Validation of the electron return effect in a low-field MR-linac using a novel phantom.

Document Type

Conference Proceeding

Publication Date


Publication Title

Med Phys


Purpose: The presence of a magnetic field may adversely impact the dose distribution, particularly at interfaces. While simulation studies have been conducted, it can be challenging to validate these effects in vivo. This study evaluates the electron return effect (ERE) caused by the presence of 0.35 T magnetic field on a 6 MV flattering-filter free (FFF) beam in a novel pelvic end to end phantom (PETE) with variable air-filled rectal status. Methods: The MR-compatible PETE phantom consists of variable fluid-filled bladder, air-filled rectum, and mold-cast anthropomorphic pelvic bones. CT simulations were conducted on the MR-compatible PETE phantom at rectal volumes ranging from 30 to 150 cc. PA, AP-PA, and 4-field box treatment plans with fixed (∼10×10 cm2) field sizes for each beam were generated for all rectal filling levels using a Monte Carlo-based dose calculation and a 1mm dose grid. Monitor units were fixed for each plan type. Plans were generated with and without the magnetic field effect simulated. Local differences were assessed at the rectal/tissue interface of the PETE phantom for each plan. Results: The PETE phantom enabled the calculation and simulation of the ERE for varied rectal filling states. Considerable hot spots were observed near the rectal air interface for the PA beam study. For rectum size 150 cc with 0.35T field, a slight increase by 14% at the entrance of the cavity followed by a striking decrease of 41% in term of relative dose. The same trend was observed for the 30cc air volume (15% increase at entrance followed by 28% decrease). Use of opposing beams (AP-PA) and 4-field box drastically reduced the ERE. Conclusion: An MR-compatible pelvic phantom with variable rectal air was implemented to assess the ERE. Future work will involve dosimetric validation using MR-compatible chambers and radiochromic film at the interface to verify simulated results.





First Page


This document is currently not available here.