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

Department of Computer Science

13 Citations (Scopus)

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
Publisher	Springer
Publication date	2018
Pages	238-250
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

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	16/09/2018 → 20/09/2018

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

Keywords

Airway segmentation
Bronchiectasis
Convolutional neural networks
CT
Data augmentation

Access to Document

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

https://arxiv.org/pdf/1808.04576.pdfLicence: Other

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 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. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) Vol. 11040 LNCS 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. Springer, 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 in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

Juarez, AGU, Tiddens, HAWM & de Bruijne, M 2018, Automatic airway segmentation in chest CT using convolutional neural networks. in 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, 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, 16/09/2018. https://doi.org/10.1007/978-3-030-00946-5_24

Juarez AGU, Tiddens HAWM, de Bruijne M. Automatic airway segmentation in chest CT using convolutional neural networks. In 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. 2018. p. 238-250. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Vol. 11040 LNCS). doi: 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. Springer, 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{d8c5079cb9734c3388fc31fd73b021c2,

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.",

keywords = "Airway segmentation, Bronchiectasis, Convolutional neural networks, CT, Data augmentation",

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

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",

pages = "238--250",

booktitle = "Image Analysis for Moving Organ, Breast, and Thoracic Images",

note = "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",

}

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.

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.

KW - Airway segmentation

KW - Bronchiectasis

KW - Convolutional neural networks

KW - CT

KW - Data augmentation

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 - Article in proceedings

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

PB - Springer

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 -

Automatic airway segmentation in chest CT using convolutional neural networks

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this