Session Item

Phantoms are a vital tool for both reference and end-to-end audits in radiotherapy. Phantom-based dosimetric audits have been shown to improve confidence and consistency of radiotherapy treatments at clinical centres. Recently, there has been an increase in the number of proton therapy centres within the UK and worldwide. However, due to differences in radiation interaction with matter between photon and proton beams, previously used X-ray phantoms and phantom materials are suboptimal for proton therapy. Research shows current tissue-equivalent materials cause large uncertainties in proton therapy measurements. Consequently, there is a need for photon and proton optimised tissue-equivalent materials. 

We have developed a mathematical model using MATLAB which enables the formulation of tissue-equivalent materials considering physical properties, photon interactions but also proton stopping power, absorption, and scattering interactions. Using a non-linear optimisation algorithm, this work has formulated new bone- and muscle-equivalent epoxy resin-based materials in terms of their photon and proton interaction parameters. However, perfectly matching all interactions is challenging, therefore, a weighted cost function was defined according to the relative importance of each photon and proton interaction. Mass density, mass attenuation and relative stopping power (RSP) were assigned the highest weightings due to their impact on the materials ability to be correctly characterised during the imaging and treatment planning process (TPS) as well as providing accurate proton dosimetry measurements.

The formulated materials have been manufactured and characterised via the use of single-and dual-energy CT as well as proton water-equivalent thickness measurements at The National Physical Laboratory (UK) and University College London Hospital. Monte Carlo simulations (FLUKA) were completed to calculate the fluence correction factors of the materials, which provides a more detailed understanding of nuclear interaction equivalence of the material. Each material’s RSP was also compared to the clinical TPS assigned RSP using the HU-RSP calibration curve to check their use for end-to-end audit purposes.

Table 1 show that the new optimised materials score better than current commercial phantom materials when considering all-important physical properties, photon interactions, proton stopping power, absorption, and scattering interactions.

Experimental results suggest the new bone and muscle formulations adequately mimic target tissue parameters such as the mass density and RSP with acceptable differences of 1-3%. 

Results suggest that the proposed bone and muscle formulations can be used for the development of future photon and proton-optimised dosimetric phantoms. This work highlights the potential of this new mathematical model as a tool in the creation of phantoms with optimised tissue materials for photon and proton beams.