ESTRO 2025 - Abstract Book

S2877

Physics - Dose prediction, optimisation and applications of photon and electron planning

ESTRO 2025

To test this workflow, we performed dose measurements with a Delta4 Phantom+ MR (ScandiDos AB, Sweden) with time-resolved readouts at 40Hz, placed on a Quasar Motion MR Platform (IBA Quasar, Canada). The dose measurements were compared to the live calculation of the accumulated dose for three different rigid motion traces. Platform-reported phantom positions were considered ground truth. Results: In Figure 2 we observe good agreement between the measurements (i.e. diodes between 60% and 85% of the max dose) and our live calculation of the accumulated dose. Furthermore, we can observe the error when not taking the rigid translation into account (i.e. the red-shaded area). From the predicted coverage (V95% of the planned dose) we could determine a point in the treatment where we expect not to reach the desired coverage and perform a baseline shift correction. Noticeably, the rate of predicted dose values increases towards the end of treatment since fewer remaining segments need to be calculated. On average calculating a segment took 152ms, trailing the live treatment by only 1 segment.