPS calculation. Scintillation images can be used as a surrogate to depth dose distribution and can enable beamlet-by-beamlet 3D dose reconstruction of MRI-Linac beams.

Funding Support, Disclosures, and Conflict of Interest: P. Bruza is a principal investigator in SBIR subaward B02463, Video Rate optical verification tool for radiotherapy treatment plans, prime award NCI R44CA199681, DoseOptics LLC. B. Pogue is founder and president of DoseOptics LLC. This work has been sponsored by NIH research grants R44CA199836, R01EB023909, and R44CA199681.


Scintillators, Dosimetry, 3D


IM- X-ray: Radiation dosimetry & risk

Contact Email