A deep symmetry convnet for stroke lesion segmentation

Yanran Wang; Aggelos K. Katsaggelos; Xue Wang; Todd B. Parrish

doi:10.1109/ICIP.2016.7532329

A deep symmetry convnet for stroke lesion segmentation

Yanran Wang, Aggelos K. Katsaggelos, Xue Wang, Todd B. Parrish

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

51 Scopus citations

Abstract

Stroke is one of the leading causes of death and disability. Clinically, to establish stroke patient prognosis, an accurate delineation of brain lesion is essential, which is time consuming and prone to subjective errors. In this paper, we propose a novel method call Deep Lesion Symmetry ConvNet to automatically segment chronic stroke lesions using MRI. An 8-layer 3D convolutional neural network is constructed to handle the MRI voxels. An additional CNN stream using the corresponding symmetric MRI voxels is combined, leading to a significant improvement in system performance. The high average dice coefficient achieved on our dataset based on data collected from three research labs demonstrates the effectiveness of our method.

Original language	English (US)
Title of host publication	2016 IEEE International Conference on Image Processing, ICIP 2016 - Proceedings
Publisher	IEEE Computer Society
Pages	111-115
Number of pages	5
ISBN (Electronic)	9781467399616
DOIs	https://doi.org/10.1109/ICIP.2016.7532329
State	Published - Aug 3 2016
Event	23rd IEEE International Conference on Image Processing, ICIP 2016 - Phoenix, United States Duration: Sep 25 2016 → Sep 28 2016

Publication series

Name	Proceedings - International Conference on Image Processing, ICIP
Volume	2016-August
ISSN (Print)	1522-4880

Other

Other	23rd IEEE International Conference on Image Processing, ICIP 2016
Country/Territory	United States
City	Phoenix
Period	9/25/16 → 9/28/16

Funding

This study was supported by NIDCD P50 DC012283 (Project Leader, Cynthia Thompson) and the Center for Translational Imaging, Northwestern University.

Keywords

Brain Quasi-symmetry
Deep Learning
Image Segmentation
MRI
Stroke

ASJC Scopus subject areas

Software
Computer Vision and Pattern Recognition
Signal Processing

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1109/ICIP.2016.7532329

Cite this

Wang, Y., Katsaggelos, A. K., Wang, X., & Parrish, T. B. (2016). A deep symmetry convnet for stroke lesion segmentation. In 2016 IEEE International Conference on Image Processing, ICIP 2016 - Proceedings (pp. 111-115). Article 7532329 (Proceedings - International Conference on Image Processing, ICIP; Vol. 2016-August). IEEE Computer Society. https://doi.org/10.1109/ICIP.2016.7532329

@inproceedings{cc169e08f8564a9390c3cb2d49929c9e,

title = "A deep symmetry convnet for stroke lesion segmentation",

abstract = "Stroke is one of the leading causes of death and disability. Clinically, to establish stroke patient prognosis, an accurate delineation of brain lesion is essential, which is time consuming and prone to subjective errors. In this paper, we propose a novel method call Deep Lesion Symmetry ConvNet to automatically segment chronic stroke lesions using MRI. An 8-layer 3D convolutional neural network is constructed to handle the MRI voxels. An additional CNN stream using the corresponding symmetric MRI voxels is combined, leading to a significant improvement in system performance. The high average dice coefficient achieved on our dataset based on data collected from three research labs demonstrates the effectiveness of our method.",

keywords = "Brain Quasi-symmetry, Deep Learning, Image Segmentation, MRI, Stroke",

author = "Yanran Wang and Katsaggelos, {Aggelos K.} and Xue Wang and Parrish, {Todd B.}",

note = "Publisher Copyright: {\textcopyright} 2016 IEEE.; 23rd IEEE International Conference on Image Processing, ICIP 2016 ; Conference date: 25-09-2016 Through 28-09-2016",

year = "2016",

month = aug,

day = "3",

doi = "10.1109/ICIP.2016.7532329",

language = "English (US)",

series = "Proceedings - International Conference on Image Processing, ICIP",

publisher = "IEEE Computer Society",

pages = "111--115",

booktitle = "2016 IEEE International Conference on Image Processing, ICIP 2016 - Proceedings",

address = "United States",

}

Wang, Y, Katsaggelos, AK, Wang, X & Parrish, TB 2016, A deep symmetry convnet for stroke lesion segmentation. in 2016 IEEE International Conference on Image Processing, ICIP 2016 - Proceedings., 7532329, Proceedings - International Conference on Image Processing, ICIP, vol. 2016-August, IEEE Computer Society, pp. 111-115, 23rd IEEE International Conference on Image Processing, ICIP 2016, Phoenix, United States, 9/25/16. https://doi.org/10.1109/ICIP.2016.7532329

A deep symmetry convnet for stroke lesion segmentation. / Wang, Yanran; Katsaggelos, Aggelos K.; Wang, Xue et al.
2016 IEEE International Conference on Image Processing, ICIP 2016 - Proceedings. IEEE Computer Society, 2016. p. 111-115 7532329 (Proceedings - International Conference on Image Processing, ICIP; Vol. 2016-August).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - A deep symmetry convnet for stroke lesion segmentation

AU - Wang, Yanran

AU - Katsaggelos, Aggelos K.

AU - Wang, Xue

AU - Parrish, Todd B.

PY - 2016/8/3

Y1 - 2016/8/3

N2 - Stroke is one of the leading causes of death and disability. Clinically, to establish stroke patient prognosis, an accurate delineation of brain lesion is essential, which is time consuming and prone to subjective errors. In this paper, we propose a novel method call Deep Lesion Symmetry ConvNet to automatically segment chronic stroke lesions using MRI. An 8-layer 3D convolutional neural network is constructed to handle the MRI voxels. An additional CNN stream using the corresponding symmetric MRI voxels is combined, leading to a significant improvement in system performance. The high average dice coefficient achieved on our dataset based on data collected from three research labs demonstrates the effectiveness of our method.

AB - Stroke is one of the leading causes of death and disability. Clinically, to establish stroke patient prognosis, an accurate delineation of brain lesion is essential, which is time consuming and prone to subjective errors. In this paper, we propose a novel method call Deep Lesion Symmetry ConvNet to automatically segment chronic stroke lesions using MRI. An 8-layer 3D convolutional neural network is constructed to handle the MRI voxels. An additional CNN stream using the corresponding symmetric MRI voxels is combined, leading to a significant improvement in system performance. The high average dice coefficient achieved on our dataset based on data collected from three research labs demonstrates the effectiveness of our method.

KW - Brain Quasi-symmetry

KW - Deep Learning

KW - Image Segmentation

KW - MRI

KW - Stroke

UR - http://www.scopus.com/inward/record.url?scp=85006757009&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85006757009&partnerID=8YFLogxK

U2 - 10.1109/ICIP.2016.7532329

DO - 10.1109/ICIP.2016.7532329

M3 - Conference contribution

AN - SCOPUS:85006757009

T3 - Proceedings - International Conference on Image Processing, ICIP

SP - 111

EP - 115

BT - 2016 IEEE International Conference on Image Processing, ICIP 2016 - Proceedings

PB - IEEE Computer Society

T2 - 23rd IEEE International Conference on Image Processing, ICIP 2016

Y2 - 25 September 2016 through 28 September 2016

ER -

A deep symmetry convnet for stroke lesion segmentation

Abstract

Publication series

Other

Funding

Keywords

ASJC Scopus subject areas

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this