PNNL

Abstract

Polyvinyl toluene (PVT) based gamma-ray scintillation detectors are cost effective for use in radiation portal monitors (RPMs) applied to screening for illicit radioactive materials at international border crossings. While PVT detectors provide good sensitivity in detecting the presence of radioactive materials, they provide poor spectral resolution, limiting their ability to identify the isotopic content of the source of radiation. Thus using only total-spectrum or gross-count alarm algorithms, PVT-based RPMs cannot distinguish innocent materials that contain low-levels of normally occurring radioactivity from special nuclear materials of concern. To reduce the number of “nuisance” alarms produced in PVT-based RPMs by innocent materials, algorithms that analyze spectra from PVT detectors must be optimized to make use of the limited information contained in their energy spectra. This paper discusses how artificial neural networks (ANNs) can be used in such an analysis. The objective was to reduce the nuisance/false alarm probability while maintaining high detection probabilities, thus allowing gross count alarm thresholds to be raised without loss of performance and sensitivity to radioactive materials of interest. The spectra used in this study were obtained from actual PVT-based RPM data, and included cases where simulated spectra were inserted into the measured spectra. This paper also includes an analysis of spectral channel importance and shows evaluations of two methods used to rebin energy spectra into smaller sets. The results show that ANNs can be used with RPMs to reduce nuisance alarms. The algorithms described can be used in analyzing PVT spectra, and potentially sodium iodide spectra.