The dose response functions of ionization chambers in photon dosimetry – Gaussian or non-Gaussian?

Depth dependence of the single chamber response function of the I’mRT MatriXX array in a 6 MV photon beam

“…Its basis is that beam broadening in the penumbra region is encountered in a 2D array I'mRT MatriXX especially for small field sizes. This effect results from the detector size and the secondary electron transport [9,10,27,32]. This effect is mathematically modeled by the convolution of the true dose profile with a convolution kernel representing the dose response function of the single detector:…”

“…The single detector's dose response function of the I'mRT MatriXX array can be defined as the convolution kernel K(x) that transforms the true dose profile D(x) into the measured signal profile M(x) and thereby characterizes the broadening of the slit-beam dose profile [16,17,31,32]. It is affected by the detector size, the replacement of water by air, the wall and the central electrode of the detector as well as the geometrical form of the chamber itself [16,32].…”

“…It is affected by the detector size, the replacement of water by air, the wall and the central electrode of the detector as well as the geometrical form of the chamber itself [16,32]. As the ranges of high-energy secondary electrons traversing each chamber are relatively large, the measured chamber signal is not only due to secondary electrons entering the chamber from the front side, but also from the ridges between the single chambers [10,32]. Therefore, each ion chamber has a dose response function which is also affected by the transport of secondary electrons.…”

“…The EBT2 film was also placed at these depths, and the high spatial resolution of the film was utilized to determine the true dose profiles D(x). In the first method, the "best fit method" used by Looe et al [32], K(x) was determined by a systematic search for the kernel that would transform D(x) into M(x), as described in section 2.2.3. The second method was to obtain K(x) by deconvolution of M(x) with D(x) as earlier performed by Poppe et al [10], Herzen et al [6] and Looe et al [32]; this is described in section 2.2.4.…”

“…The limited spatial resolution of ionization chamber arrays has been attributed to the finite size of the single detectors as well as to the transport of secondary electrons from the walls into the detector volume [8,9,31,32]. This disadvantage can be overcome by the method proposed by Poppe et al [10], who convolved the calculated dose profile provided by the TPS with the dose response function of the single ionization chamber and used the convolution product for the gammaindex based comparison with the measured 2D array profile.…”