Speaker
Description
Radiotherapy treatment verification requires the use of robust dosimeters. Radiochromic film response to radiation [1] makes it a good candidate as a bidimensional dosimeter together with the use of a flatbed scanner. Film digital image is processed into the three colour (RGB) channel independently. Dose to water calibration curve for each channel is made using the mean response of different irradiated areas. The dose of an individual irradiated pixel will be affected by the deviation of scanned level of darkness from the mean behaviour. Two different sources of deviation are established in the literature [2]: variation in film thickness/composition due to manufacturing process and scanner irregular illumination during the digitalization. The three independent dose values affected by pixel inhomogeneity can be combined in order to estimate a unique corrected dose value. This is the basic idea of the so-called multichannel dosimetry methods [3].
In this work, we present a novel description for film thickness variation based on Beer-Lambert law of optical absorption. The level of darkness is expressed in terms of net optical density (netOD) and the scanner noise effect is considered independent for each channel. This model is introduced within the framework of Bayesian estimation. The prior distribution of film inhomogeneity parameter is obtained from calibration curve. First studies, perfomed in EBT3 Gafchromic model, show that film inhomogeneity does not behave as a normal distribution random variable as is considered in previous multichannel methods [4]. We analyse and parameterize the prior distribution obtained for different film lots irradiated with standard clinical 6 MV X-ray fields at Hospital Universitario Virgen Macarena (HUVM). Then the complete Bayesian multichannel method is applied to treatment verification films and results are compared with the ones obtained with the multichannel method developed in the radiophysics unity of HUVM.
The new multichannel method is expected to be helpful in low energy proton irradiation. In this case, film variation can introduce important changes on the absorbed doses if Bragg peak region coincides with active volume of film [5].
[1] S. Devic et al. Reference radiochromic film dosimetry: review of technical aspects. Physica Medica, 32, 4, 541-556 (2016).
[2] Rulon R. Mayer et al. Enhanced dosimetry procedures and assessment for EBT2 radiochromic film. Med Phys, 39, 4, 2147-55 (2012).
[3] A. Micke et al. Multichannel dosimetry with nonuniformity correction. Medical Physics 38, 2523-2534 (2011).
[4] I. Méndez. On multichannel film dosimetry with channel-independent perturbations. Med Phys, 41, 011705 (2014).
[5] M.C. Battaglia et al. EBT3 film calibration in the Bragg peak region for proton beams below 5 MeV. Nuclear Instruments and Methods in Physics Research B 444, 117-124 (2019).
