Investigating neutron activated contrast agent imaging for tumor localization in proton therapy: a feasibility study for proton neutron gamma-x detection (PNGXD).

Proton neutron gamma-x detection (PNGXD) is a novel imaging concept being investigated for tumor localization during proton therapy that uses secondary neutron interactions with a gadolinium contrast agent (GDCA) to produce characteristic photons within the 40-200 keV energy region. The purpose of this study is to experimentally investigate the feasibility of implementing this procedure by performing experimental measurements on a passive double scattering proton treatment unit. Five experimental measurements were performed with varying concentrations and irradiation conditions. Photon spectra were measured with a 25 mm(2), 1 mm thick uncollimated X-123 CdTe spectrometer. For a 10.4 Gy administration on a 100 ml volume phantom with 10 mg g(-1) Gd solution placed in a water phantom, 1129 +/- 184 K-shell Gd counts were detected. For an administered dose of 21 Gy and the same Gd solution measured in air, resulted in 3296 +/- 256 counts. A total of 1094 +/- 171, 421 +/- 150 and 23 +/- 141 K-shell Gd counts were measured for Gd concentrations of 10 mg g(-1), 1 mg g(-1) and 0 mg g(-1) for 7 Gy dose in air. The signal to noise ratio for these five measurements were: 7, 15, 6, 3, and 0.2, respectively. The spectrum contained 43 keV K-alpha and 49 keV K-beta peaks, however a small amount of 79.5 and 181.9 keV prompt gamma rays were detected from gadolinium neutron capture. This discrepancy is due to a drop in the intrinsic detection efficiency of the CdTe spectrometer over this energy range. The measurements were compared with Monte-Carlo simulation to determine the contributions of Gd neutron capture from internal and external neutrons on a passive scattering proton therapy unit and to investigate the discrepancy in detected characteristic x-rays versus prompt gamma rays.