Implementation of MRI gel dosimetry in radiation therapy

Sammanfattning: Gel dosimetry was used together with magnetic resonance imaging (MRI) to measure three-dimensional absorbed dose distributions in radiation therapy. Two different dosimeters were studied: ferrous- and monomer gel, based on the principles of radiation-induced oxidation and polymerisation, respectively. Single clinical electron and photon beams were evaluated and gel dose distributions were mainly within 2% of conventional detector results. The ferrous-gel was also used for clinical proton beams. A decrease in signal per absorbed dose was found close to the end of the range of the protons (15-20%). This effect was explained as a linear energy transfer dependence, further supported with Monte Carlo simulations. A method for analysing and comparing data from treatment planning system (TPS) and gel measurements was developed. The method enables a new pixel by pixel evaluation, isodose comparison and dose volume histogram verification. Two standard clinical radiation therapy procedures were examined using the developed TPS verification method. The treatment regimes included several beams of different radiation qualities. The TPS calculated data were in very good agreement with the dose distribution measured by the ferrous-gel. However, in a beam abutment region, larger dose difference was found. Beam adjustment errors and a minor TPS underestimation of the lateral scatter contribution outside the primary electron beam may explain the discrepancy. The overall uncertainty in the ferrous-gel dose determination was considerably reduced using an optimised MRI acquisition protocol and a new MRI scanner. The relative dose uncertainty was found to be better than 3.3% for all dose levels (95% confidence level). Using the method developed for comparing measured gel data with calculated treatment plans, the gel dosimetry method was proven to be a useful tool for radiation treatment planning verification.