Background and Aims

Commercially available real time response dosimetric systems were shown to be unsuitable for Flash therapy, in which ultra-high Dose Per Pulse (DPP) is of interest. There is a need for novel, reliable and fast response detectors for such an application. In the present work, a comprehensive investigation is reported on the properties of diamond-based detectors specifically designed for Flash therapy.

Methods

Several diamond Schottky diode prototypes were produced at Rome Tor Vergata University, in cooperation with PTW Freiburg. Quite a few relevant design and electronical parameters were systematically varied, tuning the device’s overall performance and meet the stringent requirements of ultra-high DPP irradiation. The resulting prototypes were tested by using an ElectronFllsh linac (SIT S.P.A.), by using a 9 MeV electron beam and DPPs up to 12.5 Gy/pulse.

Results

The performance of the prototypes, investigated under a wide range of irradiation parameters, was found to depend on their specific structural properties and designs, showing a response linearity up to a DPP of 12.5 Gy/pulse in the best case.

Conclusions

First results of this systematic investigation, clearly demonstrate the feasibility of a diamond based detector for Flash therapy applications. The analysis of the obtained experimental results will be used as an input for further development and optimization procedure of the investigated prototypes.

The present work is part of the 18HLT04 UHDpulse project which has received funding from the EMPIR programme co-financed by the Participating States and from the European Union’s Horizon 2020 research and innovation programme.

Background and Aims

Fast and accurate active dosimeters are the key component to perform quantitative measurements in FLASH radiotherapy. No reference active dosimeters are currently available since most detectors show non recoverable saturation effects for dose-per-pulse (DPP) values typical of FLASH (1Gy/p or higher). Aim of this study is to develop and test a detector prototype based on inorganic scintillators for FLASH-IORT active dosimetry.

Methods

The detector prototype is composed of a LYSO scintillating crystal (2x2x10 mm^3) wrapped in 5 Teflon layers and coupled to an optical fiber 1.2 m long (0.980 mm diameter, PMMA core). The other end of the fiber is connected to a Photodiode (PD, Thorlabs - SM05PD7A) read out by a multimeter (Keithley 617) that integrates the PD photocurrent. The detector was placed in a PMMA support and covered with solid water slabs. The phantom was irradiated with a 7 MeV IORT electron LINAC (NOVAC7 from SIT, Aprilia, Italy) at different SSDs and depths in solid water to vary DPP at the detector position. The DPP spanned from conventional (3 cGy/pulse) to FLASH values (250 cGy/pulse). The DPP was evaluated by means of calibrated Gafchromic films.

Results

The PD integrated charge was linear in the whole range of DPP values explored and 0.46 nC/Gy in sensitivity was measured.

Conclusions

Further measurements are planned to fully characterize the detector such as extend the DPP upper limit and investigate the dependency on the instantaneous DPP, but these first results indicate LYSO based detectors as promising candidates for FLASH-IORT active dosimetry.

Background and Aims

One of the challenges in investigating FLASH radiation therapy is the reliable measurement of the absorbed dose at ultra-high dose per pulse (DPP). This work aims to evaluate a probe-type graphite calorimeter (Aerrow) at ultra-high DPP.

Methods

The DPP was varied between 0.5 to 5.5 Gy. The signal was compared to the pulse charge measured by a non-destructive Integrating Current Transformer (ICT). In addition, a depth dose curve was measured and compared to an ionization chamber measurement and a Monte-Carlo simulation.

Results

The response of the calorimeter was proportional to the ICT signal. On average, the standard deviation of the calorimeter response was 0.2 % in high dose rate and reached 1% at the lower rate. In figure 1, the depth dose measurement with calorimeter and ionization chamber is compared to Monte-Carlo simulation. The half-value depths in water R50 obtained from measurement and simulation agree within 1 mm.

Figure 1: Depth dose in water measured and simulated.

Conclusions

Calorimetry is showing promising results for absorbed dose measurements. Calorimetry gets simpler at FLASH dose rates as the dose delivery is in a few seconds or less. The preliminary results show that advanced thermal insulation of the calorimeter is not required, nor the use of a heat lost correction factor.

Acknowledgement: This project 18HLT04 UHDpulse has received funding from the EMPIR programme co-financed by the Participating States and from the European Union’s Horizon 2020 research and innovation programme.

Background and Aims

Ionization chamber models developed during the last century by Boag have been and continue to be of great interest for daily
clinical practice. However, its simplicity is achieved at the cost of ignoring higher order effects that do not play
an important role in the dose rates found in conventional radiotherapy. However, when we get into the ultra
high dose rate regime, these models have shown important discrepancies.

Methods

A computational model was developed whose main objective is to solve the differential equations governing
charge transport along the ionization chambers. Charge collection efficiency and instantaneous current of different ionization chambers obtained from the model are currently evaluated with a set of measurements carried
out in collaboration with the PTB.

Results

Figure 1: Charge collection efficiency (CCE) for different analytical models currently used for saturation correction of ionization chambers and their simulation counterpart for a PPC40 chamber. CCE was evaluated for a plane-parallel ionization chamber with 2 mm gap at 300 V and a pulse duration of 2.5 μs at reference temperature and pressure conditions.

Figure 2: Comparison between instantaneous current from simulation and measured at 4 Gy/pulse and 400 V using a ROOS chamber.

Conclusions

Our computational model, which is still under development, shows better agreement to experimental data than existing analytical models and a preliminary version of this software has been released for testing.

Acknowledgement
This project 18HLT04 UHDpulse has received funding from the EMPIR programme co-financed by the Participating States and from the European Union’s Horizon 2020 research and innovation programme.

Background and Aims

Ionization chambers (IC) represent the standard for performing the commissioning of medical linacs. Nevertheless, their use in the UHDR range is not currently possible, due to the amount of charge produced by each pulse.

For dose-per-pulse (dpp) above 0.5 cGy/p, the approach implemented by international protocols for modelling ion recombination failed, because the free electron fraction p contribution must be considered. We modify the approach of Di Martino (2005) in order to obtain p by means of ionometric measurements only.

Methods

According to the proposed model:

where

q^col is the charge collected by IC;

V is the voltage applied to IC;

q^gen is the charge generated by the pulse;

A and λ are parameters depending on the IC..

By varying the voltage applied V, such function can be determined versus the unknown parameters (q^gen, A and λ ); then, such parameters can be determined by means of the non-linear regression method.

Once all parameters are known, p is calculated and and then k_sat is determined as:

being α = A / V .

Results

The method has been adopted for estimating k_sat for the Adv. Markus, both with the beam produced by ElectronFlash and by LIAC HWL.The fit provided an agreement better than 1% within IORT range and better than 5% with 0.6 Gy/p.

Conclusions

Once k_sat is known, ionometric measurements at the larger distance might become the central element of a reliable Quality Assurance program for any Flash linac.

Background and Aims

Optical calorimetry (OC) is a technique based on laser interferometry that spatially reconstructs energy absorbed in a transparent medium. The proposed approach measures absorbed dose to water without requiring correction factors, making it a promising technique for ultra-high-dose-rate beams. To investigate OC as a FLASH dosimeter, a standard linear accelerator was converted to deliver dose rates > 40 Gy/s.

Methods

The electron gun parameters of an Elekta Synergy linac were optimised to increase dose output of a 6 MV photon beam. Further modifications included removing the target, flattening filter, and scattering foils. An in-house built optical calorimeter built around a water phantom was placed at central axis. The image-based dosimeter was operated at a fixed frame rate. An in-house constructed pulse counting device measured the number of radiation pulses delivered per frame. The dose per frame was determined and used to calculate the dose rate achieved at the point of measurement.

Results

In the converted linac beam the OC system measured a dose/frame of 4.2 ± 0.4 Gy/frame. The dose rate was determined to be 210 ± 20 Gy/s at the point of measurement, which is within the expected range reported in the literature. A 10% measurement variation was observed between trials.

Conclusions

A recently decommissioned linac was successfully converted to deliver ultra-high FLASH dose rates. The optical calorimeter determined absolute dose in a FLASH beam without the need for correction factors. Further improvements to reduce the measurement variability would enhance the usefulness of OC as a dosimetry system for FLASH radiotherapy.