A novel methodology for gamma-ray spectra dataset procurement over varying standoff distances and source activities

Journal Article · Fri Aug 02 00:00:00 EDT 2024 · Nuclear Instruments and Methods in Physics Research. Section A, Accelerators, Spectrometers, Detectors and Associated Equipment

DOI:https://doi.org/10.1016/j.nima.2024.169681· OSTI ID:2585042

Fjeldsted, Aaron P. ^[1]; Morrow, Tyler J. ^[2]; Scott, Clayton ^[3]; Zhu, Yilun ^[3]; Holland, Darren E. ^[4]; Hanks, Ephraim M. ^[1]; Lintereur, Azaree T. ^[5]; Wolfe, Douglas E. ^[1]

---

1. Pennsylvania State Univ., University Park, PA (United States)
2. Sandia National Lab. (SNL-NM), Albuquerque, NM (United States)
3. Univ. of Michigan, Ann Arbor, MI (United States)
4. Air Force Institute of Technology, Dayton, OH (United States)
5. Los Alamos National Laboratory (LANL), Los Alamos, NM (United States)

The adoption of machine learning approaches for gamma-ray spectroscopy has received considerable attention in the literature. Many studies have investigated the deployment of various algorithm architectures to a specific task. However, little attention has been afforded to the development of the datasets leveraged to train the models. Such training datasets typically span a set of environmental or detector parameters to encompass a problem space of interest to a user. Variations in these measurement parameters will also induce fluctuations in the detector response, including expected pile-up and ground scatter effects. Fundamental to this work is the understanding that 1) the underlying spectral shape varies as the measurement parameters change and 2) the statistical uncertainties associated with two spectra impact their level of similarity. While previous studies attribute some arbitrary discretization to the measurement parameters for the generation of their synthetic training data, this work introduces a principled methodology for efficient spectral-based discretization of a problem space. A signal-to-noise ratio (SNR) respective spectral comparison measure and a Gaussian Process Regression (GPR) model are used to predict the spectral similarity across a range of measurement parameters. This innovative approach effectively showcased its capability by dividing a problem space, ranging from 5 cm to 100 cm standoff distances and 5 μCi–100 μCi of ¹³⁷Cs, into three unique combinations of measurement parameters. The findings from this work will aid in creating more robust datasets, which incorporate many possible measurement scenarios, reduce the number of required experimental test set measurements, and possibly enable experimental training data collection for gamma-ray spectroscopy.

• Research Organization: Sandia National Laboratories (SNL-NM), Albuquerque, NM (United States)
• Sponsoring Organization: Defense Threat Reduction Agency; USDOE National Nuclear Security Administration (NNSA)
• Grant/Contract Number: NA0003525
• OSTI ID: 2585042
• Report Number(s): SAND2025--07241J; 1740821
• Journal Information: Nuclear Instruments and Methods in Physics Research. Section A, Accelerators, Spectrometers, Detectors and Associated Equipment, Journal Name: Nuclear Instruments and Methods in Physics Research. Section A, Accelerators, Spectrometers, Detectors and Associated Equipment Journal Issue: 1 Vol. 1067; ISSN 0168-9002
• Publisher: Elsevier
• Country of Publication: United States
• Language: English

Related Subjects

data science
gamma-ray spectra
machine learning
pile-up