AB026. Development of real-time gamma-ray spectrum/dose monitor: investigation of true real-time convergence

Background: Our research group is developing a portable gamma-ray monitor that can measure the energy spectrum and dose rate in real time. The detector configuration consists mainly of a CsI(Tl) crystal and a multi-pixel photon counter (MPPC). The ability of the monitor to estimate the energy spectrum and dose really in real-time has been evaluated using a digital pulse processor (DP5, Amptek, Inc., Bedford, MA, USA) instead of a multi-channel analyzer (MCA8000D, Amptek, Inc.) and standard gamma-ray sources, in various conditions. The objective of this study is to experimentally verify the real-time gamma-ray energy spectrum and dose estimation, and compare the results with those obtained using the previous method.

Methods: The method utilized for the estimation is an improved sequential Bayesian estimation method. Specifically, the pulse height distribution of gamma-rays over a time period is measured and converted to energy spectrum. The dose is then calculated through application of the dose conversion coefficient to the estimated energy spectrum. Until now, the measured pulse height spectrum was used as the population, and by resampling each time from the population, the signal was revised, and the energy spectrum was estimated post-experimentally in “real-time”. However, this is not a true real-time estimation during measurements. In the current research, by replacing the multi-channel analyzer component with a digital pulse processor (DP5), true real-time estimation was possible. This is achieved by processing the pulse height and time stamp data continuously with the improved sequential Bayesian estimation method. The monitor is able to display on the PC two graphs of the energy spectrum and dose rate over time. Measurements with standard sources of Barium-133 (¹³³Ba), Caesium-137 (¹³⁷Cs), and Cobalt-60 (⁶⁰Co) were performed at 5 cm from the detector, and the results of the true real-time method were compared with those of the resampling method.

Results: Figures of the energy spectrum over energy were obtained for both methods and showed a similar shape and peaks. Figures of the dose rate over time were also obtained and showed similar values. For ¹³³Ba, ¹³⁷Cs, and ⁶⁰Co, the resampling and true real-time result of the dose rate matched at around 1.1, 3.6, and 0.9 μSv/h, respectively.

Conclusions: The results of the two methods were in good agreement, indicating that it is possible to use the present monitor to estimate the energy spectrum and dose in true real-time.

Keywords: Radiation detection; dose rate; gamma-rays; Bayesian estimation