Characterization of the energy response and backscatter contribution for two electronic personal dosimeter models

We characterized the energy response of personal dose equivalent (Hp(10) in mrem) and the contribution of backscatter to the readings of two electronic personal dosimeter (EPD) models with radionuclides commonly used in a nuclear medicine clinic. The EPD models characterized were the RADOS RAD-60R, and the SAIC PD-10i. The experimental setup and calculation of EPD energy response was based on ANSI/HPS N13.11-2009. Fifteen RAD-60R and 2 PD-10i units were irradiated using (99m)Tc, (131)I, and (18)F radionuclides with emission energies at 140 keV, 364 keV, and 511 keV, respectively. At each energy, the EPDs output in Hp(10) [mrem] were recorded with 15 inch thick PMMA to simulate backscatter form the torso. Simultaneous free-in-air exposure rate measurements were also performed using two Victoreen ionization survey meters to calculate the expected EPD Hp(10) values per ANSI/HPS N13.11-2009. The energy response was calculated by taking the ratio of the EPD Hp(10) readings with the expected Hp(10) readings and a two-tailed z-test was used to determine the significance of the ratio deviating away from unity. The contribution from backscatter was calculated by taking the ratio of the EPD Hp(10) readings with and without backscatter material. A paired, two-tailed t-test was used to determine the significance of change in EPD Hp(10) readings. The RAD-60R mean energy response at 140 keV was 0.85, and agreed to within 5% and 11% at 364 and 511 keV, respectively. The PD-10i mean energy response at 140 keV was 1.20, and agreed to within 5% at 364 and 511 keV, respectively. On average, in the presence of acrylic, RAD-60R values increased by 32%, 12%, and 14%, at 140, 364, and 511 keV, respectively; all increases were statistically significant. The PD-10i increased by 25%, 19%, and 10% at 140 keV, 364 keV, and 511 keV, respectively; however, only the 140 keV measurement was statistically significant. Although both EPD models performed within the manufacturers' specifications of ± 25% in the energy ranges used, they fell outside of our criteria of 10% at lower energies, suggesting the need to calculate energy-dependent correction factors, depending on the intended EPD use.