Auto‐segmentation of thoraco‐abdominal organs in pediatric dynamic MRI

Purpose Dynamic magnetic resonance imaging (dMRI) is a practical imaging modality for capturing information about regional thoracic‐abdominal components and their dynamics in healthy children and pediatric patients with thoracic insufficiency syndrome (TIS). We propose an auto‐segmentation set‐up fo...

Full description

Saved in:
Bibliographic Details
Published in:Medical physics (Lancaster) Vol. 52; no. 11; pp. e70104 - n/a
Main Authors: Akhtar, Yusuf, Udupa, Jayaram K., Tong, Yubing, Liu, Tiange, Wu, Caiyun, Kogan, Rachel, Al‐Noury, Mostafa, Hosseini, Mahdie, Tong, Leihui, Mannikeri, Samarth, Odhner, Dewey, Mcdonough, Joseph M., Lott, Carina, Clark, Abigail, Cahill, Patrick J., Anari, Jason B., Torigian, Drew A.
Format: Journal Article
Language:English
Published: United States 01.11.2025
Subjects:
ISSN:0094-2405, 2473-4209, 2473-4209
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Abstract Purpose Dynamic magnetic resonance imaging (dMRI) is a practical imaging modality for capturing information about regional thoracic‐abdominal components and their dynamics in healthy children and pediatric patients with thoracic insufficiency syndrome (TIS). We propose an auto‐segmentation set‐up for the lungs, kidneys, liver, spleen, and thoraco‐abdominal skin outer boundary (Skn) in dMRI images. Methods The segmentation setup has been implemented in two steps, recognition and delineation, using two deep neural network (DL) architectures, DL‐R and DL‐D for the recognition and delineation steps, respectively. The encoder‐decoder framework in DL‐D utilizes features at four different resolution levels to counter the challenges involved in segmentation. dMRI sagittal slice acquisitions of 189 (near‐)normal subjects were evaluated, with an in‐plane spatial resolution of roughly 1 × 1 mm2 with 6.00 mm spacing between slices. We utilized images from 89 and 10 subjects at end inspiration for training and validation, respectively. For testing, we experimented with three scenarios utilizing: (1) the images of the 90 (=189‐89‐10) remaining subjects at end inspiration for testing, (2) the images of the remaining 90 subjects at end expiration for testing, and (3) the images of the other 99 (=89+10) subjects at end expiration for testing. In some situations, we can take advantage of the already available ground truth (GT) segmentation for an object in a subject at a particular respiratory phase to automatically segment the same object in the same subject at a different respiratory phase, and then refine the segmentation to create the final GT for all respiratory phases in the image of a subject. We anticipate that this process of creating GT would require minimal post hoc correction. In this spirit, we conducted separate experiments where we assumed to have GT of test subjects at the end expiration for scenario (1), end inspiration for (2), and end inspiration for (3). A major contribution in this paper is the different scenarios of training and testing that we have extensively evaluated with respect to respiratory phases and the subjects to which the images in the training and testing sets belong. Results Among these three scenarios of testing, for DL‐R, we achieve the best average location error (LE) of about 1 voxel for the lungs, kidneys, and spleen, and 1.5 voxels for the liver and Skn. The standard deviation (SD) of LE is about 1 or 2 voxels. For DL‐D, we achieve an average Dice coefficient (DC) of about 0.92 to 0.94 for the lungs, 0.82 for the kidneys, 0.90 for the liver, 0.81 for the spleen, and 0.93 for Skn. The SD of DC is lower (0.02 to 0.07) for the lungs, liver, and Skn and slightly higher (0.06 to 0.12) for the spleen and kidneys. Conclusions Motivated by applications in surgical planning for disorders such as TIS, adolescent idiopathic scoliosis, and early onset scoliosis, we have created an auto‐segmentation system for thoraco‐abdominal organs in dMRI acquisitions. This proposed setup copes with the challenges posed by low resolution, motion blur, inadequate contrast, and image intensity non‐standardness in dMRI images quite well.
AbstractList Dynamic magnetic resonance imaging (dMRI) is a practical imaging modality for capturing information about regional thoracic-abdominal components and their dynamics in healthy children and pediatric patients with thoracic insufficiency syndrome (TIS). We propose an auto-segmentation set-up for the lungs, kidneys, liver, spleen, and thoraco-abdominal skin outer boundary (Skn) in dMRI images.PURPOSEDynamic magnetic resonance imaging (dMRI) is a practical imaging modality for capturing information about regional thoracic-abdominal components and their dynamics in healthy children and pediatric patients with thoracic insufficiency syndrome (TIS). We propose an auto-segmentation set-up for the lungs, kidneys, liver, spleen, and thoraco-abdominal skin outer boundary (Skn) in dMRI images.The segmentation setup has been implemented in two steps, recognition and delineation, using two deep neural network (DL) architectures, DL-R and DL-D for the recognition and delineation steps, respectively. The encoder-decoder framework in DL-D utilizes features at four different resolution levels to counter the challenges involved in segmentation. dMRI sagittal slice acquisitions of 189 (near-)normal subjects were evaluated, with an in-plane spatial resolution of roughly 1 × 1 mm2 with 6.00 mm spacing between slices. We utilized images from 89 and 10 subjects at end inspiration for training and validation, respectively. For testing, we experimented with three scenarios utilizing: (1) the images of the 90 (=189-89-10) remaining subjects at end inspiration for testing, (2) the images of the remaining 90 subjects at end expiration for testing, and (3) the images of the other 99 (=89+10) subjects at end expiration for testing. In some situations, we can take advantage of the already available ground truth (GT) segmentation for an object in a subject at a particular respiratory phase to automatically segment the same object in the same subject at a different respiratory phase, and then refine the segmentation to create the final GT for all respiratory phases in the image of a subject. We anticipate that this process of creating GT would require minimal post hoc correction. In this spirit, we conducted separate experiments where we assumed to have GT of test subjects at the end expiration for scenario (1), end inspiration for (2), and end inspiration for (3). A major contribution in this paper is the different scenarios of training and testing that we have extensively evaluated with respect to respiratory phases and the subjects to which the images in the training and testing sets belong.METHODSThe segmentation setup has been implemented in two steps, recognition and delineation, using two deep neural network (DL) architectures, DL-R and DL-D for the recognition and delineation steps, respectively. The encoder-decoder framework in DL-D utilizes features at four different resolution levels to counter the challenges involved in segmentation. dMRI sagittal slice acquisitions of 189 (near-)normal subjects were evaluated, with an in-plane spatial resolution of roughly 1 × 1 mm2 with 6.00 mm spacing between slices. We utilized images from 89 and 10 subjects at end inspiration for training and validation, respectively. For testing, we experimented with three scenarios utilizing: (1) the images of the 90 (=189-89-10) remaining subjects at end inspiration for testing, (2) the images of the remaining 90 subjects at end expiration for testing, and (3) the images of the other 99 (=89+10) subjects at end expiration for testing. In some situations, we can take advantage of the already available ground truth (GT) segmentation for an object in a subject at a particular respiratory phase to automatically segment the same object in the same subject at a different respiratory phase, and then refine the segmentation to create the final GT for all respiratory phases in the image of a subject. We anticipate that this process of creating GT would require minimal post hoc correction. In this spirit, we conducted separate experiments where we assumed to have GT of test subjects at the end expiration for scenario (1), end inspiration for (2), and end inspiration for (3). A major contribution in this paper is the different scenarios of training and testing that we have extensively evaluated with respect to respiratory phases and the subjects to which the images in the training and testing sets belong.Among these three scenarios of testing, for DL-R, we achieve the best average location error (LE) of about 1 voxel for the lungs, kidneys, and spleen, and 1.5 voxels for the liver and Skn. The standard deviation (SD) of LE is about 1 or 2 voxels. For DL-D, we achieve an average Dice coefficient (DC) of about 0.92 to 0.94 for the lungs, 0.82 for the kidneys, 0.90 for the liver, 0.81 for the spleen, and 0.93 for Skn. The SD of DC is lower (0.02 to 0.07) for the lungs, liver, and Skn and slightly higher (0.06 to 0.12) for the spleen and kidneys.RESULTSAmong these three scenarios of testing, for DL-R, we achieve the best average location error (LE) of about 1 voxel for the lungs, kidneys, and spleen, and 1.5 voxels for the liver and Skn. The standard deviation (SD) of LE is about 1 or 2 voxels. For DL-D, we achieve an average Dice coefficient (DC) of about 0.92 to 0.94 for the lungs, 0.82 for the kidneys, 0.90 for the liver, 0.81 for the spleen, and 0.93 for Skn. The SD of DC is lower (0.02 to 0.07) for the lungs, liver, and Skn and slightly higher (0.06 to 0.12) for the spleen and kidneys.Motivated by applications in surgical planning for disorders such as TIS, adolescent idiopathic scoliosis, and early onset scoliosis, we have created an auto-segmentation system for thoraco-abdominal organs in dMRI acquisitions. This proposed setup copes with the challenges posed by low resolution, motion blur, inadequate contrast, and image intensity non-standardness in dMRI images quite well.CONCLUSIONSMotivated by applications in surgical planning for disorders such as TIS, adolescent idiopathic scoliosis, and early onset scoliosis, we have created an auto-segmentation system for thoraco-abdominal organs in dMRI acquisitions. This proposed setup copes with the challenges posed by low resolution, motion blur, inadequate contrast, and image intensity non-standardness in dMRI images quite well.
Purpose Dynamic magnetic resonance imaging (dMRI) is a practical imaging modality for capturing information about regional thoracic‐abdominal components and their dynamics in healthy children and pediatric patients with thoracic insufficiency syndrome (TIS). We propose an auto‐segmentation set‐up for the lungs, kidneys, liver, spleen, and thoraco‐abdominal skin outer boundary (Skn) in dMRI images. Methods The segmentation setup has been implemented in two steps, recognition and delineation, using two deep neural network (DL) architectures, DL‐R and DL‐D for the recognition and delineation steps, respectively. The encoder‐decoder framework in DL‐D utilizes features at four different resolution levels to counter the challenges involved in segmentation. dMRI sagittal slice acquisitions of 189 (near‐)normal subjects were evaluated, with an in‐plane spatial resolution of roughly 1 × 1 mm2 with 6.00 mm spacing between slices. We utilized images from 89 and 10 subjects at end inspiration for training and validation, respectively. For testing, we experimented with three scenarios utilizing: (1) the images of the 90 (=189‐89‐10) remaining subjects at end inspiration for testing, (2) the images of the remaining 90 subjects at end expiration for testing, and (3) the images of the other 99 (=89+10) subjects at end expiration for testing. In some situations, we can take advantage of the already available ground truth (GT) segmentation for an object in a subject at a particular respiratory phase to automatically segment the same object in the same subject at a different respiratory phase, and then refine the segmentation to create the final GT for all respiratory phases in the image of a subject. We anticipate that this process of creating GT would require minimal post hoc correction. In this spirit, we conducted separate experiments where we assumed to have GT of test subjects at the end expiration for scenario (1), end inspiration for (2), and end inspiration for (3). A major contribution in this paper is the different scenarios of training and testing that we have extensively evaluated with respect to respiratory phases and the subjects to which the images in the training and testing sets belong. Results Among these three scenarios of testing, for DL‐R, we achieve the best average location error (LE) of about 1 voxel for the lungs, kidneys, and spleen, and 1.5 voxels for the liver and Skn. The standard deviation (SD) of LE is about 1 or 2 voxels. For DL‐D, we achieve an average Dice coefficient (DC) of about 0.92 to 0.94 for the lungs, 0.82 for the kidneys, 0.90 for the liver, 0.81 for the spleen, and 0.93 for Skn. The SD of DC is lower (0.02 to 0.07) for the lungs, liver, and Skn and slightly higher (0.06 to 0.12) for the spleen and kidneys. Conclusions Motivated by applications in surgical planning for disorders such as TIS, adolescent idiopathic scoliosis, and early onset scoliosis, we have created an auto‐segmentation system for thoraco‐abdominal organs in dMRI acquisitions. This proposed setup copes with the challenges posed by low resolution, motion blur, inadequate contrast, and image intensity non‐standardness in dMRI images quite well.
Dynamic magnetic resonance imaging (dMRI) is a practical imaging modality for capturing information about regional thoracic-abdominal components and their dynamics in healthy children and pediatric patients with thoracic insufficiency syndrome (TIS). We propose an auto-segmentation set-up for the lungs, kidneys, liver, spleen, and thoraco-abdominal skin outer boundary (Skn) in dMRI images. The segmentation setup has been implemented in two steps, recognition and delineation, using two deep neural network (DL) architectures, DL-R and DL-D for the recognition and delineation steps, respectively. The encoder-decoder framework in DL-D utilizes features at four different resolution levels to counter the challenges involved in segmentation. dMRI sagittal slice acquisitions of 189 (near-)normal subjects were evaluated, with an in-plane spatial resolution of roughly 1 × 1 mm with 6.00 mm spacing between slices. We utilized images from 89 and 10 subjects at end inspiration for training and validation, respectively. For testing, we experimented with three scenarios utilizing: (1) the images of the 90 (=189-89-10) remaining subjects at end inspiration for testing, (2) the images of the remaining 90 subjects at end expiration for testing, and (3) the images of the other 99 (=89+10) subjects at end expiration for testing. In some situations, we can take advantage of the already available ground truth (GT) segmentation for an object in a subject at a particular respiratory phase to automatically segment the same object in the same subject at a different respiratory phase, and then refine the segmentation to create the final GT for all respiratory phases in the image of a subject. We anticipate that this process of creating GT would require minimal post hoc correction. In this spirit, we conducted separate experiments where we assumed to have GT of test subjects at the end expiration for scenario (1), end inspiration for (2), and end inspiration for (3). A major contribution in this paper is the different scenarios of training and testing that we have extensively evaluated with respect to respiratory phases and the subjects to which the images in the training and testing sets belong. Among these three scenarios of testing, for DL-R, we achieve the best average location error (LE) of about 1 voxel for the lungs, kidneys, and spleen, and 1.5 voxels for the liver and Skn. The standard deviation (SD) of LE is about 1 or 2 voxels. For DL-D, we achieve an average Dice coefficient (DC) of about 0.92 to 0.94 for the lungs, 0.82 for the kidneys, 0.90 for the liver, 0.81 for the spleen, and 0.93 for Skn. The SD of DC is lower (0.02 to 0.07) for the lungs, liver, and Skn and slightly higher (0.06 to 0.12) for the spleen and kidneys. Motivated by applications in surgical planning for disorders such as TIS, adolescent idiopathic scoliosis, and early onset scoliosis, we have created an auto-segmentation system for thoraco-abdominal organs in dMRI acquisitions. This proposed setup copes with the challenges posed by low resolution, motion blur, inadequate contrast, and image intensity non-standardness in dMRI images quite well.
Author Tong, Yubing
Tong, Leihui
Odhner, Dewey
Mcdonough, Joseph M.
Al‐Noury, Mostafa
Kogan, Rachel
Torigian, Drew A.
Wu, Caiyun
Anari, Jason B.
Lott, Carina
Hosseini, Mahdie
Udupa, Jayaram K.
Cahill, Patrick J.
Mannikeri, Samarth
Clark, Abigail
Akhtar, Yusuf
Liu, Tiange
Author_xml – sequence: 1
  givenname: Yusuf
  surname: Akhtar
  fullname: Akhtar, Yusuf
  organization: Vellore Institute of Technology, Katpadi
– sequence: 2
  givenname: Jayaram K.
  surname: Udupa
  fullname: Udupa, Jayaram K.
  email: jay@pennmedicine.upenn.edu
  organization: University of Pennsylvania
– sequence: 3
  givenname: Yubing
  surname: Tong
  fullname: Tong, Yubing
  organization: University of Pennsylvania
– sequence: 4
  givenname: Tiange
  surname: Liu
  fullname: Liu, Tiange
  organization: University of Science and Technology Beijing
– sequence: 5
  givenname: Caiyun
  surname: Wu
  fullname: Wu, Caiyun
  organization: University of Pennsylvania
– sequence: 6
  givenname: Rachel
  surname: Kogan
  fullname: Kogan, Rachel
  organization: University of Pennsylvania
– sequence: 7
  givenname: Mostafa
  surname: Al‐Noury
  fullname: Al‐Noury, Mostafa
  organization: University of Pennsylvania
– sequence: 8
  givenname: Mahdie
  surname: Hosseini
  fullname: Hosseini, Mahdie
  organization: University of Pennsylvania
– sequence: 9
  givenname: Leihui
  surname: Tong
  fullname: Tong, Leihui
  organization: Boston University
– sequence: 10
  givenname: Samarth
  surname: Mannikeri
  fullname: Mannikeri, Samarth
  organization: University of Pennsylvania
– sequence: 11
  givenname: Dewey
  surname: Odhner
  fullname: Odhner, Dewey
  organization: University of Pennsylvania
– sequence: 12
  givenname: Joseph M.
  surname: Mcdonough
  fullname: Mcdonough, Joseph M.
  organization: Children's Hospital of Philadelphia
– sequence: 13
  givenname: Carina
  surname: Lott
  fullname: Lott, Carina
  organization: Children's Hospital of Philadelphia
– sequence: 14
  givenname: Abigail
  surname: Clark
  fullname: Clark, Abigail
  organization: Children's Hospital of Philadelphia
– sequence: 15
  givenname: Patrick J.
  surname: Cahill
  fullname: Cahill, Patrick J.
  organization: Children's Hospital of Philadelphia
– sequence: 16
  givenname: Jason B.
  surname: Anari
  fullname: Anari, Jason B.
  organization: Children's Hospital of Philadelphia
– sequence: 17
  givenname: Drew A.
  surname: Torigian
  fullname: Torigian, Drew A.
  organization: University of Pennsylvania
BackLink https://www.ncbi.nlm.nih.gov/pubmed/41206361$$D View this record in MEDLINE/PubMed
BookMark eNp10EtOwzAQBmALFdEHSJwAZckmZWwncbKsKh5FrUAI1pZjOyUotkOcCHXHETgjJyGlBVas_pHm00j_jNHAOqsROsUwxQDkwtRTBhiiAzQiEaNhRCAboBFAFoUkgniIxt6_AEBCYzhCwwiTfkzwCN3OutZ9vn94vTbatqItnQ1cEbTPrhFyuxG5cqa0ogpcsxbWB6UNaq1K0TalDNTGCtPn6mFxjA4LUXl9ss8Jerq6fJzfhMu768V8tgwlxWkU4oyljCqS5ZQUhci00goLkDFomqhYSCyLhGVpnBOghcRZjiGVTClgQiVK0Ak6392tG_faad9yU3qpq0pY7TrPKWF98TghrKdne9rlRiteN6URzYb_9P-7JRvnfaOLX4KBb1_LTc2_X9vTcEffykpv_nV8db_zX6HLemM
Cites_doi 10.1002/mp.14141
10.2214/AJR.05.1446
10.1038/s41592‐020‐01008‐z
10.1007/s10278‐007‐9060‐5
10.1002/nbm.4131
10.1007/s12178‐023‐09848‐w
10.1016/j.neucom.2022.07.070
10.1109/CVPR.2016.90
10.1016/j.cviu.2009.06.003
10.3978/j.issn.2223‐3652.2014.03.01
10.1002/mp.15854
10.1109/TPAMI.2021.3059968
10.1109/TPAMI.2018.2858826
10.1073/pnas.89.12.5675
10.1109/3DV.2016.79
10.1117/1.JMI.4.1.014504
10.1186/s13244‐020‐00868‐5
10.1152/japplphysiol.01216.2009
10.1109/CVPR.2019.00326
10.1038/s41598‐018‐22941‐2
10.1109/CVPR.2018.00913
10.3390/electronics12051199
10.1109/42.836373
10.1007/s00586‐009‐1183‐6
10.3390/jcm11247276
10.1109/JBHI.2013.2282183
10.1016/j.media.2022.102527
10.1016/j.inffus.2022.09.031
10.1109/TMI.2014.2377694
10.1016/j.apmr.2024.02.692
10.1109/CVPR.2017.243
10.1007/978-3-319-55524-9_14
10.1016/j.media.2021.102088
10.2214/AJR.06.1329
10.1016/j.neuroimage.2022.118961
10.1148/radiol.2019181731
10.1109/TMI.2018.2806309
10.1109/ICCV.2017.322
10.1016/j.imu.2024.101504
10.1016/j.media.2014.04.003
ContentType Journal Article
Copyright 2025 The Author(s). published by Wiley Periodicals LLC on behalf of American Association of Physicists in Medicine.
2025 The Author(s). Medical Physics published by Wiley Periodicals LLC on behalf of American Association of Physicists in Medicine.
Copyright_xml – notice: 2025 The Author(s). published by Wiley Periodicals LLC on behalf of American Association of Physicists in Medicine.
– notice: 2025 The Author(s). Medical Physics published by Wiley Periodicals LLC on behalf of American Association of Physicists in Medicine.
DBID 24P
AAYXX
CITATION
CGR
CUY
CVF
ECM
EIF
NPM
7X8
DOI 10.1002/mp.70104
DatabaseName Wiley Online Library Open Access
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: 24P
  name: Wiley Online Library Open Access
  url: https://authorservices.wiley.com/open-science/open-access/browse-journals.html
  sourceTypes: Publisher
– sequence: 2
  dbid: NPM
  name: PubMed
  url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed
  sourceTypes: Index Database
– sequence: 3
  dbid: 7X8
  name: MEDLINE - Academic
  url: https://search.proquest.com/medline
  sourceTypes: Aggregation Database
DeliveryMethod fulltext_linktorsrc
Discipline Medicine
Physics
EISSN 2473-4209
EndPage n/a
ExternalDocumentID 41206361
10_1002_mp_70104
MP70104
Genre researchArticle
Journal Article
GrantInformation_xml – fundername: National Institutes of Health
  funderid: R01HL150147
– fundername: NIH HHS
  grantid: R01HL150147
GroupedDBID ---
--Z
-DZ
.GJ
0R~
1OB
1OC
24P
29M
2WC
33P
36B
3O-
4.4
53G
5GY
5RE
5VS
AAHQN
AAIPD
AAMMB
AAMNL
AANLZ
AAQQT
AASGY
AAXRX
AAYCA
AAZKR
ABCUV
ABDPE
ABEFU
ABJNI
ABLJU
ABQWH
ABUFD
ABXGK
ACAHQ
ACBEA
ACCZN
ACGFO
ACGFS
ACGOF
ACPOU
ACXBN
ACXQS
ADBBV
ADBTR
ADKYN
ADMLS
ADOZA
ADXAS
ADZMN
AEFGJ
AEGXH
AEIGN
AENEX
AEUYR
AEYWJ
AFBPY
AFFPM
AFWVQ
AGHNM
AGXDD
AGYGG
AHBTC
AIACR
AIAGR
AIDQK
AIDYY
AIQQE
AITYG
AIURR
ALMA_UNASSIGNED_HOLDINGS
ALVPJ
AMYDB
ASPBG
BFHJK
C45
CS3
DCZOG
DRFUL
DRMAN
DRSTM
DU5
EBD
EBS
EJD
EMB
EMOBN
F5P
HDBZQ
HGLYW
I-F
KBYEO
LATKE
LEEKS
LH4
LOXES
LUTES
LYRES
MEWTI
O9-
OVD
P2P
P2W
PALCI
PHY
RJQFR
RNS
ROL
SAMSI
SUPJJ
SV3
TEORI
TN5
TWZ
USG
WOHZO
WXSBR
ZGI
ZVN
ZXP
ZY4
ZZTAW
AAYXX
CITATION
CGR
CUY
CVF
ECM
EIF
NPM
7X8
ID FETCH-LOGICAL-c3184-197873d29b32ffa9eded1a0c50e36d5ac1cf67985b203fc19b108c7dd07ad6da3
IEDL.DBID 24P
ISICitedReferencesCount 0
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001609635500001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 0094-2405
2473-4209
IngestDate Mon Nov 10 00:57:17 EST 2025
Fri Nov 14 01:40:50 EST 2025
Thu Nov 27 00:46:09 EST 2025
Sat Nov 15 09:20:45 EST 2025
IsDoiOpenAccess true
IsOpenAccess true
IsPeerReviewed true
IsScholarly true
Issue 11
Keywords thoracic insufficiency syndrome
image segmentation
early onset scoliosis
deep neural networks
adolescent idiopathic scoliosis
dynamic MRI
thoraco‐abdominal organs
Language English
License Attribution
2025 The Author(s). Medical Physics published by Wiley Periodicals LLC on behalf of American Association of Physicists in Medicine.
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c3184-197873d29b32ffa9eded1a0c50e36d5ac1cf67985b203fc19b108c7dd07ad6da3
Notes ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
OpenAccessLink https://onlinelibrary.wiley.com/doi/abs/10.1002%2Fmp.70104
PMID 41206361
PQID 3270095627
PQPubID 23479
PageCount 18
ParticipantIDs proquest_miscellaneous_3270095627
pubmed_primary_41206361
crossref_primary_10_1002_mp_70104
wiley_primary_10_1002_mp_70104_MP70104
PublicationCentury 2000
PublicationDate November 2025
PublicationDateYYYYMMDD 2025-11-01
PublicationDate_xml – month: 11
  year: 2025
  text: November 2025
PublicationDecade 2020
PublicationPlace United States
PublicationPlace_xml – name: United States
PublicationTitle Medical physics (Lancaster)
PublicationTitleAlternate Med Phys
PublicationYear 2025
References 2015; 34
2007; 189
47
2022; 250
2010; 109
2017; 4
2023; 12
2020; 42
2010; 19
2024; 105
2023; 16
2007; 188
2009; 113
2016; 10154
2020; 11
2025
90
2021; 72
2022; 49
2018; 8
2014; 4
2000; 19
2023
2022; 81
2022
2021; 18
2022; 35
2019
2018
2020; 47
2017
2022 Jul; 44
2016
2014; 18
2022; 11
2007; 20
2019; 292
1992; 89
506
2018; 37
e_1_2_9_30_1
Çiçek Ö (e_1_2_9_43_1) 2016
e_1_2_9_31_1
e_1_2_9_11_1
Akhtar Y (e_1_2_9_19_1) 2023
e_1_2_9_34_1
e_1_2_9_10_1
e_1_2_9_35_1
e_1_2_9_13_1
e_1_2_9_32_1
e_1_2_9_12_1
e_1_2_9_33_1
Tong Y (e_1_2_9_45_1) 2025
e_1_2_9_15_1
e_1_2_9_38_1
e_1_2_9_14_1
e_1_2_9_39_1
e_1_2_9_17_1
e_1_2_9_36_1
e_1_2_9_37_1
e_1_2_9_18_1
e_1_2_9_41_1
e_1_2_9_42_1
e_1_2_9_20_1
e_1_2_9_40_1
e_1_2_9_22_1
e_1_2_9_21_1
e_1_2_9_24_1
e_1_2_9_23_1
e_1_2_9_44_1
e_1_2_9_8_1
e_1_2_9_7_1
Xie L (e_1_2_9_16_1) 2022
e_1_2_9_6_1
e_1_2_9_5_1
e_1_2_9_4_1
e_1_2_9_3_1
e_1_2_9_2_1
e_1_2_9_9_1
e_1_2_9_26_1
e_1_2_9_25_1
e_1_2_9_28_1
e_1_2_9_27_1
e_1_2_9_29_1
38766023 - medRxiv. 2024 May 06:2024.05.04.24306582. doi: 10.1101/2024.05.04.24306582.
References_xml – volume: 250
  start-page: 1
  year: 2022
  end-page: 12
  article-title: Dynamic MR imaging of cerebral perfusion during bicycling exercise
  publication-title: Neuroimage
– volume: 20
  start-page: 101
  year: 2007
  end-page: 118
  article-title: CAVASS: a computer‐assisted visualization and analysis software system
  publication-title: J Digit Imaging
– volume: 8
  start-page: 1
  issue: 6159
  year: 2018
  end-page: 9
  article-title: An automated skin segmentation of breasts in dynamic contrast‐enhanced magnetic resonance imaging
  publication-title: Sci Rep
– volume: 81
  year: 2022
  article-title: Object recognition in medical images via anatomy‐guided deep learning
  publication-title: Med Image Anal
– volume: 49
  start-page: 7118
  issue: 11
  year: 2022
  end-page: 7149
  article-title: Combining natural and artificial intelligence for robust automatic anatomy segmentation: application in neck and thorax auto‐contouring
  publication-title: Med Phys
– volume: 105
  issue: 4
  year: 2024
  article-title: The impact of patient specific scoliosis exercises on outcomes for adolescent idiopathic scoliosis: a systematic review with meta‐analysis
  publication-title: Arch Phys Med Rehabil
– volume: 72
  year: 2021
  article-title: OFx: a method of 4D image construction from free‐breathing non‐gated MRI slice acquisitions of the thorax via optical flux
  publication-title: Med Image Anal
– start-page: 565
  year: 2016
  end-page: 571
  article-title: V‐net: fully convolutional neural networks for volumetric medical image segmentation
– volume: 11
  start-page: 1
  issue: 66
  year: 2020
  end-page: 10
  article-title: Dynamic MRI for articulating joint evaluation on 1.5 T and 3.0 T scanners: setup, protocols, and real‐time sequences
  publication-title: Insights Imaging
– start-page: 1
  year: 2023
  end-page: 7
– volume: 19
  start-page: 91
  year: 2010
  end-page: 95
  article-title: Accuracy of free‐hand placement of thoracic pedicle screws in adolescent idiopathic scoliosis: how much of a difference does surgeon experience make?
  publication-title: Eur Spine J
– volume: 11
  start-page: 7276
  year: 2022
  article-title: Characteristics of diaphragmatic and chest wall motion in people with normal pulmonary function: a study with free‐breathing dynamic MRI
  publication-title: J Clin Med
– volume: 42
  start-page: 318
  issue: 2
  year: 2020
  end-page: 327
  article-title: Focal loss for dense object detection
  publication-title: IEEE Trans Pattern Anal Mach Intell
– volume: 89
  start-page: 5675
  year: 1992
  end-page: 5679
  article-title: Dynamic magnetic resonance imaging of human brain activity during primary sensory stimulation
  publication-title: Proc Natl Acad Sci
– volume: 44
  start-page: 3523
  issue: 7
  year: 2022 Jul
  end-page: 3542
  article-title: Image segmentation using deep learning: a survey
  publication-title: IEEE Trans Pattern Anal Mach Intell
– volume: 34
  start-page: 1993
  issue: 10
  year: 2015
  end-page: 2024
  article-title: The multimodal brain tumor image Segmentation Benchmark (BRATS)
  publication-title: IEEE Trans Med Imaging
– volume: 189
  start-page: 192
  year: 2007
  end-page: 201
  article-title: Dynamic contrast‐enhanced MRI of prostate cancer at 3 T: a study of pharmacokinetic parameters
  publication-title: Genitourinary Imaging
– volume: 506
  start-page: 313
  end-page: 335
  article-title: A review on the use of deep learning for medical images segmentation
  publication-title: Neurocomputing
– volume: 113
  start-page: 1095
  issue: 10
  year: 2009
  end-page: 1103
  article-title: Intensity standardization simplifies brain MR image segmentation
  publication-title: Comput Vis Image Underst
– volume: 292
  start-page: 206
  year: 2019
  end-page: 213
  article-title: Quantitative dynamic thoracic MRI: application to thoracic insufficiency syndrome
  publication-title: Radiology
– year: 2016
– start-page: 770
  year: 2016
  end-page: 778
  article-title: Deep residual learning for image recognition
– volume: 35
  start-page: 1
  year: 2022
  end-page: 16
  article-title: DIMENSION: dynamic MR imaging with both k‐space and spatial prior knowledge obtained via multi‐supervised network
  publication-title: NMR Biomed
– volume: 90
  start-page: 316
  end-page: 352
  article-title: Medical image segmentation using deep semantic‐based methods: a review of techniques, applications and emerging trends
  publication-title: Information Fusion
– volume: 12
  start-page: 1199
  issue: 5
  year: 2023
  article-title: Techniques and challenges of image segmentation: a review
  publication-title: Electronics
– volume: 19
  start-page: 143
  issue: 2
  year: 2000
  end-page: 150
  article-title: New variants of a method of MRI scale standardization
  publication-title: IEEE Trans Med Imaging
– volume: 10154
  start-page: 138
  year: 2016
  end-page: 149
– start-page: 3146
  year: 2019
  end-page: 3154
  article-title: Dual attention network for scene segmentation
– volume: 47
  start-page: 2352
  end-page: 9148
  article-title: Deep learning for medical image segmentation: state‐of‐the‐art advancements and challenges
  publication-title: Inf Med Unlock
– volume: 109
  start-page: 1064
  year: 2010
  end-page: 1071
  article-title: Stabilizing function of the diaphragm: dynamic MRI and synchronized spirometric assessment
  publication-title: J Appl Physiol
– volume: 16
  start-page: 447
  year: 2023
  end-page: 456
  article-title: Update on the diagnosis and management of early‐onset scoliosis
  publication-title: Curr Rev Musculoskel Med
– year: 2025
– year: 2022
– volume: 18
  start-page: 203
  year: 2021
  end-page: 211
  article-title: nnU‐Net: a self‐configuring method for deep learning‐based biomedical image segmentation
  publication-title: Nat Methods
– volume: 188
  start-page: 26
  year: 2007
  end-page: 36
  article-title: Dynamic MRI of solitary pulmonary nodules: comparison of enhancement patterns of malignant and benign small peripheral lung lesions
  publication-title: Chest Imag
– start-page: 8759
  year: 2018
  end-page: 8768
  article-title: Path aggregation network for instance segmentation
– volume: 4
  start-page: 1
  issue: 1
  year: 2017
  end-page: 9
  article-title: Automatic vertebra segmentation on dynamic magnetic resonance imaging
  publication-title: J Med Imaging
– start-page: 2980
  year: 2017
  end-page: 2988
  article-title: Mask R‐CNN
– volume: 18
  start-page: 752
  issue: 5
  year: 2014
  end-page: 771
  article-title: Body‐wide hierarchical fuzzy modeling, recognition, and delineation of anatomy in medical images
  publication-title: Med Image Anal
– volume: 4
  start-page: 147
  issue: 2
  year: 2014
  end-page: 164
  article-title: Dynamic contrast‐enhanced magnetic resonance imaging: fundamentals and application to the evaluation of the peripheral perfusion
  publication-title: Cardiovasc Diagn Ther
– volume: 18
  start-page: 969
  issue: 3
  year: 2014
  end-page: 977
  article-title: A spatiotemporal‐based scheme for efficient registration‐based segmentation of thoracic 4‐D MRI
  publication-title: IEEE J Biomed Health Inform
– volume: 37
  start-page: 1822
  issue: 8
  year: 2018
  end-page: 1834
  article-title: Automatic multi‐organ segmentation on abdominal CT with dense V‐Networks
  publication-title: IEEE Trans Med Imaging
– volume: 47
  start-page: 2986
  issue: 7
  year: 2020
  end-page: 2999
  article-title: ABCNet: a new efficient 3D dense‐structure network for segmentation and analysis of body tissue composition on body‐torso‐wide CT images
  publication-title: Med Phys
– start-page: 2261
  year: 2017
  end-page: 2269
  article-title: Densely connected convolutional networks
– ident: e_1_2_9_22_1
  doi: 10.1002/mp.14141
– ident: e_1_2_9_15_1
  doi: 10.2214/AJR.05.1446
– ident: e_1_2_9_31_1
  doi: 10.1038/s41592‐020‐01008‐z
– volume-title: A normative database of free‐breathing pediatric thoracic 4D dynamic MRI images
  year: 2025
  ident: e_1_2_9_45_1
– ident: e_1_2_9_36_1
  doi: 10.1007/s10278‐007‐9060‐5
– ident: e_1_2_9_20_1
  doi: 10.1002/nbm.4131
– ident: e_1_2_9_5_1
  doi: 10.1007/s12178‐023‐09848‐w
– ident: e_1_2_9_27_1
  doi: 10.1016/j.neucom.2022.07.070
– ident: e_1_2_9_37_1
  doi: 10.1109/CVPR.2016.90
– volume-title: Medical Image Computing and Computer‐Assisted Intervention—MICCAI
  year: 2016
  ident: e_1_2_9_43_1
– ident: e_1_2_9_34_1
  doi: 10.1016/j.cviu.2009.06.003
– ident: e_1_2_9_6_1
  doi: 10.3978/j.issn.2223‐3652.2014.03.01
– ident: e_1_2_9_44_1
  doi: 10.1002/mp.15854
– ident: e_1_2_9_26_1
  doi: 10.1109/TPAMI.2021.3059968
– ident: e_1_2_9_39_1
  doi: 10.1109/TPAMI.2018.2858826
– ident: e_1_2_9_11_1
  doi: 10.1073/pnas.89.12.5675
– ident: e_1_2_9_42_1
  doi: 10.1109/3DV.2016.79
– ident: e_1_2_9_8_1
  doi: 10.1117/1.JMI.4.1.014504
– ident: e_1_2_9_7_1
  doi: 10.1186/s13244‐020‐00868‐5
– ident: e_1_2_9_14_1
  doi: 10.1152/japplphysiol.01216.2009
– ident: e_1_2_9_24_1
  doi: 10.1109/CVPR.2019.00326
– ident: e_1_2_9_12_1
  doi: 10.1038/s41598‐018‐22941‐2
– start-page: 1
  volume-title: SPIE Medical Imaging
  year: 2023
  ident: e_1_2_9_19_1
– ident: e_1_2_9_23_1
  doi: 10.1109/CVPR.2018.00913
– ident: e_1_2_9_28_1
  doi: 10.3390/electronics12051199
– ident: e_1_2_9_33_1
  doi: 10.1109/42.836373
– ident: e_1_2_9_3_1
  doi: 10.1007/s00586‐009‐1183‐6
– ident: e_1_2_9_13_1
  doi: 10.3390/jcm11247276
– ident: e_1_2_9_17_1
  doi: 10.1109/JBHI.2013.2282183
– ident: e_1_2_9_21_1
  doi: 10.1016/j.media.2022.102527
– ident: e_1_2_9_25_1
  doi: 10.1016/j.inffus.2022.09.031
– ident: e_1_2_9_32_1
  doi: 10.1109/TMI.2014.2377694
– ident: e_1_2_9_4_1
  doi: 10.1016/j.apmr.2024.02.692
– ident: e_1_2_9_38_1
  doi: 10.1109/CVPR.2017.243
– ident: e_1_2_9_40_1
  doi: 10.1007/978-3-319-55524-9_14
– ident: e_1_2_9_18_1
  doi: 10.1016/j.media.2021.102088
– ident: e_1_2_9_10_1
  doi: 10.2214/AJR.06.1329
– ident: e_1_2_9_9_1
  doi: 10.1016/j.neuroimage.2022.118961
– ident: e_1_2_9_2_1
  doi: 10.1148/radiol.2019181731
– ident: e_1_2_9_41_1
  doi: 10.1109/TMI.2018.2806309
– ident: e_1_2_9_30_1
  doi: 10.1109/ICCV.2017.322
– ident: e_1_2_9_29_1
  doi: 10.1016/j.imu.2024.101504
– volume-title: SPIE Medical Imaging
  year: 2022
  ident: e_1_2_9_16_1
– ident: e_1_2_9_35_1
  doi: 10.1016/j.media.2014.04.003
– reference: 38766023 - medRxiv. 2024 May 06:2024.05.04.24306582. doi: 10.1101/2024.05.04.24306582.
SSID ssj0006350
Score 2.4807134
Snippet Purpose Dynamic magnetic resonance imaging (dMRI) is a practical imaging modality for capturing information about regional thoracic‐abdominal components and...
Dynamic magnetic resonance imaging (dMRI) is a practical imaging modality for capturing information about regional thoracic-abdominal components and their...
SourceID proquest
pubmed
crossref
wiley
SourceType Aggregation Database
Index Database
Publisher
StartPage e70104
SubjectTerms Abdomen - diagnostic imaging
Adolescent
adolescent idiopathic scoliosis
Automation
Child
Child, Preschool
deep neural networks
dynamic MRI
early onset scoliosis
Humans
Image Processing, Computer-Assisted - methods
image segmentation
Magnetic Resonance Imaging
thoracic insufficiency syndrome
thoraco‐abdominal organs
Thorax - diagnostic imaging
Title Auto‐segmentation of thoraco‐abdominal organs in pediatric dynamic MRI
URI https://onlinelibrary.wiley.com/doi/abs/10.1002%2Fmp.70104
https://www.ncbi.nlm.nih.gov/pubmed/41206361
https://www.proquest.com/docview/3270095627
Volume 52
WOSCitedRecordID wos001609635500001&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: PRVWIB
  databaseName: Wiley Online Library Full Collection 2020
  customDbUrl:
  eissn: 2473-4209
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0006350
  issn: 0094-2405
  databaseCode: DRFUL
  dateStart: 19970101
  isFulltext: true
  titleUrlDefault: https://onlinelibrary.wiley.com
  providerName: Wiley-Blackwell
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV1JT9wwFH4qW9ULe-mwyUgVt0Bsx3FyRMCoIAaNUKnmFnkFDpOMyAzn_oT-xv6S2k4yCFVIlXpJDrGtyG_383sfwFeGpTMcNo00U_60KqeRtLGOpDM-NhbSktCs-scNv73NRqN82N6q9LUwTX-I-YGbl4ygr72AC1mfvjYNHU9OuA8mFmAJY5p52AaSDOda2BnSpvwkT3wGgXWNZ2Ny2s18a4r-8i_fuqvB3vTX_udP12G19TLRWcMWG_DBlJvwcdDm0TdhJVz8VPUWXJ_NptXvn79q8zBuy5BKVFk0fXS8ofwXIXUVkL9QQICq0VOJJh3AB9INoj0a3F1tw33_8vv5t6jFV4iUk-Qkwi6C5FSTXFJirciNNhqLWLHY0FQzobCyPknDJImpVTiXOM4U1zrmQqda0M-wWFal-QIoTUxmmNJCJS7W5qkUWZ5yyjRzCiSzeQ-Ouq0uJk0bjaJpmEyK8aQI2-PGdDQoHI_7xIUoTTWrC-qz4y6QI7wHOw1x5qskmDhap7gHx4EG7y5fDIbhvfuvA_fgE_Fgv6HwcB8Wp88zcwDL6mX6VD8fBm5zTz7KDmHp4q5_f_MHJXXZMw
linkProvider Wiley-Blackwell
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV1JS8NAFH5oXS_uS11HEG-xySSTBU8iStW2FFHxFmZVD02KaT37E_yN_hJnJk1FRBA85ZDJEObt7817H8Ah8Zg2HCp0BOEmW5X4DlOucJg2PsqlTGE7rPq-FXU68cND0p2Ak6oXppwPMU64Gcmw-toIuElIN76mhvb6x5GJJiZhKtBGxjA5DrpjNawtadl_kgSmhECqybMublRffrdFPxzM7_6qNTgXi__61SVYGPmZ6LRkjGWYkNkKzLZHlfQVmLFXP3mxClenw0H-8fZeyMfeqBEpQ7lCgyfNHdy8oUzkFvsLWQyoAj1nqF9BfCBRYtqj9s3lGtxdnN-eNZ0RwoLDtSwHjqdjyMgXOGE-VoomUkjhUZcTV_qhIJR7XJkyDWHY9RX3Eua5MY-EcCMqQkH9dahleSY3AYWBjCXhgvJAR9tRyGichJFPBNEqJFZJHQ6qs0775SCNtByZjNNeP7XHo9dUREg1l5vSBc1kPixS39THdSiHozpslNQZ7xJ4WBM79OpwZInw6_Zpu2ufW39duA9zzdt2K21ddq63YR4b6F_bhrgDtcHLUO7CNH8dPBcve5b1PgExo9qW
linkToPdf http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV1JT-MwFH5iygziAgxrWT3SiFto4sRZxAkBFUtbVWgYcYu8AocmEWk58xP4jfwSbCcpQggJiVMOcazIb_Pz8_s-gL_EYzpwqNARhJvTqsR3mHKFw3TwUS5lCluw6v-9aDCIb26S4QwcNr0wFT7E9MDNWIb118bAZSFU5w01dFQcRCab-AGzgeGQacHsyVX3ujd1xDqWVh0oSWCKCKTBnnVxp_n2fTT6sMV8v2O1Iae7-K2fXYKFeqeJjirV-A0zMluGuX5dS1-GX_byJy9X4OJoMs5fnp5LeTuqW5EylCs0vtP6wc0bykRu2b-QZYEq0X2GiobkA4mK1R71r85X4bp7-u_4zKk5FhyurTlwPJ1FRr7ACfOxUjSRQgqPupy40g8FodzjyhRqCMOur7iXMM-NeSSEG1ERCuqvQSvLM7kBKAxkLAkXlAc6345CRuMkjHwiiHYisUra8KdZ67SooDTSCjQZp6MitcujxzRCSLWem-IFzWQ-KVPfVMh1MoejNqxX0pnOEnhYCzv02rBvhfDp9Gl_aJ-bXx24B3PDk27aOx9cbsE8Nty_tg9xG1rjh4ncgZ_8cXxfPuzWuvcK-aTbrA
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=Auto%E2%80%90segmentation+of+thoraco%E2%80%90abdominal+organs+in+pediatric+dynamic+MRI&rft.jtitle=Medical+physics+%28Lancaster%29&rft.au=Akhtar%2C+Yusuf&rft.au=Udupa%2C+Jayaram+K.&rft.au=Tong%2C+Yubing&rft.au=Liu%2C+Tiange&rft.date=2025-11-01&rft.issn=0094-2405&rft.eissn=2473-4209&rft.volume=52&rft.issue=11&rft.epage=n%2Fa&rft_id=info:doi/10.1002%2Fmp.70104&rft.externalDBID=10.1002%252Fmp.70104&rft.externalDocID=MP70104
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0094-2405&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0094-2405&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0094-2405&client=summon