Investigation of the Dosimetry Characteristics of the GAFCHROMIC® EBT3 Film Response to Alpha Particle Irradiation

The purpose of this study was to investigate the dosimetric characteristics of the GAFCHROMIC® EBT3 film responding to alpha particle irradiation. Unlaminated GAFCHROMIC® EBT3 film pieces, were irradiated with a 30.055 kBq 241Am alpha source, at eight different dose levels, between 0 and 509 Gy. The irradiations were performed inside an enclosed box. Epson Expression 10000 XL scanner in transmission mode was used to digitize irradiated films 24 hours post-irradiations as 16-bit RGB images in tagged image file format (TIFF). Optical density (OD) values were obtained by following the OD theorem. Raw and normalized pixel values (PVnorm) from the red, green, and blue colour channels were sampled from a 3 × 3 mm2 region of interest. Calibration curves were created for both data sets (OD and PVnorm) and were fit accordingly. Monte Carlo simulations with the Geant4 toolkit were performed to establish a dose rate at the point within the sensitive layer of the film. An alpha dosimetry protocol for EBT3 films was obtained from the Monte Carlo calculated dose rate and dose calibration curves for alpha radiation were created. It is necessary to extend this study for different film types and compare to photon dosimetry calibration curves.


Introduction
GAFCHROMIC ® EBT3 films are designed and used for the measurement of absorbed doses of ionizing radiation especially suited for high-energy photons.This film model has optimal dose ranges from 0.2 Gy to 10 Gy and can be developed in real time without post-exposure treatment [1].EBT3 films need a color photo scanner to acquire 16-bit per channel of the red, green, and blue color components of light transmitted through the film.The EBT3 film model offers high spatial resolution and allows for depth dose measurements [2].The EBT3 film has a structure comprised of an active layer (28 µm thick), sandwiched between two matte-polyester substrates (125 µm).The active layer contains an active component, a marker dye, stabilizers, and other elements [1].The change in optical properties of GAFCHROMIC ® films depends on the absorbed dose and the linear energy transfer (LET) of the ionizing radiation.The current experience of GAFCHROMIC ® film dosimetry is based on low LET photon beams.Lately, the use of alpha emitting radionuclides such as the Alpha Tau Diffusing Alpha-emitters Radiation Therapy (DaRT) [3] for the treatment of cancer, has increased.This novel treatment modality relies on temporary or permanently implantable seeds impregnated with a small activity of 224 Ra, which are placed inside the tumours.Short-lived alpha-particle emitting atoms are released in the decay chain of 224 Ra and diffuse inside the tumour [3].In this study, we investigate the GAFCHROMIC ® EBT3 film model response to eight different alpha particle doses from an 241 Am source.However, since the range of alpha particles is very short (up to 100 micrometers), unlaminated films were used to directly expose the active layer to the alpha particles [4].

Materials and Methods
In this study, unlaminated GAFCHROMIC ® EBT3 films were used for alpha particle dosimetry.The film was cut into 6.35 x 5.08 cm 2 pieces using a sharp cutter with a board for paper trimming.The films were cut longitudinally, and a landscape orientation was followed throughout the entirety of the study.The radioactive source emitting alpha particles, 241 Am was placed inside a metal box, mimicking a dark box environment.The source has a gold cover of approximately 51.8 nm to protect anything from being in direct contact with the source.The films were placed on the 241 Am source downwards, with the active layer in contact with the gold cover for varying exposure times to absorbed doses in the range of 0 Gy to 509 Gy. Figure 1 presents the experimental design.The absorbed doses were determined by calculating the dose rate of the experimental design using Monte Carlo Simulations performed with Geant4 simulation toolkit.The dose rate was calculated to be 28.26Gy/h for irradiation at the level of the active layer of the film [5].The film was scanned using Epson Expression 10000 XL scanner.The image was saved as a Tagged Image File Format (tiff).All scanned images were analysed using an inhouse python script in terms of pixel value (PV) for the red, green, and blue channels.The PV of the exposed region of the films were converted into average net optical densities (ODnet).The Equation 1 below indicates the conversion of 16-bit pixel channel numbers converted into ODnet.The film irradiations were repeated to lower error probabilities.It represents the EBT3 film at red, green, and blue colour channels.The calibration curves, i.e., the dependence delivered dose on the normalized PVs (Equation 2) and ODnet, were fitted to a power function (Equation 3) and exponential function (Equation 4) respectively.
Where normalized PV is calculated below: Where b, c, and n are fitting parameters, as a polynomial correction to a linear form.Finally, the fit for the ODnet data as mentioned, was fit to an exponential function: where , , and  are fitting parameters.The plotted uncertainties were calculated based on total uncertainties from the relative fit parameter uncertainties and experimental uncertainties added in quadrature.

Results and Discussion
The dependence of the normalized PV and ODnet on the absorbed dose have been investigated.The result from the normalized PV concludes that the red colour channel has the highest sensitivity than the green and blue channels, however only useful for small doses (up to ~8 Gy).The higher sensitivity for the red channel is due to this channel having the highest absorption for low dose ranges up to approximately 8 Gy, however, the higher dose ranges up to 100 Gy, will saturate the red channel sensitivity curve.The green channel can provide a smaller uncertainty than the red channel when normalized PV is used as a response function for higher doses.The absorbed dose (Gy) as a function of normalized PV for all three colour channels were fit to the polynomial fit function, seen as equation 3 above.The absorbed dose (Gy) as a function of ODnet for all three colour channels, were fit to an exponential function.The errors bars associated with each plot are the calculated total uncertainties from the relative fit parameter uncertainties and experimental uncertainties, added in quadrature.Through further analysis, caution must be taken when looking at particular colour channels where the raw PV is less than 10 000.In this case, the capabilities of a flatbed optical scanner are interpreted rather than the function of the film.The measured response value is dependent on the light transmission through the film but for response values less than 10 000, it is corrupted by a signal from scattered light in the measurement system.This signal is now independent of the transmission of the film.Therefore, any dose above 50 Gy cannot represent response values appropriately for EBT3 film.

Conclusion
EBT3 film can measure the alpha radiation doses with high accuracy up to 50 Gy.This film model cannot be used for dosimetry higher than 50 Gy, therefore, above this dose limit, another film model such as HD-V2 film (high-dose) must be used.Further analysis is currently in completion with HD-V2 film to be used in combination with EBT3.In future studies, unlaminated EBT3 film will be used to benchmark autoradiography measurements for novel animal studies involving alpha emitting sources used for the treatment of cancer.
)And the normalized data was fit to: 3