Source localization for neutron imaging systems using convolutional neural networks

The nuclear imaging system at the National Ignition Facility (NIF) is a crucial diagnostic for determining the geometry of inertial confinement fusion implosions. The geometry is reconstructed from a neutron aperture image via a set of reconstruction algorithms using an iterative Bayesian inference...

Full description

Saved in:
Bibliographic Details
Published inReview of scientific instruments Vol. 95; no. 6
Main Authors Saavedra, Gary, Geppert-Kleinrath, Verena, Danly, Chris, Durocher, Mora, Wilde, Carl, Fatherley, Valerie, Mendoza, Emily, Tafoya, Landon, Volegov, Petr, Fittinghoff, David, Rubery, Michael, Freeman, Matthew S.
Format Journal Article
LanguageEnglish
Published United States 01.06.2024
Online AccessGet full text

Cover

Abstract The nuclear imaging system at the National Ignition Facility (NIF) is a crucial diagnostic for determining the geometry of inertial confinement fusion implosions. The geometry is reconstructed from a neutron aperture image via a set of reconstruction algorithms using an iterative Bayesian inference approach. An important step in these reconstruction algorithms is finding the fusion source location within the camera field-of-view. Currently, source localization is achieved via an iterative optimization algorithm. In this paper, we introduce a machine learning approach for source localization. Specifically, we train a convolutional neural network to predict source locations given a neutron aperture image. We show that this approach decreases computation time by several orders of magnitude compared to the current optimization-based source localization while achieving similar accuracy on both synthetic data and a collection of recent NIF deuterium–tritium shots.
AbstractList The nuclear imaging system at the National Ignition Facility (NIF) is a crucial diagnostic for determining the geometry of inertial confinement fusion implosions. The geometry is reconstructed from a neutron aperture image via a set of reconstruction algorithms using an iterative Bayesian inference approach. An important step in these reconstruction algorithms is finding the fusion source location within the camera field-of-view. Currently, source localization is achieved via an iterative optimization algorithm. In this paper, we introduce a machine learning approach for source localization. Specifically, we train a convolutional neural network to predict source locations given a neutron aperture image. We show that this approach decreases computation time by several orders of magnitude compared to the current optimization-based source localization while achieving similar accuracy on both synthetic data and a collection of recent NIF deuterium-tritium shots.
Author Fittinghoff, David
Fatherley, Valerie
Danly, Chris
Geppert-Kleinrath, Verena
Mendoza, Emily
Freeman, Matthew S.
Durocher, Mora
Tafoya, Landon
Volegov, Petr
Rubery, Michael
Wilde, Carl
Saavedra, Gary
Author_xml – sequence: 1
  givenname: Gary
  surname: Saavedra
  fullname: Saavedra, Gary
  organization: Los Alamos National Laboratory
– sequence: 2
  givenname: Verena
  surname: Geppert-Kleinrath
  fullname: Geppert-Kleinrath, Verena
  organization: Los Alamos National Laboratory
– sequence: 3
  givenname: Chris
  surname: Danly
  fullname: Danly, Chris
  organization: Los Alamos National Laboratory
– sequence: 4
  givenname: Mora
  surname: Durocher
  fullname: Durocher, Mora
  organization: Los Alamos National Laboratory
– sequence: 5
  givenname: Carl
  surname: Wilde
  fullname: Wilde, Carl
  organization: Los Alamos National Laboratory
– sequence: 6
  givenname: Valerie
  surname: Fatherley
  fullname: Fatherley, Valerie
  organization: Los Alamos National Laboratory
– sequence: 7
  givenname: Emily
  surname: Mendoza
  fullname: Mendoza, Emily
  organization: Los Alamos National Laboratory
– sequence: 8
  givenname: Landon
  surname: Tafoya
  fullname: Tafoya, Landon
  organization: Los Alamos National Laboratory
– sequence: 9
  givenname: Petr
  surname: Volegov
  fullname: Volegov, Petr
  organization: Lawrence Livermore National Laboratory
– sequence: 10
  givenname: David
  surname: Fittinghoff
  fullname: Fittinghoff, David
  organization: Lawrence Livermore National Laboratory
– sequence: 11
  givenname: Michael
  surname: Rubery
  fullname: Rubery, Michael
  organization: Lawrence Livermore National Laboratory
– sequence: 12
  givenname: Matthew S.
  surname: Freeman
  fullname: Freeman, Matthew S.
  organization: Los Alamos National Laboratory
BackLink https://www.ncbi.nlm.nih.gov/pubmed/38888398$$D View this record in MEDLINE/PubMed
BookMark eNo9kEtLw0AUhQep2LS68A9I1kLqPDKvpRRfUHBRXYebmUmJppkykyj115u01bM598LH5Z4zQ5PWtw6ha4IXBAt2xxeYYp5LeoYSgpXOpKBsghKMWZ4JmaspmsX4gQdxQi7QlKlBTKsErde-D8aljTfQ1D_Q1b5NKx_S1vVdGOZ6C5u63aRxHzu3jWkfx8349ss3_UhDM7LhYN23D5_xEp1X0ER3dfI5en98eFs-Z6vXp5fl_SrzVIgu01iCrixYm0sgQquKV6UCIzixhgHnUjhOqLVEMGuIqIA4IUUJQDnTVrM5ujne3fXl1tliF4Znw774SzcAt0cgmro7RPtnCC7G6gpenKpjv708Yfg
CODEN RSINAK
ContentType Journal Article
Copyright Author(s)
2024 Author(s). All article content, except where otherwise noted, is licensed under a Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Copyright_xml – notice: Author(s)
– notice: 2024 Author(s). All article content, except where otherwise noted, is licensed under a Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
DBID AJDQP
NPM
DOI 10.1063/5.0205472
DatabaseName AIP Open Access Journals
PubMed
DatabaseTitle PubMed
DatabaseTitleList PubMed
Database_xml – sequence: 1
  dbid: NPM
  name: PubMed
  url: https://proxy.k.utb.cz/login?url=http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed
  sourceTypes: Index Database
– sequence: 2
  dbid: AJDQP
  name: AIP Open Access Journals
  url: https://publishing.aip.org/librarians/open-access-policy
  sourceTypes: Publisher
DeliveryMethod fulltext_linktorsrc
Discipline Sciences (General)
EISSN 1089-7623
ExternalDocumentID 38888398
rsi
Genre Journal Article
GrantInformation_xml – fundername: U.S. Department of Energy
  grantid: 89233218CNA000001
  funderid: https://doi.org/10.13039/100000015
GroupedDBID ---
-DZ
-~X
.DC
123
2-P
29P
4.4
5RE
5VS
85S
A9.
AAAAW
AABDS
AAEUA
AAPUP
AAYIH
ABFTF
ABJNI
ACBEA
ACBRY
ACGFO
ACGFS
ACLYJ
ACNCT
ACZLF
ADCTM
AEGXH
AEJMO
AENEX
AFATG
AFHCQ
AGKCL
AGLKD
AGMXG
AGTJO
AHSDT
AIAGR
AJDQP
AJJCW
AJQPL
ALEPV
ALMA_UNASSIGNED_HOLDINGS
AQWKA
ATXIE
AWQPM
BPZLN
CS3
DU5
EBS
ESX
F5P
FDOHQ
FFFMQ
HAM
L7B
M43
M6X
M71
M73
N9A
NPSNA
O-B
P2P
RIP
RNS
RQS
TAE
TN5
VQA
WH7
XSW
YNT
YZZ
~02
53G
BDMKI
NPM
ID FETCH-LOGICAL-o266t-907a9fdadd47a1698f5fb8ac651dc3a5576e512dd163dc16fa1e676baa2539d93
IEDL.DBID AJDQP
ISSN 0034-6748
IngestDate Thu Oct 24 09:56:35 EDT 2024
Tue Jun 25 17:50:09 EDT 2024
IsDoiOpenAccess true
IsOpenAccess true
IsPeerReviewed true
IsScholarly true
Issue 6
Language English
License All article content, except where otherwise noted, is licensed under a Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
2024 Author(s). All article content, except where otherwise noted, is licensed under a Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-o266t-907a9fdadd47a1698f5fb8ac651dc3a5576e512dd163dc16fa1e676baa2539d93
ORCID 0000-0003-0361-7500
0000-0002-6869-5772
0000-0002-0108-6200
0000-0002-3496-0419
0000-0003-2169-0306
0000-0002-9366-8258
0000-0002-8787-1671
0000-0003-0793-6621
0000-0001-6427-9555
0000-0002-6089-8911
0000-0003-4415-2818
OpenAccessLink http://dx.doi.org/10.1063/5.0205472
PMID 38888398
PageCount 10
ParticipantIDs pubmed_primary_38888398
scitation_primary_10_1063_5_0205472
PublicationCentury 2000
PublicationDate 2024-06-01
PublicationDateYYYYMMDD 2024-06-01
PublicationDate_xml – month: 06
  year: 2024
  text: 2024-06-01
  day: 01
PublicationDecade 2020
PublicationPlace United States
PublicationPlace_xml – name: United States
PublicationTitle Review of scientific instruments
PublicationTitleAlternate Rev Sci Instrum
PublicationYear 2024
References Wilson, Grim, Tregillis, Wilke, Patel, Sepke, Morgan, Hatarik, Loomis, Wilde (c16) 2010; 81
Volegov, Danly, Fittinghoff, Guler, Merrill, Wilde (c7) 2014; 85
Yang, Wang (c11) 2020; 107
Abu-Shawareb, Acree, Adams, Adams, Addis, Aden, Adrian, Afeyan, Aggleton, Aghaian (c1) 2022; 129
Danly, Christensen, Fatherley, Fittinghoff, Grim, Hibbard, Izumi, Jedlovec, Merrill, Schmidt (c5) 2016; 87
Volegov, Danly, Fittinghoff, Grim, Guler, Izumi, Ma, Merrill, Warrick, Wilde, Wilson (c6) 2014; 85
Dong, Yu, Cao, Shi, Ma (c23) 2020; 14
Guler, Volegov, Danly, Grim, Merrill, Wilde (c10) 2012; 83
Mahesh (c21) 2013; 40
Geppert-Kleinrath, Freeman, Hurlbut, Merrill, Tinsley, Volegov, Wilde (c4) 2018; 89
Merrill, Bower, Buckles, Clark, Danly, Drury, Dzenitis, Fatherley, Fittinghoff, Gallegos (c2) 2012; 83
Nelder, Mead (c8) 1965; 7
Moore, Schlossberg, Appelbe, Chandler, Crilly, Eckart, Forrest, Glebov, Grim, Hartouni (c17) 2023; 94
Shanker, Hu, Hung (c18) 1996; 24
Zhuang, Qi, Duan, Xi, Zhu, Zhu, Xiong, He (c22) 2021; 109
Li, Liu, Yang, Peng, Zhou (c12) 2022; 33
Yang, Shami (c15) 2020; 415
Valente, António, Mora, Jardim (c3) 2023; 9
Price, Coope, Byatt (c9) 2002; 113
References_xml – volume: 107
  start-page: 1048
  year: 2020
  ident: c11
  article-title: Artificial neural networks for neuroscientists: A primer
  publication-title: Neuron
  contributor:
    fullname: Wang
– volume: 14
  start-page: 241
  year: 2020
  ident: c23
  article-title: A survey on ensemble learning
  publication-title: Front. Comput. Sci.
  contributor:
    fullname: Ma
– volume: 415
  start-page: 295
  year: 2020
  ident: c15
  article-title: On hyperparameter optimization of machine learning algorithms: Theory and practice
  publication-title: Neurocomputing
  contributor:
    fullname: Shami
– volume: 109
  start-page: 43
  year: 2021
  ident: c22
  article-title: A comprehensive survey on transfer learning
  publication-title: Proc. IEEE
  contributor:
    fullname: He
– volume: 7
  start-page: 308
  year: 1965
  ident: c8
  article-title: A simplex method for function minimization
  publication-title: Comput. J.
  contributor:
    fullname: Mead
– volume: 83
  start-page: 10D316
  year: 2012
  ident: c10
  article-title: Simultaneous usage of pinhole and penumbral apertures for imaging small scale neutron sources from inertial confinement fusion experiments
  publication-title: Rev. Sci. Instrum.
  contributor:
    fullname: Wilde
– volume: 40
  start-page: 077301
  year: 2013
  ident: c21
  article-title: The essential physics of medical imaging, third edition
  publication-title: Med. Phys.
  contributor:
    fullname: Mahesh
– volume: 94
  start-page: 061102
  year: 2023
  ident: c17
  article-title: Neutron time of flight (nToF) detectors for inertial fusion experiments
  publication-title: Rev. Sci. Instrum.
  contributor:
    fullname: Hartouni
– volume: 129
  start-page: 075001
  year: 2022
  ident: c1
  article-title: Lawson criterion for ignition exceeded in an inertial fusion experiment
  publication-title: Phys. Rev. Lett.
  contributor:
    fullname: Aghaian
– volume: 9
  start-page: 207
  year: 2023
  ident: c3
  article-title: Developments in image processing using deep learning and reinforcement learning
  publication-title: J. Imaging
  contributor:
    fullname: Jardim
– volume: 113
  start-page: 5
  year: 2002
  ident: c9
  article-title: A convergent variant of the Nelder–Mead algorithm
  publication-title: J. Optim. Theory Appl.
  contributor:
    fullname: Byatt
– volume: 33
  start-page: 6999
  year: 2022
  ident: c12
  article-title: A survey of convolutional neural networks: Analysis, applications, and prospects
  publication-title: IEEE Trans. Neural Networks Learn. Syst.
  contributor:
    fullname: Zhou
– volume: 89
  start-page: 10I142
  year: 2018
  ident: c4
  article-title: A liquid VI scintillator cell for fast-gated neutron imaging
  publication-title: Rev. Sci. Instrum.
  contributor:
    fullname: Wilde
– volume: 81
  start-page: 10D335
  year: 2010
  ident: c16
  article-title: Modeling the national ignition facility neutron imaging system
  publication-title: Rev. Sci. Instrum.
  contributor:
    fullname: Wilde
– volume: 83
  start-page: 10D317
  year: 2012
  ident: c2
  article-title: The neutron imaging diagnostic at NIF (invited)
  publication-title: Rev. Sci. Instrum.
  contributor:
    fullname: Gallegos
– volume: 85
  start-page: 023508
  year: 2014
  ident: c6
  article-title: Neutron source reconstruction from pinhole imaging at national ignition facility
  publication-title: Rev. Sci. Instrum.
  contributor:
    fullname: Wilson
– volume: 24
  start-page: 385
  year: 1996
  ident: c18
  article-title: Effect of data standardization on neural network training
  publication-title: Omega
  contributor:
    fullname: Hung
– volume: 87
  start-page: 11D703
  year: 2016
  ident: c5
  article-title: Combined neutron and x-ray imaging at the National Ignition Facility (invited)
  publication-title: Rev. Sci. Instrum.
  contributor:
    fullname: Schmidt
– volume: 85
  start-page: 123506
  year: 2014
  ident: c7
  article-title: Self characterization of a coded aperture array for neutron source imaging
  publication-title: Rev. Sci. Instrum.
  contributor:
    fullname: Wilde
SSID ssj0000511
Score 2.4896967
Snippet The nuclear imaging system at the National Ignition Facility (NIF) is a crucial diagnostic for determining the geometry of inertial confinement fusion...
SourceID pubmed
scitation
SourceType Index Database
Publisher
Title Source localization for neutron imaging systems using convolutional neural networks
URI http://dx.doi.org/10.1063/5.0205472
https://www.ncbi.nlm.nih.gov/pubmed/38888398
Volume 95
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
link http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwjV09T8MwED2VdoAFUT7LR2WpDDAY4jh24hEBVVWpVVGp1C1yYgcxkCLa_n8ujildkJgyOJGlc3zvne_8DuA6zGLLjDFUcVbQSAhOE60ltTZmuQ2KOHCXwkZjOZhFw7mYN6D3RwZf8ntxh4xGRDH62VYYqyBqQuth-PQy-XW4gtWN8XhEq94ZPwJC2x9vAcwuYkud5t5Ckv4B7HsKSB7qNWtDw5aH0PabbEluvBL07RFMp-5onTjA8RcmCbJMUtp1dYRN3j9ckyFS6zEvSVXF_kaqSnL_R-E8lWKle7h67-UxzPrPr48D6rsg0AWC54pi9KpVYdAPRbFmUiWFKLJE51Iwk3MtMGCwiNrGILMyOZOFZlbGMtM6FFwZxU-gWS5KewYEqRs3iQrykFvcu6HOAiPC3CZBhm4mKDpwWhsp_aylLlKOATJSqKQDvY3VNoMufy15KlJv5PN_vXUBeyEyg7re6hKaq6-1vUJkX2Vdv7Jd2BlPRt-XUKBd
link.rule.ids 315,783,787,27904,27938,27939,76745
linkProvider American Institute of Physics
linkToHtml http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwjV3PT8IwFG4MHPBixJ_4s4ke9FBd17XbjkQliEA0QMJt6dbOkOggDv5_X7sKHD3tsC1rXtf3fa_v9XsI3fppqKlSisSM5iTgnJFISkG0DmmmvTz07KGwwVB0J0FvyqeuNsechYFBlA9ytrBJfBO2OwOSL-Ccq8VGcECwR_4AXIcHIXjgutGk4TVUb_eeP943rpjTqmUeC4jpqvEnLbT98hb0NAB1qgT4FsZ09tGeI4e4XQ2miXZ0cYCabvmV-M5pRN8fotHIbrpjC0XuKCUG_okLvTKb23j2bdsP4UqpucSmvv0Tmxpz96_Bd4yWpb3YSvDyCE06L-OnLnH9EcgcYHVJIK6Vca7AQwWhpCKOcp6nkcwEpypjkkMooQHPlQLOpTIqckm1CEUqpc9ZrGJ2jGrFvNCnCAOpYyqKvcxnGla1L1NPcT_TkZeCA_LyFjqpjJQsKhGMhEHoDOQqaqGbtdXWN21mW7CEJ87IZ_966ho1uuNBP-m_Dt_O0a4P_KGqyrpAteXPSl8C_i_TKzfLv6CXrG0
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Source+localization+for+neutron+imaging+systems+using+convolutional+neural+networks&rft.jtitle=Review+of+scientific+instruments&rft.au=Saavedra%2C+Gary&rft.au=Geppert-Kleinrath%2C+Verena&rft.au=Danly%2C+Chris&rft.au=Durocher%2C+Mora&rft.date=2024-06-01&rft.issn=0034-6748&rft.eissn=1089-7623&rft.volume=95&rft.issue=6&rft_id=info:doi/10.1063%2F5.0205472&rft.externalDocID=rsi
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0034-6748&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0034-6748&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0034-6748&client=summon