Multi-scale neural network for automatic segmentation of ischemic strokes on acute perfusion images

Christian Lucas; Andre Kemmling; Amir Madany Mamlouk; Mattias P. Heinrich

doi:10.1109/ISBI.2018.8363767

Multi-scale neural network for automatic segmentation of ischemic strokes on acute perfusion images

Christian Lucas, Andre Kemmling, Amir Madany Mamlouk, Mattias P. Heinrich

Abstract

Perfusion imaging is very important for early assessment of strokes due to its ability to measure blood flow, transition times and dispersion. Deep learning approaches are able to perform automatic segmentations, but have had limited accuracy so far for clinically important small lesions. We present an extension to the popular U-Net architecture that concatenates higher-level features and improves their propagation throughout the network using additional skip connections. The new method is evaluated on public perfusion datasets and achieves substantially improved accuracy (33% lower surface distance than U-Net) and robustness in particular for smaller stroke lesions that are difficult to detect.

Original language	English
Title of host publication	2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018)
Number of pages	4
Volume	2018-April
Publisher	IEEE
Publication date	23.05.2018
Pages	1118-1121
ISBN (Print)	978-1-5386-3635-0, 978-1-5386-3637-4
ISBN (Electronic)	978-1-5386-3636-7
DOIs	https://doi.org/10.1109/ISBI.2018.8363767
Publication status	Published - 23.05.2018
Event	2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018) - Washington, DC, United States Duration: 04.04.2018 → 07.04.2018

UN SDGs

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

SDG 3 Good Health and Well-being
SDG 9 Industry, Innovation, and Infrastructure

Access to Document

10.1109/ISBI.2018.8363767

Cite this

Lucas, C., Kemmling, A., Mamlouk, A. M., & Heinrich, M. P. (2018). Multi-scale neural network for automatic segmentation of ischemic strokes on acute perfusion images. In 2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018) (Vol. 2018-April, pp. 1118-1121). (Proceedings - International Symposium on Biomedical Imaging). IEEE. https://doi.org/10.1109/ISBI.2018.8363767

@inproceedings{6f5409d318094de6b1f8d84bea0d3d25,

title = "Multi-scale neural network for automatic segmentation of ischemic strokes on acute perfusion images",

abstract = "Perfusion imaging is very important for early assessment of strokes due to its ability to measure blood flow, transition times and dispersion. Deep learning approaches are able to perform automatic segmentations, but have had limited accuracy so far for clinically important small lesions. We present an extension to the popular U-Net architecture that concatenates higher-level features and improves their propagation throughout the network using additional skip connections. The new method is evaluated on public perfusion datasets and achieves substantially improved accuracy (33\% lower surface distance than U-Net) and robustness in particular for smaller stroke lesions that are difficult to detect.",

author = "Christian Lucas and Andre Kemmling and Mamlouk, \{Amir Madany\} and Heinrich, \{Mattias P.\}",

note = "Funding Information: This work was supported by the Graduate School for Computing in Medicine and Life Sciences funded by Germany{\textquoteright}s Excellence Initiative [DFG GSC 235/2]. We would also like to thank Nvidia Corporation for their support by providing us with a Titan Xp graphics card. Publisher Copyright: {\textcopyright} 2018 IEEE. Copyright: Copyright 2018 Elsevier B.V., All rights reserved.; 2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018) ; Conference date: 04-04-2018 Through 07-04-2018",

year = "2018",

month = may,

day = "23",

doi = "10.1109/ISBI.2018.8363767",

language = "English",

isbn = "978-1-5386-3635-0",

volume = "2018-April",

series = "Proceedings - International Symposium on Biomedical Imaging",

publisher = "IEEE",

pages = "1118--1121",

booktitle = "2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018)",

}

Lucas, C, Kemmling, A, Mamlouk, AM & Heinrich, MP 2018, Multi-scale neural network for automatic segmentation of ischemic strokes on acute perfusion images. in 2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018). vol. 2018-April, Proceedings - International Symposium on Biomedical Imaging, IEEE, pp. 1118-1121, 2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018), Washington, DC, United States, 04.04.18. https://doi.org/10.1109/ISBI.2018.8363767

Multi-scale neural network for automatic segmentation of ischemic strokes on acute perfusion images. / Lucas, Christian; Kemmling, Andre; Mamlouk, Amir Madany et al.
2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018). Vol. 2018-April IEEE, 2018. p. 1118-1121 (Proceedings - International Symposium on Biomedical Imaging).

Research output: Chapters in Books/Reports/Conference Proceedings › Conference contribution › peer-review

TY - GEN

T1 - Multi-scale neural network for automatic segmentation of ischemic strokes on acute perfusion images

AU - Lucas, Christian

AU - Kemmling, Andre

AU - Mamlouk, Amir Madany

AU - Heinrich, Mattias P.

N1 - Funding Information: This work was supported by the Graduate School for Computing in Medicine and Life Sciences funded by Germany’s Excellence Initiative [DFG GSC 235/2]. We would also like to thank Nvidia Corporation for their support by providing us with a Titan Xp graphics card. Publisher Copyright: © 2018 IEEE. Copyright: Copyright 2018 Elsevier B.V., All rights reserved.

PY - 2018/5/23

Y1 - 2018/5/23

N2 - Perfusion imaging is very important for early assessment of strokes due to its ability to measure blood flow, transition times and dispersion. Deep learning approaches are able to perform automatic segmentations, but have had limited accuracy so far for clinically important small lesions. We present an extension to the popular U-Net architecture that concatenates higher-level features and improves their propagation throughout the network using additional skip connections. The new method is evaluated on public perfusion datasets and achieves substantially improved accuracy (33% lower surface distance than U-Net) and robustness in particular for smaller stroke lesions that are difficult to detect.

AB - Perfusion imaging is very important for early assessment of strokes due to its ability to measure blood flow, transition times and dispersion. Deep learning approaches are able to perform automatic segmentations, but have had limited accuracy so far for clinically important small lesions. We present an extension to the popular U-Net architecture that concatenates higher-level features and improves their propagation throughout the network using additional skip connections. The new method is evaluated on public perfusion datasets and achieves substantially improved accuracy (33% lower surface distance than U-Net) and robustness in particular for smaller stroke lesions that are difficult to detect.

UR - https://www.scopus.com/pages/publications/85048097151

U2 - 10.1109/ISBI.2018.8363767

DO - 10.1109/ISBI.2018.8363767

M3 - Conference contribution

SN - 978-1-5386-3635-0

SN - 978-1-5386-3637-4

VL - 2018-April

T3 - Proceedings - International Symposium on Biomedical Imaging

SP - 1118

EP - 1121

BT - 2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018)

PB - IEEE

T2 - 2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018)

Y2 - 4 April 2018 through 7 April 2018

ER -

Multi-scale neural network for automatic segmentation of ischemic strokes on acute perfusion images

Abstract

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this