Absolute prompt-gamma yield measurements for ion beam therapy monitoring

Abstract

Prompt-gamma emission detection is a promising technique for hadrontherapy monitoring purposes. In this regard, obtaining prompt-gamma yields that can be used to develop monitoring systems based on this principle is of utmost importance since any camera design must cope with the available signal. Herein, a comprehensive study of the data from ten single-slit experiments is presented, five consisting in the irradiation of either PMMA or water targets with lower and higher energy carbon ions, and another five experiments using PMMA targets and proton beams. Analysis techniques such as background subtraction methods, geometrical normalization, and systematic uncertainty estimation were applied to the data in order to obtain absolute prompt-gamma yields in units of prompt-gamma counts per incident ion, unit of field of view, and unit of solid angle. At the entrance of a PMMA target, where the contribution of secondary nuclear reactions is negligible, prompt-gamma counts per incident ion, per millimetre and per steradian equal to (124 ± 0.7stat ± 30sys) × 10(-6) for 95 MeV u(-1) carbon ions, (79 ± 2stat ± 23sys) × 10(-6) for 310 MeV u(-1) carbon ions, and (16 ± 0.07stat ± 1sys) × 10(-6) for 160 MeV protons were found for prompt gammas with energies higher than 1 MeV. This shows a factor 5 between the yields of two different ions species with the same range in water (160 MeV protons and 310 MeV u(-1) carbon ions). The target composition was also found to influence the prompt-gamma yield since, for 300/310 MeV u(-1) carbon ions, a 42% greater yield ((112 ± 1stat ± 22sys) × 10(-6) counts ion(-1) mm(-1) sr(-1)) was obtained with a water target compared to a PMMA one.

Prompt-Gamma Monitoring of Proton- and Carbon-Therapy. Combined Development of Time-of-Flight Collimated- and Compton-Cameras

“…Consistent analysis, using the same residual background subtraction method, has led to typical PG yields above 1 MeV of ≈3 × 10 −3 cm −1 per proton and ≈2×10 −2 cm −1 per carbon ion in a PMMA target [12]. These yields are those directly induced by the projectiles (i.e.…”

Benchmarking Geant4 hadronic models for prompt‐γ monitoring in carbon ion therapy

“…25 The GEANT4 simulation of the two experiments confirmed this result, predicting a factor 4.6 AE 0.2 between the photon emission yield at 80 and 220 MeV/u. These values can also be compared with those presented in Pinto et al, 50 which gave differential yields in carbon ions per solid angle and mm for a 310 MeV/u 12 C beam that can be used to extrapolate an integrated yield over the whole range. Using this method, we obtained an integrated yield of 0.99 9 10 À3 sr À1 /N C , which was compatible with the result from Table III. The overestimation of the yields by a factor 2.2 for QMD and 2.4 for BIC at 60°was also in agreement with the results of Dedes et al, 26 where the overestimation by a factor ranging from 1.8 to 2.8 was observed when using GEANT4 simulations for 12 C ions between 95 and 310 MeV/u.…”

Application of variance‐based uncertainty and sensitivity analysis to biological modeling in carbon ion treatment plans

“…Currently, a lot of effort is invested into the quantification and subsequent minimization of dose-related uncertainties for proton and ion therapy. [2][3][4] This includes both the quantification of these physical uncertainties [5][6][7][8][9] and how to account for them in the optimization process. [10][11][12][13][14][15][16] Little is known about the actual size and influence of uncertainties in RBE predictions.…”