Advanced

An Efficient Optimization Framework for Multi-Region Segmentation based on Lagrangian Duality

Ulén, Johannes LU ; Strandmark, Petter LU and Kahl, Fredrik LU (2013) In IEEE Transactions on Medical Imaging 32(2). p.178-188
Abstract
We introduce a multi-region model for simultaneous segmentation of medical images. In contrast to many other models, geometric constraints such as inclusion and exclusion between the regions are enforced, which makes it possible to correctly segment different regions even if the intensity distributions are identical. We efficiently optimize the model using a combination of graph cuts and Lagrangian duality which is faster and more memory efficient than current state of the art. As the method is based on global optimization techniques, the resulting segmentations are independent of initialization. We apply our framework to the segmentation of the left and right ventricles, myocardium and the left ventricular papillary muscles in MRI and to... (More)
We introduce a multi-region model for simultaneous segmentation of medical images. In contrast to many other models, geometric constraints such as inclusion and exclusion between the regions are enforced, which makes it possible to correctly segment different regions even if the intensity distributions are identical. We efficiently optimize the model using a combination of graph cuts and Lagrangian duality which is faster and more memory efficient than current state of the art. As the method is based on global optimization techniques, the resulting segmentations are independent of initialization. We apply our framework to the segmentation of the left and right ventricles, myocardium and the left ventricular papillary muscles in MRI and to lung segmentation in full-body X-ray CT. We evaluate our approach on a publicly available benchmark with competitive results. (Less)
Please use this url to cite or link to this publication:
author
; and
organization
publishing date
type
Contribution to journal
publication status
published
subject
keywords
Cardiac segmentation, discrete optimization, image segmentation, lung segmentation
in
IEEE Transactions on Medical Imaging
volume
32
issue
2
pages
178 - 188
publisher
IEEE - Institute of Electrical and Electronics Engineers Inc.
external identifiers
  • wos:000314367100004
  • scopus:84873282421
  • pmid:22987510
ISSN
1558-254X
DOI
10.1109/TMI.2012.2218117
language
English
LU publication?
yes
id
76b55772-9953-40c8-97ad-2a61f85d7972 (old id 3459310)
alternative location
http://www.maths.lth.se/vision/publdb/reports/pdf/ulen-strandmark-etal-itmi-12.pdf
http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=6298014
date added to LUP
2016-04-01 11:07:25
date last changed
2020-02-12 03:14:20
@article{76b55772-9953-40c8-97ad-2a61f85d7972,
  abstract     = {We introduce a multi-region model for simultaneous segmentation of medical images. In contrast to many other models, geometric constraints such as inclusion and exclusion between the regions are enforced, which makes it possible to correctly segment different regions even if the intensity distributions are identical. We efficiently optimize the model using a combination of graph cuts and Lagrangian duality which is faster and more memory efficient than current state of the art. As the method is based on global optimization techniques, the resulting segmentations are independent of initialization. We apply our framework to the segmentation of the left and right ventricles, myocardium and the left ventricular papillary muscles in MRI and to lung segmentation in full-body X-ray CT. We evaluate our approach on a publicly available benchmark with competitive results.},
  author       = {Ulén, Johannes and Strandmark, Petter and Kahl, Fredrik},
  issn         = {1558-254X},
  language     = {eng},
  number       = {2},
  pages        = {178--188},
  publisher    = {IEEE - Institute of Electrical and Electronics Engineers Inc.},
  series       = {IEEE Transactions on Medical Imaging},
  title        = {An Efficient Optimization Framework for Multi-Region Segmentation based on Lagrangian Duality},
  url          = {http://dx.doi.org/10.1109/TMI.2012.2218117},
  doi          = {10.1109/TMI.2012.2218117},
  volume       = {32},
  year         = {2013},
}