To address this shortfall, a team headed up by James Renaud at McGill University has designed a portable water calorimetry system for use with non-standard particle beams with reference depths of 6–20 mm. The short-range water calorimeter (SHREWcal) operates without a large water phantom, instead using a small glass vessel filled with pure water as the absorber. As a result, SHREWcal requires collimated radiation fields with diameters no greater than 70 mm and can accommodate electron energies as low as 6 MeV.

Renaud and colleagues (also from Université Catholique de Louvain, University of Toronto, EBG MedAustron, National Physical Laboratory and Clatterbridge Cancer Centre) constructed a SHREWcal prototype and tested its reproducibility with low-energy electron and proton beams (Phys. Med. Biol. 61 6602).

Compact construction

SHREWcal comprises a sealed Pyrex vessel, with 79 mm diameter, 1.1 mm thick front and back circular windows, encased in a block of expanded polystyrene. The space between the windows is 22.7 mm. Two opposing ports in the sidewall allow insertion of thermistor probes, which define the point of measurement. The entire set-up is sealed in an airtight container and cooled to 4 °C.

"A radiation-induced temperature rise in the water will result in a change in thermistor resistance, which is measured using a Wheatstone bridge circuit," Renaud explained. "Dose to water is determined by multiplying this temperature rise by the specific heat capacity of water, along with some necessary correction factors."

The researchers used SHREWcal to measure absorbed dose to water in clinical 6 and 8 MeV electron beams, with field sizes of 10 × 10 cm. The calorimeter entrance window was positioned at a source-to-surface distance of 1000 mm and left to cool overnight. The thermistors were positioned at a depth of 10.8 mm relative to the outer front face of the vessel. They repeatedly delivered 60 s irradiations in sets of three to six.

They then acquired measurements in a 60 MeV proton beam used for ocular treatments at Clatterbridge. Here, the thermistor depth was 7.5 mm. The team measured both monoenergetic and range-modulated proton beams, collimated to 30 mm diameter fields. The calorimeter was repeatedly exposed to 20 s irradiations in sets of two to five.

For an average of 25 repeated measurements, the standard uncertainty on the mean measured dose to water was 0.2% for proton beams and 0.4% for electron beams. In terms of thermal stability, drifts were a couple of hundred µK per minute, with a short-term variation of 5–10 µK.

The researchers also calculated corrections for conductive heat transfer. This is needed to account for other sources of heat loss or gain (due to the presence of non-water materials or dose gradients, for example) that perturb the radiation-induced temperature rise. For 60 s electron irradiations, the correction was 1.042 and 1.049, for 6 and 8 MeV beams, respectively. For 20 s proton irradiations, it was 1.021 for monoenergetic and 1.024 for modulated beams.

They note that by simulating heat transfer a priori, the thermistor positioning within the vessel can be optimized to minimize this uncertainty, which is strongly dependent upon dose gradient. For the proton beams used in this study, the flat dose distributions within the vessel resulted in an optimal thermistor position of about 11 mm, while for electrons with more severe dose gradients, the optimal position was about 6 mm.

Overall uncertainty

To determine the overall dose measurement uncertainty, the researchers combined beam-independent uncertainties (specific heat capacity of water, absolute temperature measurement, and bridge and thermistor calibration factors) and beam-dependent uncertainties (thermistor positioning and reproducibility of the SHREWcal reading).

The combined uncertainty was 0.6% for electron beams and the monoenergetic proton beam, and 0.7% for the modulated proton beam. The largest contributions to uncertainty were thermistor positioning (up to 0.5%) and the conductive heat transfer correction (up to 0.4%). For comparison, absorbed dose to water can be established with an uncertainty of 0.2–0.3% in high-energy photon beams using primary standard water calorimeters.

Renaud and colleagues concluded that the SHREWcal water calorimeter could form the basis of a transportable dose standard for direct calibration of ionization chambers in a user's beam. They are currently completing a comparison study in which SHREWcal and a graphite calorimeter are operated side-by-side to measure the dose from the 60 MeV proton beam at the Clatterbridge Cancer Centre. "We have also developed a probe-format graphite calorimeter, designed specifically for direct use by physicists in the radiotherapy clinic," Renaud told medicalphysicsweb.

Related stories

• GEM detector lines up for proton therapy
• OSL detectors accurately measure ion dose
• A step towards 3D proton beam dosimetry
• Pulsed protons no problem for calorimetry