Application of Bayes' theorem for pulse shape discrimination
A Bayesian approach is proposed for pulse shape discrimination of photons and neutrons in liquid organic scinitillators. Instead of drawing a decision boundary, each pulse is assigned a photon or neutron confidence probability. In addition, this allows for photon and neutron classification on an eventbyevent basis. The sum of those confidence probabilities is used to estimate the number of photon and neutron instances in the data. An iterative scheme, similar to an expectationmaximization algorithm for Gaussian mixtures, is used to infer the ratio of photonstoneutrons in each measurement. Therefore, the probability space adapts to data with varying photontoneutron ratios. A timecorrelated measurement of Am–Be and separate measurements of ^{137}Cs, ^{60}Co and ^{232}Th photon sources were used to construct libraries of neutrons and photons. These libraries were then used to produce synthetic data sets with varying ratios of photonstoneutrons. Probability weighted method that we implemented was found to maintain neutron acceptance rate of up to 90% up to photontoneutron ratio of 2000, and performed 9% better than the decision boundary approach. Furthermore, the iterative approach appropriately changed the probability space with an increasing number of photons which kept the neutron population estimate from unrealistically increasing.
 Authors:

^{[1]};
^{[2]};
^{[2]};
^{[2]}
 Sandia National Lab. (SNLCA), Livermore, CA (United States)
 Univ. of Michigan, Ann Arbor, MI (United States)
 Publication Date:
 Report Number(s):
 SAND20151190J
Journal ID: ISSN 01689002; 567086
 Grant/Contract Number:
 AC0494AL85000
 Type:
 Accepted Manuscript
 Journal Name:
 Nuclear Instruments and Methods in Physics Research. Section A, Accelerators, Spectrometers, Detectors and Associated Equipment
 Additional Journal Information:
 Journal Volume: 795; Journal Issue: C; Journal ID: ISSN 01689002
 Publisher:
 Elsevier
 Research Org:
 Sandia National Lab. (SNLCA), Livermore, CA (United States)
 Sponsoring Org:
 USDOE National Nuclear Security Administration (NNSA), Office of Defense Nuclear Nonproliferation (NA20)
 Country of Publication:
 United States
 Language:
 English
 Subject:
 46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND TECHNOLOGY; pulse shape discrimination; liquid scintillator; Bayes׳ theorem; expectationmaximization
 OSTI Identifier:
 1235311
 Alternate Identifier(s):
 OSTI ID: 1422673
Marleau, Peter, Monterial, Mateusz, Clarke, Shaun, and Pozzi, Sara. Application of Bayes' theorem for pulse shape discrimination. United States: N. p.,
Web. doi:10.1016/j.nima.2015.06.014.
Marleau, Peter, Monterial, Mateusz, Clarke, Shaun, & Pozzi, Sara. Application of Bayes' theorem for pulse shape discrimination. United States. doi:10.1016/j.nima.2015.06.014.
Marleau, Peter, Monterial, Mateusz, Clarke, Shaun, and Pozzi, Sara. 2015.
"Application of Bayes' theorem for pulse shape discrimination". United States.
doi:10.1016/j.nima.2015.06.014. https://www.osti.gov/servlets/purl/1235311.
@article{osti_1235311,
title = {Application of Bayes' theorem for pulse shape discrimination},
author = {Marleau, Peter and Monterial, Mateusz and Clarke, Shaun and Pozzi, Sara},
abstractNote = {A Bayesian approach is proposed for pulse shape discrimination of photons and neutrons in liquid organic scinitillators. Instead of drawing a decision boundary, each pulse is assigned a photon or neutron confidence probability. In addition, this allows for photon and neutron classification on an eventbyevent basis. The sum of those confidence probabilities is used to estimate the number of photon and neutron instances in the data. An iterative scheme, similar to an expectationmaximization algorithm for Gaussian mixtures, is used to infer the ratio of photonstoneutrons in each measurement. Therefore, the probability space adapts to data with varying photontoneutron ratios. A timecorrelated measurement of Am–Be and separate measurements of 137Cs, 60Co and 232Th photon sources were used to construct libraries of neutrons and photons. These libraries were then used to produce synthetic data sets with varying ratios of photonstoneutrons. Probability weighted method that we implemented was found to maintain neutron acceptance rate of up to 90% up to photontoneutron ratio of 2000, and performed 9% better than the decision boundary approach. Furthermore, the iterative approach appropriately changed the probability space with an increasing number of photons which kept the neutron population estimate from unrealistically increasing.},
doi = {10.1016/j.nima.2015.06.014},
journal = {Nuclear Instruments and Methods in Physics Research. Section A, Accelerators, Spectrometers, Detectors and Associated Equipment},
number = C,
volume = 795,
place = {United States},
year = {2015},
month = {6}
}