Automatic airway segmentation in chest CT using convolutional neural networks

A. Garcia Uceda Juarez; H. A.W.M. Tiddens; M. de Bruijne

doi:10.1007/978-3-030-00946-5_24

Automatic airway segmentation in chest CT using convolutional neural networks

A. Garcia Uceda Juarez^*, H. A.W.M. Tiddens, M. de Bruijne

^*Corresponding author for this work

Research output: Chapter/Conference proceeding › Conference proceeding › Academic › peer-review

40 Citations (Scopus)

2 Downloads (Pure)

Abstract

Segmentation of the airway tree from chest computed tomography (CT) images is critical for quantitative assessment of airway diseases including bronchiectasis and chronic obstructive pulmonary disease (COPD). However, obtaining an accurate segmentation of airways from CT scans is difficult due to the high complexity of airway structures. Recently, deep convolutional neural networks (CNNs) have become the state-of-the-art for many segmentation tasks, and in particular the so-called Unet architecture for biomedical images. However, its application to the segmentation of airways still remains a challenging task. This work presents a simple but robust approach based on a 3D Unet to perform segmentation of airways from chest CTs. The method is trained on a dataset composed of 12 CTs, and tested on another 6 CTs. We evaluate the influence of different loss functions and data augmentation techniques, and reach an average dice coefficient of 0.8 between the ground-truth and our automated segmentations.

Original language	English
Title of host publication	Image Analysis for Moving Organ, Breast, and Thoracic Images - Third International Workshop, RAMBO 2018, Fourth International Workshop, BIA 2018, and First International Workshop, TIA 2018, Held in Conjunction with MICCAI 2018, Proceedings
Editors	David Snead, Emanuele Trucco, Danail Stoyanov, Zeike Taylor, Lena Maier-Hein, Nasir Rajpoot, Hrvoje Bogunovic, Francesco Ciompi, Mitko Veta, Mona K. Garvin, Xin Jan Chen, Anne Martel, Jeroen van der Laak, Yanwu Xu, Stephen McKenna
Publisher	Springer-Verlag
Pages	238-250
Number of pages	13
ISBN (Print)	9783030009458
DOIs	https://doi.org/10.1007/978-3-030-00946-5_24
Publication status	Published - 2018
Event	3rd International Workshop on Reconstruction and Analysis of Moving Body Organs, RAMBO 2018, 4th International Workshop on Breast Image Analysis, BIA 2018, and 1st International Workshop on Thoracic Image Analysis, TIA 2018, held in conjunction with 21st International Conference on Medical Imaging and Computer-Assisted Intervention, MICCAI 2018 - Granada, Spain Duration: 16 Sept 2018 → 20 Sept 2018

Publication series

Series	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	11040 LNCS
ISSN	0302-9743

Conference

Conference	3rd International Workshop on Reconstruction and Analysis of Moving Body Organs, RAMBO 2018, 4th International Workshop on Breast Image Analysis, BIA 2018, and 1st International Workshop on Thoracic Image Analysis, TIA 2018, held in conjunction with 21st International Conference on Medical Imaging and Computer-Assisted Intervention, MICCAI 2018
Country/Territory	Spain
City	Granada
Period	16/09/18 → 20/09/18

Bibliographical note

Acknowledgements:
Acknowledgments. This work has been funded by the EU Innovative Medicines Initiative (IMI). We would like to thank F. Dubost for his help with the experiments and with writing of this manuscript. We would also like to thank F. Calvet for sharing with us his implementation of elastic image deformation.

Publisher Copyright:
© Springer Nature Switzerland AG 2018.

Access to Document

10.1007/978-3-030-00946-5_24

1808.04576Accepted author manuscript, 4.03 MB

Cite this

Juarez, A. G. U., Tiddens, H. A. W. M., & de Bruijne, M. (2018). Automatic airway segmentation in chest CT using convolutional neural networks. In D. Snead, E. Trucco, D. Stoyanov, Z. Taylor, L. Maier-Hein, N. Rajpoot, H. Bogunovic, F. Ciompi, M. Veta, M. K. Garvin, X. J. Chen, A. Martel, J. van der Laak, Y. Xu, & S. McKenna (Eds.), Image Analysis for Moving Organ, Breast, and Thoracic Images - Third International Workshop, RAMBO 2018, Fourth International Workshop, BIA 2018, and First International Workshop, TIA 2018, Held in Conjunction with MICCAI 2018, Proceedings (pp. 238-250). Springer-Verlag. https://doi.org/10.1007/978-3-030-00946-5_24

Juarez, A. Garcia Uceda ; Tiddens, H. A.W.M. ; de Bruijne, M. / Automatic airway segmentation in chest CT using convolutional neural networks. Image Analysis for Moving Organ, Breast, and Thoracic Images - Third International Workshop, RAMBO 2018, Fourth International Workshop, BIA 2018, and First International Workshop, TIA 2018, Held in Conjunction with MICCAI 2018, Proceedings. editor / David Snead ; Emanuele Trucco ; Danail Stoyanov ; Zeike Taylor ; Lena Maier-Hein ; Nasir Rajpoot ; Hrvoje Bogunovic ; Francesco Ciompi ; Mitko Veta ; Mona K. Garvin ; Xin Jan Chen ; Anne Martel ; Jeroen van der Laak ; Yanwu Xu ; Stephen McKenna. Springer-Verlag, 2018. pp. 238-250 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Vol. 11040 LNCS).

@inproceedings{e7922448d54c4109b7d880f578ec51ba,

title = "Automatic airway segmentation in chest CT using convolutional neural networks",

abstract = "Segmentation of the airway tree from chest computed tomography (CT) images is critical for quantitative assessment of airway diseases including bronchiectasis and chronic obstructive pulmonary disease (COPD). However, obtaining an accurate segmentation of airways from CT scans is difficult due to the high complexity of airway structures. Recently, deep convolutional neural networks (CNNs) have become the state-of-the-art for many segmentation tasks, and in particular the so-called Unet architecture for biomedical images. However, its application to the segmentation of airways still remains a challenging task. This work presents a simple but robust approach based on a 3D Unet to perform segmentation of airways from chest CTs. The method is trained on a dataset composed of 12 CTs, and tested on another 6 CTs. We evaluate the influence of different loss functions and data augmentation techniques, and reach an average dice coefficient of 0.8 between the ground-truth and our automated segmentations.",

author = "Juarez, {A. Garcia Uceda} and Tiddens, {H. A.W.M.} and {de Bruijne}, M.",

note = "Acknowledgements: Acknowledgments. This work has been funded by the EU Innovative Medicines Initiative (IMI). We would like to thank F. Dubost for his help with the experiments and with writing of this manuscript. We would also like to thank F. Calvet for sharing with us his implementation of elastic image deformation. Publisher Copyright: {\textcopyright} Springer Nature Switzerland AG 2018.; 3rd International Workshop on Reconstruction and Analysis of Moving Body Organs, RAMBO 2018, 4th International Workshop on Breast Image Analysis, BIA 2018, and 1st International Workshop on Thoracic Image Analysis, TIA 2018, held in conjunction with 21st International Conference on Medical Imaging and Computer-Assisted Intervention, MICCAI 2018 ; Conference date: 16-09-2018 Through 20-09-2018",

year = "2018",

doi = "10.1007/978-3-030-00946-5_24",

language = "English",

isbn = "9783030009458",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer-Verlag",

pages = "238--250",

editor = "David Snead and Emanuele Trucco and Danail Stoyanov and Zeike Taylor and Lena Maier-Hein and Nasir Rajpoot and Hrvoje Bogunovic and Francesco Ciompi and Mitko Veta and Garvin, {Mona K.} and Chen, {Xin Jan} and Anne Martel and {van der Laak}, Jeroen and Yanwu Xu and Stephen McKenna",

booktitle = "Image Analysis for Moving Organ, Breast, and Thoracic Images - Third International Workshop, RAMBO 2018, Fourth International Workshop, BIA 2018, and First International Workshop, TIA 2018, Held in Conjunction with MICCAI 2018, Proceedings",

address = "Germany",

}

Juarez, AGU, Tiddens, HAWM & de Bruijne, M 2018, Automatic airway segmentation in chest CT using convolutional neural networks. in D Snead, E Trucco, D Stoyanov, Z Taylor, L Maier-Hein, N Rajpoot, H Bogunovic, F Ciompi, M Veta, MK Garvin, XJ Chen, A Martel, J van der Laak, Y Xu & S McKenna (eds), Image Analysis for Moving Organ, Breast, and Thoracic Images - Third International Workshop, RAMBO 2018, Fourth International Workshop, BIA 2018, and First International Workshop, TIA 2018, Held in Conjunction with MICCAI 2018, Proceedings. Springer-Verlag, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 11040 LNCS, pp. 238-250, 3rd International Workshop on Reconstruction and Analysis of Moving Body Organs, RAMBO 2018, 4th International Workshop on Breast Image Analysis, BIA 2018, and 1st International Workshop on Thoracic Image Analysis, TIA 2018, held in conjunction with 21st International Conference on Medical Imaging and Computer-Assisted Intervention, MICCAI 2018, Granada, Spain, 16/09/18. https://doi.org/10.1007/978-3-030-00946-5_24

Automatic airway segmentation in chest CT using convolutional neural networks. / Juarez, A. Garcia Uceda; Tiddens, H. A.W.M.; de Bruijne, M.
Image Analysis for Moving Organ, Breast, and Thoracic Images - Third International Workshop, RAMBO 2018, Fourth International Workshop, BIA 2018, and First International Workshop, TIA 2018, Held in Conjunction with MICCAI 2018, Proceedings. ed. / David Snead; Emanuele Trucco; Danail Stoyanov; Zeike Taylor; Lena Maier-Hein; Nasir Rajpoot; Hrvoje Bogunovic; Francesco Ciompi; Mitko Veta; Mona K. Garvin; Xin Jan Chen; Anne Martel; Jeroen van der Laak; Yanwu Xu; Stephen McKenna. Springer-Verlag, 2018. p. 238-250 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Vol. 11040 LNCS).

Research output: Chapter/Conference proceeding › Conference proceeding › Academic › peer-review

TY - GEN

T1 - Automatic airway segmentation in chest CT using convolutional neural networks

AU - Juarez, A. Garcia Uceda

AU - Tiddens, H. A.W.M.

AU - de Bruijne, M.

N1 - Acknowledgements: Acknowledgments. This work has been funded by the EU Innovative Medicines Initiative (IMI). We would like to thank F. Dubost for his help with the experiments and with writing of this manuscript. We would also like to thank F. Calvet for sharing with us his implementation of elastic image deformation. Publisher Copyright: © Springer Nature Switzerland AG 2018.

PY - 2018

Y1 - 2018

N2 - Segmentation of the airway tree from chest computed tomography (CT) images is critical for quantitative assessment of airway diseases including bronchiectasis and chronic obstructive pulmonary disease (COPD). However, obtaining an accurate segmentation of airways from CT scans is difficult due to the high complexity of airway structures. Recently, deep convolutional neural networks (CNNs) have become the state-of-the-art for many segmentation tasks, and in particular the so-called Unet architecture for biomedical images. However, its application to the segmentation of airways still remains a challenging task. This work presents a simple but robust approach based on a 3D Unet to perform segmentation of airways from chest CTs. The method is trained on a dataset composed of 12 CTs, and tested on another 6 CTs. We evaluate the influence of different loss functions and data augmentation techniques, and reach an average dice coefficient of 0.8 between the ground-truth and our automated segmentations.

AB - Segmentation of the airway tree from chest computed tomography (CT) images is critical for quantitative assessment of airway diseases including bronchiectasis and chronic obstructive pulmonary disease (COPD). However, obtaining an accurate segmentation of airways from CT scans is difficult due to the high complexity of airway structures. Recently, deep convolutional neural networks (CNNs) have become the state-of-the-art for many segmentation tasks, and in particular the so-called Unet architecture for biomedical images. However, its application to the segmentation of airways still remains a challenging task. This work presents a simple but robust approach based on a 3D Unet to perform segmentation of airways from chest CTs. The method is trained on a dataset composed of 12 CTs, and tested on another 6 CTs. We evaluate the influence of different loss functions and data augmentation techniques, and reach an average dice coefficient of 0.8 between the ground-truth and our automated segmentations.

UR - http://www.scopus.com/inward/record.url?scp=85053890171&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-00946-5_24

DO - 10.1007/978-3-030-00946-5_24

M3 - Conference proceeding

AN - SCOPUS:85053890171

SN - 9783030009458

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 238

EP - 250

BT - Image Analysis for Moving Organ, Breast, and Thoracic Images - Third International Workshop, RAMBO 2018, Fourth International Workshop, BIA 2018, and First International Workshop, TIA 2018, Held in Conjunction with MICCAI 2018, Proceedings

A2 - Snead, David

A2 - Trucco, Emanuele

A2 - Stoyanov, Danail

A2 - Taylor, Zeike

A2 - Maier-Hein, Lena

A2 - Rajpoot, Nasir

A2 - Bogunovic, Hrvoje

A2 - Ciompi, Francesco

A2 - Veta, Mitko

A2 - Garvin, Mona K.

A2 - Chen, Xin Jan

A2 - Martel, Anne

A2 - van der Laak, Jeroen

A2 - Xu, Yanwu

A2 - McKenna, Stephen

PB - Springer-Verlag

T2 - 3rd International Workshop on Reconstruction and Analysis of Moving Body Organs, RAMBO 2018, 4th International Workshop on Breast Image Analysis, BIA 2018, and 1st International Workshop on Thoracic Image Analysis, TIA 2018, held in conjunction with 21st International Conference on Medical Imaging and Computer-Assisted Intervention, MICCAI 2018

Y2 - 16 September 2018 through 20 September 2018

ER -

Juarez AGU, Tiddens HAWM, de Bruijne M. Automatic airway segmentation in chest CT using convolutional neural networks. In Snead D, Trucco E, Stoyanov D, Taylor Z, Maier-Hein L, Rajpoot N, Bogunovic H, Ciompi F, Veta M, Garvin MK, Chen XJ, Martel A, van der Laak J, Xu Y, McKenna S, editors, Image Analysis for Moving Organ, Breast, and Thoracic Images - Third International Workshop, RAMBO 2018, Fourth International Workshop, BIA 2018, and First International Workshop, TIA 2018, Held in Conjunction with MICCAI 2018, Proceedings. Springer-Verlag. 2018. p. 238-250. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Vol. 11040 LNCS). Epub 2018 Sept 12. doi: 10.1007/978-3-030-00946-5_24

Automatic airway segmentation in chest CT using convolutional neural networks

Abstract

Publication series

Conference

Bibliographical note

Access to Document

Other files and links

Fingerprint

Cite this