Eintrag weiter verarbeiten
Low‐complexity atlas‐based prostate segmentation by combining global, regional, and local metrics
Gespeichert in:
Zeitschriftentitel: | Medical Physics |
---|---|
Personen und Körperschaften: | , |
In: | Medical Physics, 41, 2014, 4 |
Format: | E-Article |
Sprache: | Englisch |
veröffentlicht: |
Wiley
|
Schlagwörter: |
author_facet |
Xie, Qiuliang Ruan, Dan Xie, Qiuliang Ruan, Dan |
---|---|
author |
Xie, Qiuliang Ruan, Dan |
spellingShingle |
Xie, Qiuliang Ruan, Dan Medical Physics Low‐complexity atlas‐based prostate segmentation by combining global, regional, and local metrics General Medicine |
author_sort |
xie, qiuliang |
spelling |
Xie, Qiuliang Ruan, Dan 0094-2405 2473-4209 Wiley General Medicine http://dx.doi.org/10.1118/1.4867855 <jats:sec><jats:title><jats:bold>Purpose:</jats:bold></jats:title><jats:p>To improve the efficiency of atlas‐based segmentation without compromising accuracy, and to demonstrate the validity of the proposed method on MRI‐based prostate segmentation application.</jats:p></jats:sec><jats:sec><jats:title><jats:bold>Methods:</jats:bold></jats:title><jats:p>Accurate and efficient automatic structure segmentation is an important task in medical image processing. Atlas‐based methods, as the state‐of‐the‐art, provide good segmentation at the cost of a large number of computationally intensive nonrigid registrations, for anatomical sites/structures that are subject to deformation. In this study, the authors propose to utilize a combination of global, regional, and local metrics to improve the accuracy yet significantly reduce the number of required nonrigid registrations. The authors first perform an affine registration to minimize the global mean squared error (gMSE) to coarsely align each atlas image to the target. Subsequently, a<jats:italic>target‐specific</jats:italic> regional MSE (rMSE), demonstrated to be a good surrogate for dice similarity coefficient (DSC), is used to select a relevant subset from the training atlas. Only within this subset are nonrigid registrations performed between the training images and the target image, to minimize a weighted combination of gMSE and rMSE. Finally, structure labels are propagated from the selected training samples to the target via the estimated deformation fields, and label fusion is performed based on a weighted combination of rMSE and local MSE (lMSE) discrepancy, with proper total‐variation‐based spatial regularization.</jats:p></jats:sec><jats:sec><jats:title><jats:bold>Results:</jats:bold></jats:title><jats:p>The proposed method was applied to a public database of 30 prostate MR images with expert‐segmented structures. The authors’ method, utilizing only eight nonrigid registrations, achieved a performance with a median/mean DSC of over 0.87/0.86, outperforming the state‐of‐the‐art full‐fledged atlas‐based segmentation approach of which the median/mean DSC was 0.84/0.82 when applying to their data set.</jats:p></jats:sec><jats:sec><jats:title><jats:bold>Conclusions:</jats:bold></jats:title><jats:p>The proposed method requires a fixed number of nonrigid registrations, independent of atlas size, providing desirable scalability especially important for a large or growing atlas. When applied to prostate segmentation, the method achieved better performance to the state‐of‐the‐art atlas‐based approaches, with significant improvement in computation efficiency. The proposed rationale of utilizing jointly global, regional, and local metrics, based on the information characteristic and surrogate behavior for registration and fusion subtasks, can be extended naturally to similarity metrics beyond MSE, such as correlation or mutual information types.</jats:p></jats:sec> Low‐complexity atlas‐based prostate segmentation by combining global, regional, and local metrics Medical Physics |
doi_str_mv |
10.1118/1.4867855 |
facet_avail |
Online |
format |
ElectronicArticle |
fullrecord |
blob:ai-49-aHR0cDovL2R4LmRvaS5vcmcvMTAuMTExOC8xLjQ4Njc4NTU |
id |
ai-49-aHR0cDovL2R4LmRvaS5vcmcvMTAuMTExOC8xLjQ4Njc4NTU |
institution |
DE-D161 DE-Gla1 DE-Zi4 DE-15 DE-Pl11 DE-Rs1 DE-105 DE-14 DE-Ch1 DE-L229 DE-D275 DE-Bn3 DE-Brt1 |
imprint |
Wiley, 2014 |
imprint_str_mv |
Wiley, 2014 |
issn |
0094-2405 2473-4209 |
issn_str_mv |
0094-2405 2473-4209 |
language |
English |
mega_collection |
Wiley (CrossRef) |
match_str |
xie2014lowcomplexityatlasbasedprostatesegmentationbycombiningglobalregionalandlocalmetrics |
publishDateSort |
2014 |
publisher |
Wiley |
recordtype |
ai |
record_format |
ai |
series |
Medical Physics |
source_id |
49 |
title |
Low‐complexity atlas‐based prostate segmentation by combining global, regional, and local metrics |
title_unstemmed |
Low‐complexity atlas‐based prostate segmentation by combining global, regional, and local metrics |
title_full |
Low‐complexity atlas‐based prostate segmentation by combining global, regional, and local metrics |
title_fullStr |
Low‐complexity atlas‐based prostate segmentation by combining global, regional, and local metrics |
title_full_unstemmed |
Low‐complexity atlas‐based prostate segmentation by combining global, regional, and local metrics |
title_short |
Low‐complexity atlas‐based prostate segmentation by combining global, regional, and local metrics |
title_sort |
low‐complexity atlas‐based prostate segmentation by combining global, regional, and local metrics |
topic |
General Medicine |
url |
http://dx.doi.org/10.1118/1.4867855 |
publishDate |
2014 |
physical |
|
description |
<jats:sec><jats:title><jats:bold>Purpose:</jats:bold></jats:title><jats:p>To improve the efficiency of atlas‐based segmentation without compromising accuracy, and to demonstrate the validity of the proposed method on MRI‐based prostate segmentation application.</jats:p></jats:sec><jats:sec><jats:title><jats:bold>Methods:</jats:bold></jats:title><jats:p>Accurate and efficient automatic structure segmentation is an important task in medical image processing. Atlas‐based methods, as the state‐of‐the‐art, provide good segmentation at the cost of a large number of computationally intensive nonrigid registrations, for anatomical sites/structures that are subject to deformation. In this study, the authors propose to utilize a combination of global, regional, and local metrics to improve the accuracy yet significantly reduce the number of required nonrigid registrations. The authors first perform an affine registration to minimize the global mean squared error (gMSE) to coarsely align each atlas image to the target. Subsequently, a<jats:italic>target‐specific</jats:italic> regional MSE (rMSE), demonstrated to be a good surrogate for dice similarity coefficient (DSC), is used to select a relevant subset from the training atlas. Only within this subset are nonrigid registrations performed between the training images and the target image, to minimize a weighted combination of gMSE and rMSE. Finally, structure labels are propagated from the selected training samples to the target via the estimated deformation fields, and label fusion is performed based on a weighted combination of rMSE and local MSE (lMSE) discrepancy, with proper total‐variation‐based spatial regularization.</jats:p></jats:sec><jats:sec><jats:title><jats:bold>Results:</jats:bold></jats:title><jats:p>The proposed method was applied to a public database of 30 prostate MR images with expert‐segmented structures. The authors’ method, utilizing only eight nonrigid registrations, achieved a performance with a median/mean DSC of over 0.87/0.86, outperforming the state‐of‐the‐art full‐fledged atlas‐based segmentation approach of which the median/mean DSC was 0.84/0.82 when applying to their data set.</jats:p></jats:sec><jats:sec><jats:title><jats:bold>Conclusions:</jats:bold></jats:title><jats:p>The proposed method requires a fixed number of nonrigid registrations, independent of atlas size, providing desirable scalability especially important for a large or growing atlas. When applied to prostate segmentation, the method achieved better performance to the state‐of‐the‐art atlas‐based approaches, with significant improvement in computation efficiency. The proposed rationale of utilizing jointly global, regional, and local metrics, based on the information characteristic and surrogate behavior for registration and fusion subtasks, can be extended naturally to similarity metrics beyond MSE, such as correlation or mutual information types.</jats:p></jats:sec> |
container_issue |
4 |
container_start_page |
0 |
container_title |
Medical Physics |
container_volume |
41 |
format_de105 |
Article, E-Article |
format_de14 |
Article, E-Article |
format_de15 |
Article, E-Article |
format_de520 |
Article, E-Article |
format_de540 |
Article, E-Article |
format_dech1 |
Article, E-Article |
format_ded117 |
Article, E-Article |
format_degla1 |
E-Article |
format_del152 |
Buch |
format_del189 |
Article, E-Article |
format_dezi4 |
Article |
format_dezwi2 |
Article, E-Article |
format_finc |
Article, E-Article |
format_nrw |
Article, E-Article |
_version_ |
1792333034831216648 |
geogr_code |
not assigned |
last_indexed |
2024-03-01T14:06:21.487Z |
geogr_code_person |
not assigned |
openURL |
url_ver=Z39.88-2004&ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fvufind.svn.sourceforge.net%3Agenerator&rft.title=Low%E2%80%90complexity+atlas%E2%80%90based+prostate+segmentation+by+combining+global%2C+regional%2C+and+local+metrics&rft.date=2014-04-01&genre=article&issn=2473-4209&volume=41&issue=4&jtitle=Medical+Physics&atitle=Low%E2%80%90complexity+atlas%E2%80%90based+prostate+segmentation+by+combining+global%2C+regional%2C+and+local+metrics&aulast=Ruan&aufirst=Dan&rft_id=info%3Adoi%2F10.1118%2F1.4867855&rft.language%5B0%5D=eng |
SOLR | |
_version_ | 1792333034831216648 |
author | Xie, Qiuliang, Ruan, Dan |
author_facet | Xie, Qiuliang, Ruan, Dan, Xie, Qiuliang, Ruan, Dan |
author_sort | xie, qiuliang |
container_issue | 4 |
container_start_page | 0 |
container_title | Medical Physics |
container_volume | 41 |
description | <jats:sec><jats:title><jats:bold>Purpose:</jats:bold></jats:title><jats:p>To improve the efficiency of atlas‐based segmentation without compromising accuracy, and to demonstrate the validity of the proposed method on MRI‐based prostate segmentation application.</jats:p></jats:sec><jats:sec><jats:title><jats:bold>Methods:</jats:bold></jats:title><jats:p>Accurate and efficient automatic structure segmentation is an important task in medical image processing. Atlas‐based methods, as the state‐of‐the‐art, provide good segmentation at the cost of a large number of computationally intensive nonrigid registrations, for anatomical sites/structures that are subject to deformation. In this study, the authors propose to utilize a combination of global, regional, and local metrics to improve the accuracy yet significantly reduce the number of required nonrigid registrations. The authors first perform an affine registration to minimize the global mean squared error (gMSE) to coarsely align each atlas image to the target. Subsequently, a<jats:italic>target‐specific</jats:italic> regional MSE (rMSE), demonstrated to be a good surrogate for dice similarity coefficient (DSC), is used to select a relevant subset from the training atlas. Only within this subset are nonrigid registrations performed between the training images and the target image, to minimize a weighted combination of gMSE and rMSE. Finally, structure labels are propagated from the selected training samples to the target via the estimated deformation fields, and label fusion is performed based on a weighted combination of rMSE and local MSE (lMSE) discrepancy, with proper total‐variation‐based spatial regularization.</jats:p></jats:sec><jats:sec><jats:title><jats:bold>Results:</jats:bold></jats:title><jats:p>The proposed method was applied to a public database of 30 prostate MR images with expert‐segmented structures. The authors’ method, utilizing only eight nonrigid registrations, achieved a performance with a median/mean DSC of over 0.87/0.86, outperforming the state‐of‐the‐art full‐fledged atlas‐based segmentation approach of which the median/mean DSC was 0.84/0.82 when applying to their data set.</jats:p></jats:sec><jats:sec><jats:title><jats:bold>Conclusions:</jats:bold></jats:title><jats:p>The proposed method requires a fixed number of nonrigid registrations, independent of atlas size, providing desirable scalability especially important for a large or growing atlas. When applied to prostate segmentation, the method achieved better performance to the state‐of‐the‐art atlas‐based approaches, with significant improvement in computation efficiency. The proposed rationale of utilizing jointly global, regional, and local metrics, based on the information characteristic and surrogate behavior for registration and fusion subtasks, can be extended naturally to similarity metrics beyond MSE, such as correlation or mutual information types.</jats:p></jats:sec> |
doi_str_mv | 10.1118/1.4867855 |
facet_avail | Online |
format | ElectronicArticle |
format_de105 | Article, E-Article |
format_de14 | Article, E-Article |
format_de15 | Article, E-Article |
format_de520 | Article, E-Article |
format_de540 | Article, E-Article |
format_dech1 | Article, E-Article |
format_ded117 | Article, E-Article |
format_degla1 | E-Article |
format_del152 | Buch |
format_del189 | Article, E-Article |
format_dezi4 | Article |
format_dezwi2 | Article, E-Article |
format_finc | Article, E-Article |
format_nrw | Article, E-Article |
geogr_code | not assigned |
geogr_code_person | not assigned |
id | ai-49-aHR0cDovL2R4LmRvaS5vcmcvMTAuMTExOC8xLjQ4Njc4NTU |
imprint | Wiley, 2014 |
imprint_str_mv | Wiley, 2014 |
institution | DE-D161, DE-Gla1, DE-Zi4, DE-15, DE-Pl11, DE-Rs1, DE-105, DE-14, DE-Ch1, DE-L229, DE-D275, DE-Bn3, DE-Brt1 |
issn | 0094-2405, 2473-4209 |
issn_str_mv | 0094-2405, 2473-4209 |
language | English |
last_indexed | 2024-03-01T14:06:21.487Z |
match_str | xie2014lowcomplexityatlasbasedprostatesegmentationbycombiningglobalregionalandlocalmetrics |
mega_collection | Wiley (CrossRef) |
physical | |
publishDate | 2014 |
publishDateSort | 2014 |
publisher | Wiley |
record_format | ai |
recordtype | ai |
series | Medical Physics |
source_id | 49 |
spelling | Xie, Qiuliang Ruan, Dan 0094-2405 2473-4209 Wiley General Medicine http://dx.doi.org/10.1118/1.4867855 <jats:sec><jats:title><jats:bold>Purpose:</jats:bold></jats:title><jats:p>To improve the efficiency of atlas‐based segmentation without compromising accuracy, and to demonstrate the validity of the proposed method on MRI‐based prostate segmentation application.</jats:p></jats:sec><jats:sec><jats:title><jats:bold>Methods:</jats:bold></jats:title><jats:p>Accurate and efficient automatic structure segmentation is an important task in medical image processing. Atlas‐based methods, as the state‐of‐the‐art, provide good segmentation at the cost of a large number of computationally intensive nonrigid registrations, for anatomical sites/structures that are subject to deformation. In this study, the authors propose to utilize a combination of global, regional, and local metrics to improve the accuracy yet significantly reduce the number of required nonrigid registrations. The authors first perform an affine registration to minimize the global mean squared error (gMSE) to coarsely align each atlas image to the target. Subsequently, a<jats:italic>target‐specific</jats:italic> regional MSE (rMSE), demonstrated to be a good surrogate for dice similarity coefficient (DSC), is used to select a relevant subset from the training atlas. Only within this subset are nonrigid registrations performed between the training images and the target image, to minimize a weighted combination of gMSE and rMSE. Finally, structure labels are propagated from the selected training samples to the target via the estimated deformation fields, and label fusion is performed based on a weighted combination of rMSE and local MSE (lMSE) discrepancy, with proper total‐variation‐based spatial regularization.</jats:p></jats:sec><jats:sec><jats:title><jats:bold>Results:</jats:bold></jats:title><jats:p>The proposed method was applied to a public database of 30 prostate MR images with expert‐segmented structures. The authors’ method, utilizing only eight nonrigid registrations, achieved a performance with a median/mean DSC of over 0.87/0.86, outperforming the state‐of‐the‐art full‐fledged atlas‐based segmentation approach of which the median/mean DSC was 0.84/0.82 when applying to their data set.</jats:p></jats:sec><jats:sec><jats:title><jats:bold>Conclusions:</jats:bold></jats:title><jats:p>The proposed method requires a fixed number of nonrigid registrations, independent of atlas size, providing desirable scalability especially important for a large or growing atlas. When applied to prostate segmentation, the method achieved better performance to the state‐of‐the‐art atlas‐based approaches, with significant improvement in computation efficiency. The proposed rationale of utilizing jointly global, regional, and local metrics, based on the information characteristic and surrogate behavior for registration and fusion subtasks, can be extended naturally to similarity metrics beyond MSE, such as correlation or mutual information types.</jats:p></jats:sec> Low‐complexity atlas‐based prostate segmentation by combining global, regional, and local metrics Medical Physics |
spellingShingle | Xie, Qiuliang, Ruan, Dan, Medical Physics, Low‐complexity atlas‐based prostate segmentation by combining global, regional, and local metrics, General Medicine |
title | Low‐complexity atlas‐based prostate segmentation by combining global, regional, and local metrics |
title_full | Low‐complexity atlas‐based prostate segmentation by combining global, regional, and local metrics |
title_fullStr | Low‐complexity atlas‐based prostate segmentation by combining global, regional, and local metrics |
title_full_unstemmed | Low‐complexity atlas‐based prostate segmentation by combining global, regional, and local metrics |
title_short | Low‐complexity atlas‐based prostate segmentation by combining global, regional, and local metrics |
title_sort | low‐complexity atlas‐based prostate segmentation by combining global, regional, and local metrics |
title_unstemmed | Low‐complexity atlas‐based prostate segmentation by combining global, regional, and local metrics |
topic | General Medicine |
url | http://dx.doi.org/10.1118/1.4867855 |