A comparison of machine learning methods for automated gamma-ray spectroscopy

Mark Kamuda; Jifu Zhao; Kathryn Huff

doi:10.1016/j.nima.2018.10.063

A comparison of machine learning methods for automated gamma-ray spectroscopy

Mark Kamuda
, Jifu Zhao
, Kathryn Huff

National Center for Supercomputing Applications (NCSA)

Research output: Contribution to journal › Review article › peer-review

Abstract

Pattern recognition algorithms such as artificial neural networks (NNs) and convolution neural networks (CNNs) are prime candidates to perform automated gamma-ray spectroscopy. The way these models train and operate mimic how trained spectroscopists identify spectra. These models have shown promise in identifying gamma-ray spectra with large calibration drift and unknown background radiation fields. In this work, two algorithms for mixtures of radioisotopes based on NN and CNN are presented and evaluated.

Original language	English (US)
Article number	161385
Journal	Nuclear Instruments and Methods in Physics Research, Section A: Accelerators, Spectrometers, Detectors and Associated Equipment
Volume	954
DOIs	https://doi.org/10.1016/j.nima.2018.10.063
State	Published - Feb 21 2020

Keywords

Automated isotope identification
Gamma-ray spectroscopy
Neural networks

ASJC Scopus subject areas

Nuclear and High Energy Physics
Instrumentation

Online availability

10.1016/j.nima.2018.10.063

Library availability

Discover UIUC Full Text

Cite this

@article{ca87c14929794196a2d35c21a9a9eaaa,

title = "A comparison of machine learning methods for automated gamma-ray spectroscopy",

abstract = "Pattern recognition algorithms such as artificial neural networks (NNs) and convolution neural networks (CNNs) are prime candidates to perform automated gamma-ray spectroscopy. The way these models train and operate mimic how trained spectroscopists identify spectra. These models have shown promise in identifying gamma-ray spectra with large calibration drift and unknown background radiation fields. In this work, two algorithms for mixtures of radioisotopes based on NN and CNN are presented and evaluated.",

keywords = "Automated isotope identification, Gamma-ray spectroscopy, Neural networks",

author = "Mark Kamuda and Jifu Zhao and Kathryn Huff",

note = "The authors would like to thank the National Nuclear Security Administration Consortium for Verification Technology under Department of Energy National Nuclear Security Administration award number DE-NA0002534. The authors would like to thank the National Nuclear Security Administration Consortium for Verification Technology under Department of Energy National Nuclear Security Administration award number DE-NA0002534 .",

year = "2020",

month = feb,

day = "21",

doi = "10.1016/j.nima.2018.10.063",

language = "English (US)",

volume = "954",

journal = "Nuclear Instruments and Methods in Physics Research, Section A: Accelerators, Spectrometers, Detectors and Associated Equipment",

issn = "0168-9002",

publisher = "Elsevier B.V.",

}

TY - JOUR

T1 - A comparison of machine learning methods for automated gamma-ray spectroscopy

AU - Kamuda, Mark

AU - Zhao, Jifu

AU - Huff, Kathryn

N1 - The authors would like to thank the National Nuclear Security Administration Consortium for Verification Technology under Department of Energy National Nuclear Security Administration award number DE-NA0002534. The authors would like to thank the National Nuclear Security Administration Consortium for Verification Technology under Department of Energy National Nuclear Security Administration award number DE-NA0002534 .

PY - 2020/2/21

Y1 - 2020/2/21

N2 - Pattern recognition algorithms such as artificial neural networks (NNs) and convolution neural networks (CNNs) are prime candidates to perform automated gamma-ray spectroscopy. The way these models train and operate mimic how trained spectroscopists identify spectra. These models have shown promise in identifying gamma-ray spectra with large calibration drift and unknown background radiation fields. In this work, two algorithms for mixtures of radioisotopes based on NN and CNN are presented and evaluated.

AB - Pattern recognition algorithms such as artificial neural networks (NNs) and convolution neural networks (CNNs) are prime candidates to perform automated gamma-ray spectroscopy. The way these models train and operate mimic how trained spectroscopists identify spectra. These models have shown promise in identifying gamma-ray spectra with large calibration drift and unknown background radiation fields. In this work, two algorithms for mixtures of radioisotopes based on NN and CNN are presented and evaluated.

KW - Automated isotope identification

KW - Gamma-ray spectroscopy

KW - Neural networks

UR - https://www.scopus.com/pages/publications/85055748393

UR - https://www.scopus.com/pages/publications/85055748393#tab=citedBy

U2 - 10.1016/j.nima.2018.10.063

DO - 10.1016/j.nima.2018.10.063

M3 - Review article

AN - SCOPUS:85055748393

SN - 0168-9002

VL - 954

JO - Nuclear Instruments and Methods in Physics Research, Section A: Accelerators, Spectrometers, Detectors and Associated Equipment

JF - Nuclear Instruments and Methods in Physics Research, Section A: Accelerators, Spectrometers, Detectors and Associated Equipment

M1 - 161385

ER -

A comparison of machine learning methods for automated gamma-ray spectroscopy

Abstract

Keywords

ASJC Scopus subject areas

Online availability

Library availability

Related links

Fingerprint

Cite this