Development of a transparent 2D detector (TRADERA: transparent detector for radiotherapy) for on-line control of IMRT beam: Optimization study by Monte Carlo simulations

Introduction The aim of our work is to develop a transparent 2D detector dedicated to the control of the beam at the output of accelerator for IMRT treatments. This would permit firstly to achieve a constant quality control of the beam characteristics (UM stability, homogeneity, MLC leaves’ positions, etc.), and secondly to measure the fluence delivered to the patient. The retained solution is a pixelated parallel plate ionization chamber. Detection is indirect: a part of the photons interact in the upper wall of the chamber (‘‘converter”) creating Compton electrons that will ionize the sensitive volume of the chamber on their way. Several compromise result. On the one hand, the more thicker the converter is, the more important the attenuation of the beam is. But as there are many Compton electrons crossing the sensitive volume, the sensitivity is better. On the other hand, the more thicker the sensitive volume of the chamber is, the greater the sensitivity is. Compton electrons are not emitted in the direction of the incident photon, they can cross the sensitive volume with a large lateral path and degrade the spatial resolution. And this effect increases with the sensitive volume thickness. Materials and methods Monte Carlo simulations (Geant4 9.2.p04) were conducted to evaluate beam attenuation, the sensitivity and spatial resolution of the detector for different converter and sensitive volume heights (converter material: printed circuit board). The sensitivity and spatial resolution are obtained by analyzing the response to a step function (Heaviside): energy deposited in the exposed area and width 20–80%. The phase space Elekta Precise 6 MV from IAEA was used. Results The attenuation is of the order of ∼1% per mm of converter. The sensitivity is proportional to the thickness of the sensitive volume, but not to that of the converter: the proportion of electrons reaching the sensitive volume decreases with increasing thickness of the converter. We lose a factor of 3 on the spatial resolution between sensitive volume thickness of 0.5 mm or 3 mm. Conclusion Compromises expectedwere highlightedandeffectswere quantified. The determination of optimumparameters is dependent on the choice of the electronics, which sets the required sensitivity.