Development of stereotactic radiosurgery using carbon beams (carbon-knife)

Reconstruction of dose distributions for fine carbon-ion beams using iterative approximation toward carbon-knife

For the practical application of carbon-knife with fine carbon-ion beams, the quantification of the dose distribution is essential and requires a high spatial resolution. We propose a novel method to quantify dose distributions with a spatial resolution smaller than the dosimeter size. The proposed method innovates the iterative reconstruction technique. Using a diode dosimeter with a sensitive area of 1 mm², two-dimensional dose-area-product (DAP) distributions were measured at a 0.1 mm step at the surface and near the Bragg peak depths for fine carbon-ion beams of ∼1 mm size at the full width at half maximum (FWHM). Then, the dose distributions were reconstructed with a spatial resolution of 0.1 × 0.1 mm² from the measured DAP distributions. However, an unnaturally high noise was observed in the reconstructed dose distributions, which were considered to originate from the measurement reproducibility errors of the DAP distributions estimated to be 2.5%-3%. Therefore, a low-pass filtering process was implemented to reduce the errors on the reconstructed dose distributions. The optimum cut-off frequencies of the low-pass filter were estimated depending on the amplitude of the induced noise. Using the filtering process with the obtained optimum cut-off frequency, the dose distribution was quantified with an average error of approximately 3% or less with respect to the peak value, when the actual measurement had an error of 3%. In the reconstructed dose rate distributions, a steep penumbra P_80-20 ∼ 0.2 mm was observed at the surface, and a dose rate at the center axis of ∼90 Gy s^-1 and a beam size of ∼1.1 mm at FWHM near the Bragg peak were obtained. The proposed method is expected to be useful for the measurement-based determination of microbeam models for commissioning and dose distribution calculations toward carbon-knife applications.