Characterization of prompt gamma-ray emission with respect to the Bragg peak for proton beam range verification: A Monte Carlo study

Abstract

In this paper we report a Geant4 simulation study to investigate the characteristic prompt gamma (PG) emission in a water phantom for real-time monitoring of the Bragg peak (BP) during proton beam irradiation. The PG production, emission spatial correlation with the BP, and position preference for detection with respect to the BP have been quantified in different PG energy windows as a function of proton pencil-beam energy from 100 to 200MeV. The PG response to small BP shifts was evaluated using a 2cm-thick slab with different human body materials embedded in a water phantom. Our results show that the prominent characteristic PG emissions of 4.44, 5.21 and 6.13MeV exhibit distinctive correlation with the dose deposition curve. The accuracy in BP position identification using these characteristic PG rays is highly consistent as the beam energy increases from 100 to 200MeV. There exists a position preference for PG detection with respect to the BP position, which has a strong dependence on the proton beam energy and PG energies. It was also observed that a submillimeter shift of the BP position can be realized by using PG signals. These results indicate that the characteristic PG signal is sensitive and reliable for BP tracking. Although the maximization of the PG measurement associated with the BP is difficult, it can be optimized with energy and detection position preferences.

Compton-based prompt gamma imaging using ordered origin ensemble algorithm with resolution recovery in proton therapy

“…As can be seen from the results, the PG produced by different kinds of atomic de-excitations (e.g. 14 N, 12 C, 15 O, 16 O) showed different correlations with the proton depth-dose curve, corresponding a previous study 13 . Thus, we also evaluated the reconstruction images of each characteristic, respectively.…”

“…Although the reconstructed 14 N PG had a poor agreement with that obtained by Monte Carlo simulation at the peak position, the falloff at 80% and 50% positions were accurately predicted within 1.4 mm. According to one recent study 13 , the PGs derived from 12 C and 15 O de-excitation (i.e. 4.44 MeV and 5.25 MeV) have the similar distribution to the proton depth-dose curve, while the PGs derived from 16 O de-excitation (i.e.…”

“…12 C, 15 O or 16 O). Since the 14 N PG distribution had a lower agreement with dose deposition of proton beam compared with that of 12 C and 15 O 13 , the reconstruction contained 14 N PG degraded the image quality and the accuracy of the image. But the good agreement of 80% and 50% falloff positions (within 1.4 mm) demonstrates the feasibility of this optimized OE-RR for multi-energy events, as well as the multiple interactions in detection system(e.g.…”

“…13, four main PG energy spectral peaks are: 2.31 MeV ( 14 N de-excitation), 4.44 MeV ( 12 C de-excitation), 5.25 MeV ( 15 O de-excitation) and 6.13 MeV ( 16 O de-excitation) 18 . Since PG derived from 14 N, 12 C, 15 O and 16 O de-excitation exhibits different correlations with proton energy deposited in depth 13 , we evaluated the reconstruction by them respectively. The total energy windows of coincidence events were set as within ±0.2 MeV of the four known PG energy spectral peaks, to alleviate the affect due to the incomplete absorption and background radiation.…”

Characterization of prompt gamma ray emission for in vivo range verification in particle therapy: A simulation study

“…With increasing depth of the BP position as the beam energy increases, the favorable PG detection is likely to be increasingly forward. Further analysis of proton irradiations with respect to the BP position can be found in our previous study [53]. Figure 4.…”