Dosimetric Responses of Parallel Plate Ionization Chambers in a Magnetic Field

Abstract

Radiation dosimetry has been an important issue since the discovery of X-ray. These days, the primary standards dosimetry laboratories (PSDLs) and international agencies (e.g., IAEA, WHO) disseminate the international calibration standards to users through a network of the secondary standards dosimetry laboratories (SSDL). For this purpose, various dosimeters and equipment have been developed to assure the quality of calibration traceability and accuracy. Among them, an ionization chamber has become the most relevant dosimeter for measurements of radiation doses due to its high precision and reliability. Therefore, ionization chambers have been used to provide reference dosimetry calibrations for radiotherapy.<br/> Dosimetry for magnetic resonance imaging (MRI)-guided radiotherapy has been an emerging issue in the medical radiation society. Currently developing MRI-guided radiotherapy machines adopt Co-60 or pulsed X-rays with a magnetic field of 0.35 to 1.5 T either perpendicular or parallel to the incident photon beams. The magnetic field influences the trajectories of secondary charged particles through the action of the Lorentz force such that the dose distribution in the sensitive volume of ionization chamber can be altered. Since PSDLs or SSDLs calibrate ionization chambers without a magnetic field, radiation dosimetry in a magnetic field using the ionization chambers needs an additional magnetic field correction factor. Since MRI-guided radiotherapy is still under development, these magnetic field correction factors are often calculated using Monte Carlo (MC) radiation transport codes integrated with an electromagnetic module.<br/> In order to evaluate the accuracy of electron transport algorithms of the MC codes at a static magnetic field of 0.35 to 3.0 T, the Fano cavity theorem was tested for mono-energetic electrons with energies ranging from 0.01 to 3 MeV. Four general-purpose Monte Carlo codes (EGSnrc, PENELOPE, MCNP6, and Geant4) were validated for this purpose. With transport parameters carefully selected, PENELOPE and MCNP6 could achieve the accuracy within 1.0% and 0.4%, respectively. Geant4 showed the accuracy within 1.7% except in 3.0 T. The accuracy of EGSnrc with the enhanced electromagnetic field macros was within 0.2%. Owing to its superior accuracy, the following simulations in this study were performed using EGSnrc with the enhanced electromagnetic field macros.<br/> The purpose of this study was to calculate magnetic field correction factors of ionization chambers for combinations of various energies of photon beams and various strengths of magnetic fields either perpendicular or parallel to the incident photon beam. Parallel plate ionization chambers are usually used to measure the absorbed dose to water at low-energy beams for radiotherapy. It is advantageous when measuring surface doses to water and doses in a high dose gradient region, and has an easy design customization and fabrication. Three commercial parallel plate ionization chambers were selected to simulate the response variations in a magnetic field. They included the IBA NACP-02, PTW Roos (Type 34001), and Exradin A11. These chambers have the same height but different radii of the sensitive volume. For radiation sources, Eldorado 6 for a Co-60 beam as well as Varian Clinac® series for 6, 10, and 15 MV photon beams were adopted. A spectral source for a 7 MV photon beam adopted in the previous study for the MRI-guided radiotherapy machine was also simulated. The strengths of magnetic field ranged from 0.35 T to 1.5 T, which was either perpendicular or parallel to the photon beam.<br/> The responses of the parallel plate ionization chambers in magnetic fields increased by up to 18% compared to those without a magnetic field. The magnetic field correction factors of the parallel plate ionization chambers were 0.85 to 1.0. Large beam quality dependence and variation occurred in cases of the perpendicular orientation. The larger beam quality was applied, the smaller amount of magnetic field corrections were needed. Except for a Co-60 beam, the largest magnetic field correction occurred around 1.0 T rather than 1.5 T, which was the largest magnetic field strength in this study. In cases of the parallel orientation, only small magnetic field corrections less than 1% were needed. <br/> The MC calculations for segmentation of the sensitive volume of the ionization chamber showed a strong heterogeneity of the absorbed dose due to helical motions of secondary electrons in the sensitive volume. The sensitive volume of the Roos chamber was in silico modified by half or twice the radius or height to figure out the relationship between the sensitive volume and magnetic field correction. The ionization chamber with a small sensitive volume was less influenced by a magnetic field.