Accelerated Optimization of Implicit Neural Representations for CT Reconstruction

Inspired by their success in solving challenging inverse prob-lems in computer vision, implicit neural representations (INRs) have been recently proposed for reconstruction in low-dose/sparse-view X-ray computed tomography (CT). An INR represents a CT image as a small-scale neural network that takes...

Full description

Saved in:
Bibliographic Details
Published in:Proceedings (International Symposium on Biomedical Imaging) pp. 1 - 5
Main Authors: Najaf, Mahrokh, Ongie, Gregory
Format: Conference Proceeding
Language:English
Published: IEEE 14.04.2025
Subjects:
ISSN:1945-8452
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Abstract Inspired by their success in solving challenging inverse prob-lems in computer vision, implicit neural representations (INRs) have been recently proposed for reconstruction in low-dose/sparse-view X-ray computed tomography (CT). An INR represents a CT image as a small-scale neural network that takes spatial coordinates as inputs and outputs attenuation values. Fitting an INR to sinogram data is similar to classical model-based iterative reconstruction methods. However, training INRs with losses and gradient-based algorithms can be prohibitively slow, taking many thousands of iterations to converge. This paper investigates strategies to accelerate the optimization of INRs for CT reconstruction. In particular, we propose two approaches: (1) using a modified loss function with improved conditioning, and (2) an algorithm based on the alternating direction method of multipliers. We illustrate that both of these approaches significantly accelerate INR-based reconstruction of a synthetic breast CT phantom in a sparse-view setting.
AbstractList Inspired by their success in solving challenging inverse prob-lems in computer vision, implicit neural representations (INRs) have been recently proposed for reconstruction in low-dose/sparse-view X-ray computed tomography (CT). An INR represents a CT image as a small-scale neural network that takes spatial coordinates as inputs and outputs attenuation values. Fitting an INR to sinogram data is similar to classical model-based iterative reconstruction methods. However, training INRs with losses and gradient-based algorithms can be prohibitively slow, taking many thousands of iterations to converge. This paper investigates strategies to accelerate the optimization of INRs for CT reconstruction. In particular, we propose two approaches: (1) using a modified loss function with improved conditioning, and (2) an algorithm based on the alternating direction method of multipliers. We illustrate that both of these approaches significantly accelerate INR-based reconstruction of a synthetic breast CT phantom in a sparse-view setting.
Author Ongie, Gregory
Najaf, Mahrokh
Author_xml – sequence: 1
  givenname: Mahrokh
  surname: Najaf
  fullname: Najaf, Mahrokh
  organization: Marquette University,Department of Mathematical and Statistical Sciences,Milwaukee,WI,USA
– sequence: 2
  givenname: Gregory
  surname: Ongie
  fullname: Ongie, Gregory
  organization: Marquette University,Department of Mathematical and Statistical Sciences,Milwaukee,WI,USA
BookMark eNo1kMtOwzAURA0CiVLyB0j4B1J8_UjsZYkoRKqooN1XjnMjGeUlx13A1xNes5k7R6O7mGty0Q89EnIHbAXAzH25fygzpjSsOONqNSMNXMozkpjcaCFAcaZ4dk4WYKRKtVT8iiTT9M5m5VIKJhfkde0cthhsxJruxug7_2mjH3o6NLTsxtY7H-kLnoJt6RuOASfs409jos0QaHGYsZtTDCf3jW_IZWPbCZM_X5L95vFQPKfb3VNZrLeph1zH1NS5ywxzoPPGMmnyGhrO0UA9n9I5A1YK1LpmTvHK8Uy4imEloRZWV2JJbn-_ekQ8jsF3Nnwc_zcQXwBQU9k
ContentType Conference Proceeding
DBID 6IE
6IL
CBEJK
RIE
RIL
DOI 10.1109/ISBI60581.2025.10981244
DatabaseName IEEE Electronic Library (IEL) Conference Proceedings
IEEE Xplore POP ALL
IEEE Xplore All Conference Proceedings
IEEE Xplore
IEEE Proceedings Order Plans (POP All) 1998-Present
DatabaseTitleList
Database_xml – sequence: 1
  dbid: RIE
  name: IEEE/IET Electronic Library (IEL) (UW System Shared)
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
DeliveryMethod fulltext_linktorsrc
Discipline Engineering
EISBN 9798331520526
EISSN 1945-8452
EndPage 5
ExternalDocumentID 10981244
Genre orig-research
GrantInformation_xml – fundername: NSF
  grantid: CCF-2153371
  funderid: 10.13039/100000001
GroupedDBID 23N
6IE
6IF
6IK
6IL
6IN
AAJGR
AAWTH
ABLEC
ADZIZ
ALMA_UNASSIGNED_HOLDINGS
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
CBEJK
CHZPO
IEGSK
IPLJI
M43
OCL
RIE
RIL
RNS
ID FETCH-LOGICAL-i178t-9d7c690c187fa0497d1f22e91d97d4cc91a43e88d0c52bc263cb0eb41d3a8b3
IEDL.DBID RIE
ISICitedReferencesCount 0
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001546451000570&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
IngestDate Wed Jun 11 06:03:21 EDT 2025
IsPeerReviewed false
IsScholarly true
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-i178t-9d7c690c187fa0497d1f22e91d97d4cc91a43e88d0c52bc263cb0eb41d3a8b3
PageCount 5
ParticipantIDs ieee_primary_10981244
PublicationCentury 2000
PublicationDate 2025-April-14
PublicationDateYYYYMMDD 2025-04-14
PublicationDate_xml – month: 04
  year: 2025
  text: 2025-April-14
  day: 14
PublicationDecade 2020
PublicationTitle Proceedings (International Symposium on Biomedical Imaging)
PublicationTitleAbbrev ISBI
PublicationYear 2025
Publisher IEEE
Publisher_xml – name: IEEE
SSID ssj0000744304
Score 2.2884746
Snippet Inspired by their success in solving challenging inverse prob-lems in computer vision, implicit neural representations (INRs) have been recently proposed for...
SourceID ieee
SourceType Publisher
StartPage 1
SubjectTerms Computed tomography
Coor-dinate Based Neural Networks
CT Reconstruction
Fitting
Image reconstruction
Implicit Neural Representations
Iterative methods
Model-based Iterative Reconstruction
Neural networks
Optimization
Phantoms
Reconstruction algorithms
Training
X-ray imaging
Title Accelerated Optimization of Implicit Neural Representations for CT Reconstruction
URI https://ieeexplore.ieee.org/document/10981244
WOSCitedRecordID wos001546451000570&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV09T8MwELVoxQALX0V8ywNrSpw4sT1CRUUlVArt0K1y7hypEqSIFn4_ZzdtYWBgs6JESnxy3j3fvWfGrjMsLTh0kSohjmROsbCWFh4q6wrQBhGDieuj6vf1eGwGtVg9aGGcc6H5zLX9MNTycQaffquMVrgJeNRgDaXypVhrvaFCWCiJm9c9XHTrTW941_NVP08Dk6y9evrXOSoBRrp7_3yBfdbaCPL4YA01B2zLVYds94eX4BF7vgUgCPHOD8if6EfwViss-azkvdA3Pl1w78VhX_lL6H-tZUfVnFPmyjsj7rnoxlG2xYbd-1HnIarPS4imQulFZFABkV0QWpWWMn-FokwSZwTSUAIYYWXqtMYYsqSAJE-hiF0hBaZWF-kxa1azyp0wTlkjMdcy1dbb-Qks4pxoh5FGOFAZqFPW8nMzeV8aYkxW03L2x_VztuMj4IswQl6wJn2Hu2Tb8LWYzj-uQhi_ASOen4U
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV07T8MwED5BQQIWXkW88cCaEudR2yNUoEaUUmiHbpVzdqRKkCJa-P2cTdrCwMBmWYnk-OTcfb77vgO4TE2h0RobiALDIGmSLbSmg2eEtjlKZYzxIq4d0e3K4VD1KrK658JYa33xmW24oc_lmwl-uKsyOuHK-6NVWHOtsyq61uJKhbxhQui8quKih6-y_k3m8n4OCEZpY_7-r04q3pHcbf9zCTtQX1LyWG_hbHZhxZZ7sPVDTXAfnq4RyYk47QfDHulX8FpxLNmkYJmvHB_PmFPj0C_s2VfAVsSjcsoodmWtAXNodKkpW4f-3e2g1Q6qjgnBmAs5C5QRSHAXuRSFpthfGF5EkVXc0DBBVFwnsZXShJhGOUbNGPPQ5gk3sZZ5fAC1clLaQ2AUNxJ2LWKpnaAfN3nYJOChEsUtihTFEdTd3ozeviUxRvNtOf5j_gI22oOHzqiTde9PYNNZw6VkeHIKNfomewbr-DkbT9_PvUm_AKRWos4
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%3Abook&rft.genre=proceeding&rft.title=Proceedings+%28International+Symposium+on+Biomedical+Imaging%29&rft.atitle=Accelerated+Optimization+of+Implicit+Neural+Representations+for+CT+Reconstruction&rft.au=Najaf%2C+Mahrokh&rft.au=Ongie%2C+Gregory&rft.date=2025-04-14&rft.pub=IEEE&rft.eissn=1945-8452&rft.spage=1&rft.epage=5&rft_id=info:doi/10.1109%2FISBI60581.2025.10981244&rft.externalDocID=10981244