# Clinical performance of a shielded diode for small field relative dosimetry in a 1.5 T MRI‐linac using measurements and simulations

## TL;DR

This study compares the performance of different detectors for measuring radiation doses in small fields on MRI-guided linear accelerators, finding that the microDiamond detector provides the most accurate results.

## Contribution

This work is the first to investigate the Sun Nuclear EDGE shielded diode's performance in small field dosimetry under a 1.5T magnetic field using measurements and simulations.

## Key findings

- The EDGE detector under-represents the magnetic field's effect on profile shape.
- The microDiamond detector showed the best agreement with film and simulations for small field dosimetry.
- The Semiflex 3D ionization chamber over-represents the magnetic field's effect on dose profiles.

## Abstract

Small field dosimetry on magnetic‐resonance (MR) guided linear accelerators (linacs) has become increasingly more important for accurate beam modeling due to the increase in the number of patients and variety of disease sites treated on MR‐linacs such as those with small fields. Dosimetry in magnetic fields has been proven to be difficult due to the Lorentz force and intrinsic detector responses. For example, there is evidence in the literature that shielded diodes may cause a shift in the shape of the lateral dose distribution. However, there is no work in the literature investigating the shape change for the Sun Nuclear EDGE shielded diode.

The goal of this research was to use measurements and simulations to compare the performance of the EDGE diode to two other scanning detectors and film for small field relative dosimetry in a linac with an external magnetic field.

The detectors investigated in this work were the EDGE detector, the PTW 60019 microDiamond, the PTW 31021 Semiflex 3D, and Gafchromic EBT3 radiochromic film. The PTW Trufix system was used to align the sensitive volume of each detector to the radiation isocenter. To make this possible, a custom holder was manufactured for the EDGE detector which was compatible with Trufix. To enable comparison, the detector profile measurements were reproduced by Monte Carlo (MC) simulations. To simulate radiation transport in the measurement setup, detailed MC models of the examined detectors were built, together with a model of a 1.5 Tesla (T) Elekta Unity linac.

Radiochromic film was used as the ground truth for profile shape in this work. The MC beam spot size and collimation offset were tuned to match the film results. Profile shape was consistent between measurements and MC simulations for each of the scanning detectors. The EDGE detector was found to under‐represent the effect of the magnetic field on profile shape, which is consistent with findings of other shielded diodes in the literature. Repeated profiles indicate that the intra‐type variation of the active volume positioning is about 0.6 mm. The Semiflex 3D ionization chamber over‐represents the effect of the magnetic field on profiles. The microDiamond provided the best agreement to film and MC for small fields based on the profile, percent depth dose (PDD), and output factor (OF) results.

This study reinforces the importance of detector choice for characterizing small fields in the presence of magnetic fields. The EDGE detector was shown to misrepresent the shape of the crossline profile for small fields in a 1.5T magnetic field. Film is recommended for determining the dose distribution shape, but it is difficult to position precisely and thus cannot be used to determine the lateral shift with high precision. The microDiamond detector was determined to most accurately characterize both the lateral shift and shape of the lateral dose distribution of the detectors investigated. The investigated ion chamber and shielded silicon diode should be used with caution for small fields in the presence of magnetic fields due to perturbances to profile shape.