Experimental Study of Discrete Prompt Gamma Lines for In-Vivo Proton Range Verification
J Verburg*, K Riley, J Seco, Massachusetts General Hospital and Harvard Medical School, Boston, MATH-C-144-2 Thursday 10:30AM - 12:30PM Room: 144
Purpose: A novel method for in vivo proton beam range verification is presented, using energy-resolvable proton-induced prompt gamma lines.
Methods: Prompt gamma-rays emitted near the end-of-range of a clinical proton beam originate from transitions between the lower states of excited nuclear reaction products. We developed a prompt gamma-ray detection system with the aim of resolving the resulting discrete gamma lines. The ability to quantify the strength of these lines can provide valuable information for in vivo range verification: known cross sections may be used as prior knowledge, and uncertainties resulting from differences in tissue composition could be reduced. A LaBr3:Ce scintillation detector was used, which provides high energy resolution. The pulses from the photomultiplier tube were acquired with analog-to-digital converters at high sampling rates, synchronized to the cyclotron radiofrequency to separate the quasi-continuous neutron-induced background from the prompt gamma signals of interest.
Results: Preliminary measurements were performed with a 150 MeV proton pencil-beam targeted on the center of a water phantom with a 15 cm diameter. A 1 cm lead slit collimator was used perpendicular to the beam direction, centered at the end-of-range. The acquired gamma spectra show the 6.13 MeV gamma line from oxygen-16, and its escape peak, can be clearly resolved with an energy resolution of ~1% FWHM.
Conclusion: Our results show the feasibility of measuring discrete prompt gamma lines using a practical detection system. These gamma lines can be attributed to specific nuclear reactions with established properties, potentially enabling accurate proton range verification using a small number of measurement points. Experiments with different collimators and proton energies are currently being performed to determine an optimal configuration for range verification.