Small organ segmentation in whole-body mri using a two-stage fcn and weighting schemes

Vanya V. Valindria; Ioannis Lavdas; Juan Cerrolaza; Eric O. Aboagye; Andrea G. Rockall; Daniel Rueckert; Ben Glocker

doi:10.1007/978-3-030-00919-9_40

Small organ segmentation in whole-body mri using a two-stage fcn and weighting schemes

Vanya V. Valindria, Ioannis Lavdas, Juan Cerrolaza, Eric O. Aboagye, Andrea G. Rockall, Daniel Rueckert, Ben Glocker

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

16 Citations (Scopus)

Abstract

Accurate and robust segmentation of small organs in whole-body MRI is difficult due to anatomical variation and class imbalance. Recent deep network based approaches have demonstrated promising performance on abdominal multi-organ segmentations. However, the performance on small organs is still suboptimal as these occupy only small regions of the whole-body volumes with unclear boundaries and variable shapes. A coarse-to-fine, hierarchical strategy is a common approach to alleviate this problem, however, this might miss useful contextual information. We propose a two-stage approach with weighting schemes based on auto-context and spatial atlas priors. Our experiments show that the proposed approach can boost the segmentation accuracy of multiple small organs in whole-body MRI scans.

Original language	English
Title of host publication	Machine Learning in Medical Imaging - 9th International Workshop, MLMI 2018, Held in Conjunction with MICCAI 2018, Proceedings
Editors	Mingxia Liu, Heung-Il Suk, Yinghuan Shi
Publisher	Springer
Pages	346-354
Number of pages	9
ISBN (Print)	9783030009182
DOIs	https://doi.org/10.1007/978-3-030-00919-9_40
Publication status	Published - 2018
Externally published	Yes
Event	International Workshop on Machine Learning in Medical Imaging (MLMI) 2018 - Granada Conference Centre, Granada, Spain Duration: 16 Sept 2018 → 16 Sept 2018 Conference number: 9th http://mlmi2018.web.unc.edu/ https://link.springer.com/book/10.1007/978-3-030-00919-9 (Proceedings)

Publication series

Name	Lecture Notes in Computer Science
Publisher	Springer
Volume	11046
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Workshop

Workshop	International Workshop on Machine Learning in Medical Imaging (MLMI) 2018
Abbreviated title	MLMI 2018
Country/Territory	Spain
City	Granada
Period	16/09/18 → 16/09/18
Internet address	http://mlmi2018.web.unc.edu/ https://link.springer.com/book/10.1007/978-3-030-00919-9 (Proceedings)

Access to Document

10.1007/978-3-030-00919-9_40

Cite this

Valindria, V. V., Lavdas, I., Cerrolaza, J., Aboagye, E. O., Rockall, A. G., Rueckert, D., & Glocker, B. (2018). Small organ segmentation in whole-body mri using a two-stage fcn and weighting schemes. In M. Liu, H.-I. Suk, & Y. Shi (Eds.), Machine Learning in Medical Imaging - 9th International Workshop, MLMI 2018, Held in Conjunction with MICCAI 2018, Proceedings (pp. 346-354). (Lecture Notes in Computer Science; Vol. 11046). Springer. https://doi.org/10.1007/978-3-030-00919-9_40

Valindria, Vanya V. ; Lavdas, Ioannis ; Cerrolaza, Juan et al. / Small organ segmentation in whole-body mri using a two-stage fcn and weighting schemes. Machine Learning in Medical Imaging - 9th International Workshop, MLMI 2018, Held in Conjunction with MICCAI 2018, Proceedings. editor / Mingxia Liu ; Heung-Il Suk ; Yinghuan Shi. Springer, 2018. pp. 346-354 (Lecture Notes in Computer Science).

@inproceedings{8010676a81f54af0bfeae0bf8ff66301,

title = "Small organ segmentation in whole-body mri using a two-stage fcn and weighting schemes",

abstract = "Accurate and robust segmentation of small organs in whole-body MRI is difficult due to anatomical variation and class imbalance. Recent deep network based approaches have demonstrated promising performance on abdominal multi-organ segmentations. However, the performance on small organs is still suboptimal as these occupy only small regions of the whole-body volumes with unclear boundaries and variable shapes. A coarse-to-fine, hierarchical strategy is a common approach to alleviate this problem, however, this might miss useful contextual information. We propose a two-stage approach with weighting schemes based on auto-context and spatial atlas priors. Our experiments show that the proposed approach can boost the segmentation accuracy of multiple small organs in whole-body MRI scans.",

author = "Valindria, \{Vanya V.\} and Ioannis Lavdas and Juan Cerrolaza and Aboagye, \{Eric O.\} and Rockall, \{Andrea G.\} and Daniel Rueckert and Ben Glocker",

note = "Publisher Copyright: {\textcopyright} Springer Nature Switzerland AG 2018.; International Workshop on Machine Learning in Medical Imaging (MLMI) 2018, MLMI 2018 ; Conference date: 16-09-2018 Through 16-09-2018",

year = "2018",

doi = "10.1007/978-3-030-00919-9\_40",

language = "English",

isbn = "9783030009182",

series = "Lecture Notes in Computer Science",

publisher = "Springer",

pages = "346--354",

editor = "Mingxia Liu and Heung-Il Suk and Yinghuan Shi",

booktitle = "Machine Learning in Medical Imaging - 9th International Workshop, MLMI 2018, Held in Conjunction with MICCAI 2018, Proceedings",

address = "Switzerland",

url = "http://mlmi2018.web.unc.edu/, https://link.springer.com/book/10.1007/978-3-030-00919-9",

}

Valindria, VV, Lavdas, I, Cerrolaza, J, Aboagye, EO, Rockall, AG, Rueckert, D & Glocker, B 2018, Small organ segmentation in whole-body mri using a two-stage fcn and weighting schemes. in M Liu, H-I Suk & Y Shi (eds), Machine Learning in Medical Imaging - 9th International Workshop, MLMI 2018, Held in Conjunction with MICCAI 2018, Proceedings. Lecture Notes in Computer Science, vol. 11046, Springer, pp. 346-354, International Workshop on Machine Learning in Medical Imaging (MLMI) 2018, Granada, Spain, 16/09/18. https://doi.org/10.1007/978-3-030-00919-9_40

Small organ segmentation in whole-body mri using a two-stage fcn and weighting schemes. / Valindria, Vanya V.; Lavdas, Ioannis; Cerrolaza, Juan et al.
Machine Learning in Medical Imaging - 9th International Workshop, MLMI 2018, Held in Conjunction with MICCAI 2018, Proceedings. ed. / Mingxia Liu; Heung-Il Suk; Yinghuan Shi. Springer, 2018. p. 346-354 (Lecture Notes in Computer Science; Vol. 11046).

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

TY - GEN

T1 - Small organ segmentation in whole-body mri using a two-stage fcn and weighting schemes

AU - Valindria, Vanya V.

AU - Lavdas, Ioannis

AU - Cerrolaza, Juan

AU - Aboagye, Eric O.

AU - Rockall, Andrea G.

AU - Rueckert, Daniel

AU - Glocker, Ben

N1 - Conference code: 9th

PY - 2018

Y1 - 2018

N2 - Accurate and robust segmentation of small organs in whole-body MRI is difficult due to anatomical variation and class imbalance. Recent deep network based approaches have demonstrated promising performance on abdominal multi-organ segmentations. However, the performance on small organs is still suboptimal as these occupy only small regions of the whole-body volumes with unclear boundaries and variable shapes. A coarse-to-fine, hierarchical strategy is a common approach to alleviate this problem, however, this might miss useful contextual information. We propose a two-stage approach with weighting schemes based on auto-context and spatial atlas priors. Our experiments show that the proposed approach can boost the segmentation accuracy of multiple small organs in whole-body MRI scans.

AB - Accurate and robust segmentation of small organs in whole-body MRI is difficult due to anatomical variation and class imbalance. Recent deep network based approaches have demonstrated promising performance on abdominal multi-organ segmentations. However, the performance on small organs is still suboptimal as these occupy only small regions of the whole-body volumes with unclear boundaries and variable shapes. A coarse-to-fine, hierarchical strategy is a common approach to alleviate this problem, however, this might miss useful contextual information. We propose a two-stage approach with weighting schemes based on auto-context and spatial atlas priors. Our experiments show that the proposed approach can boost the segmentation accuracy of multiple small organs in whole-body MRI scans.

UR - https://www.scopus.com/pages/publications/85054499122

U2 - 10.1007/978-3-030-00919-9_40

DO - 10.1007/978-3-030-00919-9_40

M3 - Conference Paper

AN - SCOPUS:85054499122

SN - 9783030009182

T3 - Lecture Notes in Computer Science

SP - 346

EP - 354

BT - Machine Learning in Medical Imaging - 9th International Workshop, MLMI 2018, Held in Conjunction with MICCAI 2018, Proceedings

A2 - Liu, Mingxia

A2 - Suk, Heung-Il

A2 - Shi, Yinghuan

PB - Springer

T2 - International Workshop on Machine Learning in Medical Imaging (MLMI) 2018

Y2 - 16 September 2018 through 16 September 2018

ER -

Valindria VV, Lavdas I, Cerrolaza J, Aboagye EO, Rockall AG, Rueckert D et al. Small organ segmentation in whole-body mri using a two-stage fcn and weighting schemes. In Liu M, Suk HI, Shi Y, editors, Machine Learning in Medical Imaging - 9th International Workshop, MLMI 2018, Held in Conjunction with MICCAI 2018, Proceedings. Springer. 2018. p. 346-354. (Lecture Notes in Computer Science). doi: 10.1007/978-3-030-00919-9_40

Small organ segmentation in whole-body mri using a two-stage fcn and weighting schemes

Abstract

Publication series

Workshop

Access to Document

Other files and links

Cite this