Accelerating multi-coil MR image reconstruction using weak supervision

Deep-learning-based MR image reconstruction in settings where large fully sampled dataset collection is infeasible requires methods that effectively use both under-sampled and fully sampled datasets. This paper evaluates a weakly supervised, multi-coil, physics-guided approach to MR image reconstruc...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:Magma (New York, N.Y.) Ročník 38; číslo 1; s. 37 - 51
Hlavní autori: Atalık, Arda, Chopra, Sumit, Sodickson, Daniel K.
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: Cham Springer International Publishing 01.02.2025
Predmet:
ISSN:1352-8661, 1352-8661
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Abstract Deep-learning-based MR image reconstruction in settings where large fully sampled dataset collection is infeasible requires methods that effectively use both under-sampled and fully sampled datasets. This paper evaluates a weakly supervised, multi-coil, physics-guided approach to MR image reconstruction, leveraging both dataset types, to improve both the quality and robustness of reconstruction. A physics-guided end-to-end variational network (VarNet) is pretrained in a self-supervised manner using a 4 × under-sampled dataset following the self-supervised learning via data undersampling (SSDU) methodology. The pre-trained weights are transferred to another VarNet, which is fine-tuned using a smaller, fully sampled dataset by optimizing multi-scale structural similarity (MS-SSIM) loss in image space. The proposed methodology is compared with fully self-supervised and fully supervised training. Reconstruction quality improvements in SSIM, PSNR, and NRMSE when abundant training data is available (the high-data regime), and enhanced robustness when training data is scarce (the low-data regime) are demonstrated using weak supervision for knee and brain MR image reconstructions at 8 × and 10 × acceleration, respectively. Multi-coil physics-guided MR image reconstruction using both under-sampled and fully sampled datasets is achievable with transfer learning and fine-tuning. This methodology can provide improved reconstruction quality in the high-data regime and improved robustness in the low-data regime at high acceleration rates.
AbstractList Deep-learning-based MR image reconstruction in settings where large fully sampled dataset collection is infeasible requires methods that effectively use both under-sampled and fully sampled datasets. This paper evaluates a weakly supervised, multi-coil, physics-guided approach to MR image reconstruction, leveraging both dataset types, to improve both the quality and robustness of reconstruction. A physics-guided end-to-end variational network (VarNet) is pretrained in a self-supervised manner using a 4 × under-sampled dataset following the self-supervised learning via data undersampling (SSDU) methodology. The pre-trained weights are transferred to another VarNet, which is fine-tuned using a smaller, fully sampled dataset by optimizing multi-scale structural similarity (MS-SSIM) loss in image space. The proposed methodology is compared with fully self-supervised and fully supervised training. Reconstruction quality improvements in SSIM, PSNR, and NRMSE when abundant training data is available (the high-data regime), and enhanced robustness when training data is scarce (the low-data regime) are demonstrated using weak supervision for knee and brain MR image reconstructions at 8 × and 10 × acceleration, respectively. Multi-coil physics-guided MR image reconstruction using both under-sampled and fully sampled datasets is achievable with transfer learning and fine-tuning. This methodology can provide improved reconstruction quality in the high-data regime and improved robustness in the low-data regime at high acceleration rates.
Deep-learning-based MR image reconstruction in settings where large fully sampled dataset collection is infeasible requires methods that effectively use both under-sampled and fully sampled datasets. This paper evaluates a weakly supervised, multi-coil, physics-guided approach to MR image reconstruction, leveraging both dataset types, to improve both the quality and robustness of reconstruction. A physics-guided end-to-end variational network (VarNet) is pretrained in a self-supervised manner using a 4 × under-sampled dataset following the self-supervised learning via data undersampling (SSDU) methodology. The pre-trained weights are transferred to another VarNet, which is fine-tuned using a smaller, fully sampled dataset by optimizing multi-scale structural similarity (MS-SSIM) loss in image space. The proposed methodology is compared with fully self-supervised and fully supervised training. Reconstruction quality improvements in SSIM, PSNR, and NRMSE when abundant training data is available (the high-data regime), and enhanced robustness when training data is scarce (the low-data regime) are demonstrated using weak supervision for knee and brain MR image reconstructions at 8 × and 10 × acceleration, respectively. Multi-coil physics-guided MR image reconstruction using both under-sampled and fully sampled datasets is achievable with transfer learning and fine-tuning. This methodology can provide improved reconstruction quality in the high-data regime and improved robustness in the low-data regime at high acceleration rates.Deep-learning-based MR image reconstruction in settings where large fully sampled dataset collection is infeasible requires methods that effectively use both under-sampled and fully sampled datasets. This paper evaluates a weakly supervised, multi-coil, physics-guided approach to MR image reconstruction, leveraging both dataset types, to improve both the quality and robustness of reconstruction. A physics-guided end-to-end variational network (VarNet) is pretrained in a self-supervised manner using a 4 × under-sampled dataset following the self-supervised learning via data undersampling (SSDU) methodology. The pre-trained weights are transferred to another VarNet, which is fine-tuned using a smaller, fully sampled dataset by optimizing multi-scale structural similarity (MS-SSIM) loss in image space. The proposed methodology is compared with fully self-supervised and fully supervised training. Reconstruction quality improvements in SSIM, PSNR, and NRMSE when abundant training data is available (the high-data regime), and enhanced robustness when training data is scarce (the low-data regime) are demonstrated using weak supervision for knee and brain MR image reconstructions at 8 × and 10 × acceleration, respectively. Multi-coil physics-guided MR image reconstruction using both under-sampled and fully sampled datasets is achievable with transfer learning and fine-tuning. This methodology can provide improved reconstruction quality in the high-data regime and improved robustness in the low-data regime at high acceleration rates.
Deep-learning-based MR image reconstruction in settings where large fully sampled dataset collection is infeasible requires methods that effectively use both under-sampled and fully sampled datasets. This paper evaluates a weakly supervised, multi-coil, physics-guided approach to MR image reconstruction, leveraging both dataset types, to improve both the quality and robustness of reconstruction. A physics-guided end-to-end variational network (VarNet) is pretrained in a self-supervised manner using a 4 under-sampled dataset following the self-supervised learning via data undersampling (SSDU) methodology. The pre-trained weights are transferred to another VarNet, which is fine-tuned using a smaller, fully sampled dataset by optimizing multi-scale structural similarity (MS-SSIM) loss in image space. The proposed methodology is compared with fully self-supervised and fully supervised training. Reconstruction quality improvements in SSIM, PSNR, and NRMSE when abundant training data is available (the high-data regime), and enhanced robustness when training data is scarce (the low-data regime) are demonstrated using weak supervision for knee and brain MR image reconstructions at 8 and 10 acceleration, respectively. Multi-coil physics-guided MR image reconstruction using both under-sampled and fully sampled datasets is achievable with transfer learning and fine-tuning. This methodology can provide improved reconstruction quality in the high-data regime and improved robustness in the low-data regime at high acceleration rates.
Author Atalık, Arda
Sodickson, Daniel K.
Chopra, Sumit
Author_xml – sequence: 1
  givenname: Arda
  orcidid: 0000-0003-3439-7838
  surname: Atalık
  fullname: Atalık, Arda
  email: Arda.Atalik@nyu.edu
  organization: Center for Data Science, New York University, Bernard and Irene Schwartz Center for Biomedical Imaging, Department of Radiology, New York University Grossman School of Medicine, Department of Radiology, Center for Advanced Imaging Innovation and Research (CAI2R), New York University Grossman School of Medicine
– sequence: 2
  givenname: Sumit
  surname: Chopra
  fullname: Chopra, Sumit
  organization: Courant Institute of Mathematical Sciences, New York University, Bernard and Irene Schwartz Center for Biomedical Imaging, Department of Radiology, New York University Grossman School of Medicine, Department of Radiology, Center for Advanced Imaging Innovation and Research (CAI2R), New York University Grossman School of Medicine
– sequence: 3
  givenname: Daniel K.
  surname: Sodickson
  fullname: Sodickson, Daniel K.
  organization: Center for Data Science, New York University, Bernard and Irene Schwartz Center for Biomedical Imaging, Department of Radiology, New York University Grossman School of Medicine, Department of Radiology, Center for Advanced Imaging Innovation and Research (CAI2R), New York University Grossman School of Medicine
BackLink https://www.ncbi.nlm.nih.gov/pubmed/39382814$$D View this record in MEDLINE/PubMed
BookMark eNp9kE1LAzEURYNU7If-ARcySzej-e7MshSrQkUQXYdM-qZMnSY1mSj-e1OniisX4YXHuQ_OHaOBdRYQOif4imA8vQ4EM8ZzTNMjFMucHqERYYLmhZRk8Oc_ROMQNhhTIjA7QUNWsoIWhI_QYmYMtOB119h1to1t1-TGNW328JQ1W72GzINxNnQ-mq5xNothD36Afs1C3IF_b0Jan6LjWrcBzg5zgl4WN8_zu3z5eHs_ny1zQ8uiyysjSV0AFJSTUsgKcym0lpiBoLAiJYPaVNOVJBg4n0opucaCV4JUpRG1lGyCLvu7O-_eIoRObZuQBFptwcWgGCFcYFqKMqEXBzRWW1ipnU8-_lP9uCeA9oDxLgQP9S9CsNoXrPqCVSpYfResaAqxPhQSbNfg1cZFb5Pzf6kvfcZ8gQ
Cites_doi 10.1109/MSP.2019.2950640
10.1109/MSP.2019.2950557
10.1109/TIT.2006.871582
10.1109/TMI.2018.2865356
10.1109/MSP.2007.914731
10.1561/2200000016
10.1109/TIP.2003.819861
10.1109/MSP.2019.2943645
10.1002/mrm.28378
10.1109/TMI.2022.3147426
10.1002/mrm.29759
10.1109/TIP.2010.2047910
10.1109/ACSSC.2003.1292216
10.1002/mrm.10171
10.1002/mrm.28148
10.1002/nbm.4798
10.1007/978-3-319-24574-4_28
10.1148/ryai.2020190007
10.1109/TMI.2017.2760978
10.1002/mrm.26977
10.1109/TPAMI.2018.2883941
10.1109/TMI.2022.3199155
10.1007/978-3-030-59713-9_7
10.1002/mrm.21391
10.1109/TMI.2021.3075856
10.1109/CVPR.2018.00196
10.1007/978-3-031-43999-5_47
10.1109/TMI.2018.2863670
10.1002/mrm.1910380414
10.1002/(SICI)1522-2594(199911)42:5<952::AID-MRM16>3.0.CO;2-S
10.1109/TMI.2019.2927101
ContentType Journal Article
Copyright The Author(s), under exclusive licence to European Society for Magnetic Resonance in Medicine and Biology (ESMRMB) 2024 Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
2024. The Author(s), under exclusive licence to European Society for Magnetic Resonance in Medicine and Biology (ESMRMB).
Copyright_xml – notice: The Author(s), under exclusive licence to European Society for Magnetic Resonance in Medicine and Biology (ESMRMB) 2024 Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
– notice: 2024. The Author(s), under exclusive licence to European Society for Magnetic Resonance in Medicine and Biology (ESMRMB).
DBID AAYXX
CITATION
CGR
CUY
CVF
ECM
EIF
NPM
7X8
DOI 10.1007/s10334-024-01206-2
DatabaseName CrossRef
Medline
MEDLINE
MEDLINE (Ovid)
MEDLINE
MEDLINE
PubMed
MEDLINE - Academic
DatabaseTitle CrossRef
MEDLINE
Medline Complete
MEDLINE with Full Text
PubMed
MEDLINE (Ovid)
MEDLINE - Academic
DatabaseTitleList
MEDLINE - Academic
MEDLINE
Database_xml – sequence: 1
  dbid: NPM
  name: PubMed
  url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed
  sourceTypes: Index Database
– sequence: 2
  dbid: 7X8
  name: MEDLINE - Academic
  url: https://search.proquest.com/medline
  sourceTypes: Aggregation Database
DeliveryMethod fulltext_linktorsrc
Discipline Medicine
EISSN 1352-8661
EndPage 51
ExternalDocumentID 39382814
10_1007_s10334_024_01206_2
Genre Journal Article
GrantInformation_xml – fundername: National Institute of Biomedical Imaging and Bioengineering
  grantid: NIH P41 EB017183
  funderid: http://dx.doi.org/10.13039/100000070
– fundername: National Science Foundation
  grantid: 1922658
  funderid: http://dx.doi.org/10.13039/100000001
– fundername: NIBIB NIH HHS
  grantid: NIH P41 EB017183
– fundername: National Science Foundation
  grantid: 1922658
– fundername: NIBIB NIH HHS
  grantid: P41 EB017183
GroupedDBID ---
--K
-53
-5E
-5G
-BR
-EM
-Y2
-~C
.86
.VR
06C
06D
0R~
0VY
1B1
1N0
1SB
203
28-
29M
29~
2J2
2JN
2JY
2KG
2LR
2P1
2VQ
2~H
30V
3SX
3V.
4.4
406
408
409
40D
40E
53G
5GY
5QI
5VS
67Z
6NX
7X7
88E
88I
8FE
8FG
8FH
8FI
8FJ
8FW
8UJ
95-
95.
95~
96X
AAAVM
AABHQ
AACDK
AAEDT
AAHNG
AAIAL
AAJBT
AAJKR
AALRI
AANXM
AANZL
AAQFI
AAQXK
AARHV
AARTL
AASML
AATNV
AATVU
AAUYE
AAWCG
AAXUO
AAYIU
AAYQN
AAYTO
AAYZH
ABAKF
ABBBX
ABBXA
ABDZT
ABECU
ABFTV
ABHLI
ABHQN
ABJNI
ABJOX
ABKCH
ABKTR
ABMNI
ABMQK
ABNWP
ABPLI
ABQBU
ABQSL
ABSXP
ABTEG
ABTKH
ABTMW
ABULA
ABUWG
ABWNU
ABWVN
ABXPI
ACAOD
ACBXY
ACDTI
ACGFS
ACGOD
ACHSB
ACHXU
ACIUM
ACKNC
ACMDZ
ACMLO
ACOKC
ACOMO
ACPIV
ACRPL
ACSNA
ACUDM
ACZOJ
ADBBV
ADHHG
ADHIR
ADINQ
ADKNI
ADKPE
ADMUD
ADNMO
ADPHR
ADRFC
ADTPH
ADURQ
ADYFF
ADZKW
AEBTG
AEFIE
AEFQL
AEGAL
AEGNC
AEJHL
AEJRE
AEKMD
AEMSY
AENEX
AEOHA
AEPYU
AESKC
AETLH
AEUYN
AEVLU
AEXYK
AFBBN
AFEXP
AFKRA
AFLOW
AFQWF
AFWTZ
AFZKB
AGAYW
AGDGC
AGGDS
AGJBK
AGMZJ
AGQEE
AGQMX
AGRTI
AGWIL
AGWZB
AGYKE
AHAVH
AHBYD
AHKAY
AHMBA
AHSBF
AHYZX
AIAKS
AIGIU
AIIXL
AILAN
AITGF
AJBLW
AJRNO
AJZVZ
AKMHD
AKRWK
ALIPV
ALMA_UNASSIGNED_HOLDINGS
ALWAN
AMKLP
AMXSW
AMYLF
AMYQR
AOCGG
ARAPS
ARMRJ
ASPBG
AVWKF
AXYYD
AZFZN
AZQEC
B-.
BA0
BBWZM
BDATZ
BENPR
BGLVJ
BGNMA
BHPHI
BKSAR
BPHCQ
BSONS
BVXVI
CAG
CCPQU
COF
CS3
CSCUP
DDRTE
DL5
DNIVK
DPUIP
DU5
DWQXO
EBLON
EBS
EIOEI
EJD
EMOBN
EN4
EPAXT
ESBYG
F5P
FDB
FEDTE
FERAY
FFXSO
FGOYB
FIGPU
FINBP
FNLPD
FRRFC
FSGXE
FWDCC
FYUFA
G-Y
G-Z
GGCAI
GGRSB
GJIRD
GNUQQ
GNWQR
GQ6
GQ7
GQ8
GRRUI
GXS
H13
HCIFZ
HF~
HG5
HG6
HMCUK
HMJXF
HQYDN
HRMNR
HVGLF
HZ~
I09
IHE
IJ-
IKXTQ
IWAJR
IXC
IXD
IXE
IZIGR
IZQ
I~X
I~Z
J-C
J0Z
JBSCW
JCJTX
JZLTJ
KDC
KOV
KOW
KPH
LAK
LK5
LLZTM
M1P
M2P
M41
M4Y
M7R
MA-
N2Q
NB0
NDZJH
NPVJJ
NQ-
NQJWS
NU0
O9-
O93
O9G
O9I
O9J
OAM
P19
P2P
P62
P9S
PCBAR
PF0
PQQKQ
PROAC
PSQYO
PT4
PT5
Q2X
QOK
QOR
QOS
R2-
R4E
R89
R9I
RHV
RIG
RNI
RNS
ROL
RPX
RPZ
RRX
RSV
RZK
S16
S1Z
S26
S27
S28
S37
S3B
SAP
SCLPG
SDE
SDH
SEW
SHX
SISQX
SJYHP
SMD
SNE
SNPRN
SNX
SOHCF
SOJ
SPISZ
SRMVM
SSLCW
SSXJD
SSZ
STPWE
SV3
SZ9
SZN
T13
T16
TSG
TSK
TSV
TT1
TUC
U2A
U9L
UG4
UHS
UKHRP
UOJIU
UTJUX
UZXMN
VC2
VFIZW
W23
W48
WJK
WK6
WK8
YLTOR
Z45
Z7X
Z82
Z83
Z88
Z8R
Z8V
Z8W
ZMTXR
ZOVNA
~EX
AAPKM
AAYXX
ABBRH
ABDBE
ABFSG
ABRTQ
ACSTC
AEZWR
AFDZB
AFHIU
AFOHR
AHPBZ
AHWEU
AIXLP
ATHPR
AYFIA
CITATION
ADHKG
AEUPX
AFPUW
AGQPQ
CGR
CUY
CVF
ECM
EIF
NPM
PHGZM
PHGZT
PJZUB
PPXIY
PQGLB
7X8
ID FETCH-LOGICAL-c298t-bc61f8ee8241956b0465aa603e52ed193efcb7d610e4476664a054b51b9c5f663
IEDL.DBID RSV
ISICitedReferencesCount 0
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001329076400001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 1352-8661
IngestDate Thu Oct 02 06:01:46 EDT 2025
Mon Jul 21 05:59:26 EDT 2025
Sat Nov 29 03:00:50 EST 2025
Fri Feb 21 02:38:10 EST 2025
IsPeerReviewed true
IsScholarly true
Issue 1
Keywords Accelerated imaging
Weak supervision
Transfer learning
MR image reconstruction
Machine learning
Self-supervised learning
Language English
License 2024. The Author(s), under exclusive licence to European Society for Magnetic Resonance in Medicine and Biology (ESMRMB).
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c298t-bc61f8ee8241956b0465aa603e52ed193efcb7d610e4476664a054b51b9c5f663
Notes ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ORCID 0000-0003-3439-7838
PMID 39382814
PQID 3114502959
PQPubID 23479
PageCount 15
ParticipantIDs proquest_miscellaneous_3114502959
pubmed_primary_39382814
crossref_primary_10_1007_s10334_024_01206_2
springer_journals_10_1007_s10334_024_01206_2
PublicationCentury 2000
PublicationDate 2025-02-01
PublicationDateYYYYMMDD 2025-02-01
PublicationDate_xml – month: 02
  year: 2025
  text: 2025-02-01
  day: 01
PublicationDecade 2020
PublicationPlace Cham
PublicationPlace_xml – name: Cham
– name: Germany
PublicationSubtitle Official Journal of the European Society for Magnetic Resonance in Medicine and Biology
PublicationTitle Magma (New York, N.Y.)
PublicationTitleAbbrev Magn Reson Mater Phy
PublicationTitleAlternate MAGMA
PublicationYear 2025
Publisher Springer International Publishing
Publisher_xml – name: Springer International Publishing
References MA Griswold (1206_CR3) 2002; 47
Y Han (1206_CR12) 2019; 39
B Yaman (1206_CR18) 2022; 35
M Yurt (1206_CR20) 2022; 41
JA Fessler (1206_CR27) 2020; 37
S Boyd (1206_CR28) 2011; 3
EJ Candès (1206_CR6) 2008; 25
1206_CR9
Z Wang (1206_CR36) 2004; 13
MJ Muckley (1206_CR35) 2021; 40
1206_CR19
1206_CR39
J Schlemper (1206_CR7) 2017; 37
1206_CR15
1206_CR34
1206_CR33
SUH Dar (1206_CR40) 2020; 84
1206_CR10
1206_CR32
HK Aggarwal (1206_CR11) 2018; 38
1206_CR31
1206_CR30
JD Gibbons (1206_CR38) 2011
KP Pruessmann (1206_CR2) 1999; 42
D Liang (1206_CR13) 2020; 37
M Hollander (1206_CR37) 1999
Y Yang (1206_CR26) 2018; 42
F Knoll (1206_CR14) 2020; 37
K Hammernik (1206_CR8) 2018; 79
MV Afonso (1206_CR24) 2010; 19
B Yaman (1206_CR17) 2020; 84
M Lustig (1206_CR5) 2007; 58
1206_CR29
1206_CR23
C Qin (1206_CR25) 2018; 38
DK Sodickson (1206_CR1) 1997; 38
1206_CR22
DL Donoho (1206_CR4) 2006; 52
Y Korkmaz (1206_CR16) 2022; 41
AD Desai (1206_CR21) 2023; 90
References_xml – volume: 37
  start-page: 128
  issue: 1
  year: 2020
  ident: 1206_CR14
  publication-title: IEEE Signal Process Mag
  doi: 10.1109/MSP.2019.2950640
– ident: 1206_CR15
– volume: 37
  start-page: 141
  issue: 1
  year: 2020
  ident: 1206_CR13
  publication-title: IEEE Signal Process Mag
  doi: 10.1109/MSP.2019.2950557
– ident: 1206_CR32
– volume-title: Nonparametric statistical methods
  year: 1999
  ident: 1206_CR37
– volume: 52
  start-page: 1289
  issue: 4
  year: 2006
  ident: 1206_CR4
  publication-title: IEEE Trans Inf Theory
  doi: 10.1109/TIT.2006.871582
– volume: 38
  start-page: 394
  issue: 2
  year: 2018
  ident: 1206_CR11
  publication-title: IEEE Trans Med Imaging
  doi: 10.1109/TMI.2018.2865356
– volume: 25
  start-page: 21
  issue: 2
  year: 2008
  ident: 1206_CR6
  publication-title: IEEE Signal Process Mag
  doi: 10.1109/MSP.2007.914731
– volume: 3
  start-page: 1
  issue: 1
  year: 2011
  ident: 1206_CR28
  publication-title: Found Trends Mach Learn
  doi: 10.1561/2200000016
– volume: 13
  start-page: 600
  issue: 4
  year: 2004
  ident: 1206_CR36
  publication-title: IEEE Trans Image Process
  doi: 10.1109/TIP.2003.819861
– ident: 1206_CR23
– volume: 37
  start-page: 33
  issue: 1
  year: 2020
  ident: 1206_CR27
  publication-title: IEEE Signal Process Mag
  doi: 10.1109/MSP.2019.2943645
– volume: 84
  start-page: 3172
  issue: 6
  year: 2020
  ident: 1206_CR17
  publication-title: Magn Reson Med
  doi: 10.1002/mrm.28378
– volume: 41
  start-page: 1747
  issue: 7
  year: 2022
  ident: 1206_CR16
  publication-title: IEEE Trans Med Imaging
  doi: 10.1109/TMI.2022.3147426
– ident: 1206_CR19
– volume: 90
  start-page: 2052
  issue: 5
  year: 2023
  ident: 1206_CR21
  publication-title: Magn Reson Med
  doi: 10.1002/mrm.29759
– volume: 19
  start-page: 2345
  issue: 9
  year: 2010
  ident: 1206_CR24
  publication-title: IEEE Trans Image Process
  doi: 10.1109/TIP.2010.2047910
– ident: 1206_CR29
  doi: 10.1109/ACSSC.2003.1292216
– volume: 47
  start-page: 1202
  issue: 6
  year: 2002
  ident: 1206_CR3
  publication-title: Magn Reson Med
  doi: 10.1002/mrm.10171
– volume: 84
  start-page: 663
  issue: 2
  year: 2020
  ident: 1206_CR40
  publication-title: Magn Reson Med
  doi: 10.1002/mrm.28148
– volume: 35
  start-page: 4798
  issue: 12
  year: 2022
  ident: 1206_CR18
  publication-title: NMR Biomed
  doi: 10.1002/nbm.4798
– ident: 1206_CR30
  doi: 10.1007/978-3-319-24574-4_28
– ident: 1206_CR34
  doi: 10.1148/ryai.2020190007
– volume: 37
  start-page: 491
  issue: 2
  year: 2017
  ident: 1206_CR7
  publication-title: IEEE Trans Med Imaging
  doi: 10.1109/TMI.2017.2760978
– volume: 79
  start-page: 3055
  issue: 6
  year: 2018
  ident: 1206_CR8
  publication-title: Magn Reson Med
  doi: 10.1002/mrm.26977
– ident: 1206_CR33
– ident: 1206_CR39
– volume: 42
  start-page: 521
  issue: 3
  year: 2018
  ident: 1206_CR26
  publication-title: IEEE Trans Pattern Anal Mach Intell
  doi: 10.1109/TPAMI.2018.2883941
– volume: 41
  start-page: 3895
  issue: 12
  year: 2022
  ident: 1206_CR20
  publication-title: IEEE Trans Med Imaging
  doi: 10.1109/TMI.2022.3199155
– ident: 1206_CR10
  doi: 10.1007/978-3-030-59713-9_7
– volume: 58
  start-page: 1182
  issue: 6
  year: 2007
  ident: 1206_CR5
  publication-title: Magn Reson Med
  doi: 10.1002/mrm.21391
– volume: 40
  start-page: 2306
  issue: 9
  year: 2021
  ident: 1206_CR35
  publication-title: IEEE Trans Med Imaging
  doi: 10.1109/TMI.2021.3075856
– volume-title: Nonparametric statistical inference
  year: 2011
  ident: 1206_CR38
– ident: 1206_CR9
  doi: 10.1109/CVPR.2018.00196
– ident: 1206_CR22
  doi: 10.1007/978-3-031-43999-5_47
– ident: 1206_CR31
– volume: 38
  start-page: 280
  issue: 1
  year: 2018
  ident: 1206_CR25
  publication-title: IEEE Trans Med Imaging
  doi: 10.1109/TMI.2018.2863670
– volume: 38
  start-page: 591
  issue: 4
  year: 1997
  ident: 1206_CR1
  publication-title: Magn Reson Med
  doi: 10.1002/mrm.1910380414
– volume: 42
  start-page: 952
  issue: 5
  year: 1999
  ident: 1206_CR2
  publication-title: Magn Reson Med
  doi: 10.1002/(SICI)1522-2594(199911)42:5<952::AID-MRM16>3.0.CO;2-S
– volume: 39
  start-page: 377
  issue: 2
  year: 2019
  ident: 1206_CR12
  publication-title: IEEE Trans Med Imaging
  doi: 10.1109/TMI.2019.2927101
SSID ssj0021503
Score 2.3847928
Snippet Deep-learning-based MR image reconstruction in settings where large fully sampled dataset collection is infeasible requires methods that effectively use both...
SourceID proquest
pubmed
crossref
springer
SourceType Aggregation Database
Index Database
Publisher
StartPage 37
SubjectTerms Basic Science - Reconstruction algorithms and artificial intelligence
Biomedical Engineering and Bioengineering
Computer Appl. in Life Sciences
Deep Learning
Health Informatics
Humans
Image Processing, Computer-Assisted - methods
Imaging
Magnetic Resonance Imaging
Medicine
Medicine & Public Health
Radiology
Research Article
Solid State Physics
Supervised Machine Learning
Title Accelerating multi-coil MR image reconstruction using weak supervision
URI https://link.springer.com/article/10.1007/s10334-024-01206-2
https://www.ncbi.nlm.nih.gov/pubmed/39382814
https://www.proquest.com/docview/3114502959
Volume 38
WOSCitedRecordID wos001329076400001&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
journalDatabaseRights – providerCode: PRVAVX
  databaseName: Springer LINK
  customDbUrl:
  eissn: 1352-8661
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0021503
  issn: 1352-8661
  databaseCode: RSV
  dateStart: 19970101
  isFulltext: true
  titleUrlDefault: https://link.springer.com/search?facet-content-type=%22Journal%22
  providerName: Springer Nature
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwnV1LT8MwDLZgIMSF92M8piBxg0q0TdrmOCEmDmxC46HdoiRNpwnoJrrB38fJ2gEaQoJ7m1aO6-_7ascGOFWIaQhzocd5bDwaGe4p31DPtndCtEiTTLqW-Tdxp5P0evy2PBRWVNXuVUrSReovh93CEBcJbNVEgDoYA-8Swl1iBzZ07x5nMgspTlgej_n5vu8QNMcr53KiDmpa6_97yQ1YK6klaU59YRMWTL4FK-0yeb4NrabWCDJ2y_M-cZWEnh4Onkm7SwYvGFeIU8ezjrLE1sT3ybuRT6SYjGxQsb_WduChdXV_ee2VYxQ8HfBk7Ckd-VliTIJgjWpIoSJmUkYXoWGBSZHAmUyrOEUeZSiNUc5QiTxOMV9xzTJkJLtQy4e52QdCZcb8KDV2YAeVKuaZH8dcSmaTg5qldTirLCtG024Z4rMvsjWNQNMIZxoR1OGkMr5Ap7aZCpmb4aQQIao0dhFwxuuwN92V2XohD1El-rQO59UWiPK7K3552MHfLj-E1cBO-nX12UdQQ9ObY1jWb-NB8dqAxbiXNJzbfQDje8-8
linkProvider Springer Nature
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwnV1LS8NAEB6kinrx_YjPFbxpoEl2k-yxiKViW6RW6W3ZbDalqGkxrf59Z9OkKhVB78kmzEzm-77M7CzAeYSYhjDn2ZwH2qa-5nbkaGqb8U6IFnGYyHxkfjNot8Nej98Vm8Kystu9LEnmmfrLZjfPw0Vc0zXhog7GxLtIEbHMxPzO_eNMZiHF8YrtMT_f9x2C5njlXE00h5r6-v9ecgPWCmpJatNY2IQFnW7Bcqsonm9DvaYUgoxxedoneSehrYaDZ9LqkMEL5hWSq-PZRFlieuL75F3LJ5JNRiapmF9rO_BQv-5eNeziGAVbuTwc25HynSTUOkSwRjUUoSJmUvpVTzNXx0jgdKKiIEYepSkNUM5QiTwuYk7EFUuQkexCJR2meh8IlQlz_FibAzuojAKeOEHApWSmOKhYbMFFaVkxmk7LEJ9zkY1pBJpG5KYRrgVnpfEFBrWpVMhUDyeZ8FClsarLGbdgb-qV2Xoe91AlOtSCy9IFovjusl8edvC3y09hpdFtNUXzpn17CKuuOfU379U-ggq6QR_DknobD7LXkzz4PgDE6NG4
linkToPdf http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV1ZS8NAEB6kSvHF-4jnCr5paJPsNtnHohbFthQv-rZsNptS1LTYVv--szlqpSKI78kmzEz2-77MsQCnIWIawpxnc-5rm9Y0t0NHU9uMd0K0iIJYpiPzm367HXS7vDPTxZ9WuxcpyaynwUxpSsaVYRRXZhrfPA8XdE0FhYuaGDfhRWoK6Y1ev3-aSi6kO17eKvPzfd_haI5jzuVHU9hprP7_hddgJaecpJ7FyDos6GQDyq08qb4JjbpSCD4mFJIeSSsMbTXov5DWHem_4n5DUtU8nTRLTK18j3xo-UxGk6HZbMwvty14bFw9XFzb-fEKtnJ5MLZDVXPiQOsAQRxVUohKmUlZq3qauTpCYqdjFfoR8itNqY8yh0rkdyFzQq5YjExlG0rJING7QKiMmVOLtDnIg8rQ57Hj-1xKZpKGikUWnBVWFsNsiob4mpdsTCPQNCI1jXAtOCkcITDYTQZDJnowGQkP1RurupxxC3YyD03X87iH6tGhFpwX7hD59zj65WF7f7v8GMqdy4Zo3rRv92HZNYcBpyXcB1BCL-hDWFLv4_7o7SiNw0-80dqc
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=Accelerating+multi-coil+MR+image+reconstruction+using+weak+supervision&rft.jtitle=Magma+%28New+York%2C+N.Y.%29&rft.au=Atal%C4%B1k%2C+Arda&rft.au=Chopra%2C+Sumit&rft.au=Sodickson%2C+Daniel+K&rft.date=2025-02-01&rft.eissn=1352-8661&rft.volume=38&rft.issue=1&rft.spage=37&rft_id=info:doi/10.1007%2Fs10334-024-01206-2&rft_id=info%3Apmid%2F39382814&rft.externalDocID=39382814
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1352-8661&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1352-8661&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1352-8661&client=summon