Identification of COPD from Multi-View Snapshots of 3D Lung Airway Tree via Deep CNN (2024)

Du, R., Qi, S., Feng, J., Xia, S., Kang, Y., Qian, W. (2020). Identification of COPD from Multi-View Snapshots of 3D Lung Airway Tree via Deep CNN. IEEE Access, 8, 38907-38919. Article 9000819. https://doi.org/10.1109/ACCESS.2020.2974617

Du, Ran ; Qi, Shouliang ; Feng, Jie et al. / Identification of COPD from Multi-View Snapshots of 3D Lung Airway Tree via Deep CNN. In: IEEE Access. 2020 ; Vol. 8. pp. 38907-38919.

@article{36f007ce34fe4d65be0ad4a7ac84024d,

title = "Identification of COPD from Multi-View Snapshots of 3D Lung Airway Tree via Deep CNN",

abstract = "Chronic obstructive pulmonary disease (COPD) is associated with morphologic abnormalities of airways with various patterns and severities. However, the way of effectively representing these abnormalities is lacking and whether these abnormalities enable to distinguish COPD from healthy controls is unknown. We propose to use deep convolutional neural network (CNN) to assess 3D lung airway tree from the perspective of computer vision, thereby constructing models of identifying COPD. After extracting airway trees from CT images, snapshots of their 3D visualizations are obtained from ventral, dorsal and isometric views. Using snapshots of each view, one deep CNN model is constructed and further optimized by Bayesian optimization algorithm to indentify COPD. The majority voting of three views presents the final prediction. Finally, the class-discriminative localization maps have been drawn to visually explain the CNNs' decisions. The models trained with single view (ventral, dorsal and isometric) of colorful snapshots present the similar accuracy (ACC) (86.8%, 87.5% and 86.7%) and the model after voting achieves the ACC of 88.2%. The ACC of the final voting model using gray and binary snapshots achieves 88.6% and 86.4%, respectively. Our specially designed CNNs outperform the typical off-the-shelf CNNs and the pre-trained CNNs with fine tuning. The class-discriminative regions of COPD are mainly located at central airways; however, regions in HC are scattering and located at peripheral airways. It is feasible to identify COPD using snapshots of 3D lung airway tree extracted from CT images via deep CNN. The CNNs can represent the abnormalities of airway tree in COPD and make accurate CT-based diagnosis of COPD.",

keywords = "Airway, Chronic obstructive pulmonary disease (COPD), Computed tomography (CT), Convolutional neural networks, Deep learning, Image classification",

author = "Ran Du and Shouliang Qi and Jie Feng and Shuyue Xia and Yan Kang and Wei Qian and Yao, {Yu Dong}",

note = "Publisher Copyright: {\textcopyright} 2013 IEEE.",

year = "2020",

doi = "10.1109/ACCESS.2020.2974617",

language = "English",

volume = "8",

pages = "38907--38919",

}

Du, R, Qi, S, Feng, J, Xia, S, Kang, Y, Qian, W 2020, 'Identification of COPD from Multi-View Snapshots of 3D Lung Airway Tree via Deep CNN', IEEE Access, vol. 8, 9000819, pp. 38907-38919. https://doi.org/10.1109/ACCESS.2020.2974617

Identification of COPD from Multi-View Snapshots of 3D Lung Airway Tree via Deep CNN. / Du, Ran; Qi, Shouliang; Feng, Jie et al.
In: IEEE Access, Vol. 8, 9000819, 2020, p. 38907-38919.

Research output: Contribution to journal › Article › peer-review

TY - JOUR

T1 - Identification of COPD from Multi-View Snapshots of 3D Lung Airway Tree via Deep CNN

AU - Du, Ran

AU - Qi, Shouliang

AU - Feng, Jie

AU - Xia, Shuyue

AU - Kang, Yan

AU - Qian, Wei

AU - Yao, Yu Dong

N1 - Publisher Copyright:© 2013 IEEE.

PY - 2020

Y1 - 2020

N2 - Chronic obstructive pulmonary disease (COPD) is associated with morphologic abnormalities of airways with various patterns and severities. However, the way of effectively representing these abnormalities is lacking and whether these abnormalities enable to distinguish COPD from healthy controls is unknown. We propose to use deep convolutional neural network (CNN) to assess 3D lung airway tree from the perspective of computer vision, thereby constructing models of identifying COPD. After extracting airway trees from CT images, snapshots of their 3D visualizations are obtained from ventral, dorsal and isometric views. Using snapshots of each view, one deep CNN model is constructed and further optimized by Bayesian optimization algorithm to indentify COPD. The majority voting of three views presents the final prediction. Finally, the class-discriminative localization maps have been drawn to visually explain the CNNs' decisions. The models trained with single view (ventral, dorsal and isometric) of colorful snapshots present the similar accuracy (ACC) (86.8%, 87.5% and 86.7%) and the model after voting achieves the ACC of 88.2%. The ACC of the final voting model using gray and binary snapshots achieves 88.6% and 86.4%, respectively. Our specially designed CNNs outperform the typical off-the-shelf CNNs and the pre-trained CNNs with fine tuning. The class-discriminative regions of COPD are mainly located at central airways; however, regions in HC are scattering and located at peripheral airways. It is feasible to identify COPD using snapshots of 3D lung airway tree extracted from CT images via deep CNN. The CNNs can represent the abnormalities of airway tree in COPD and make accurate CT-based diagnosis of COPD.

AB - Chronic obstructive pulmonary disease (COPD) is associated with morphologic abnormalities of airways with various patterns and severities. However, the way of effectively representing these abnormalities is lacking and whether these abnormalities enable to distinguish COPD from healthy controls is unknown. We propose to use deep convolutional neural network (CNN) to assess 3D lung airway tree from the perspective of computer vision, thereby constructing models of identifying COPD. After extracting airway trees from CT images, snapshots of their 3D visualizations are obtained from ventral, dorsal and isometric views. Using snapshots of each view, one deep CNN model is constructed and further optimized by Bayesian optimization algorithm to indentify COPD. The majority voting of three views presents the final prediction. Finally, the class-discriminative localization maps have been drawn to visually explain the CNNs' decisions. The models trained with single view (ventral, dorsal and isometric) of colorful snapshots present the similar accuracy (ACC) (86.8%, 87.5% and 86.7%) and the model after voting achieves the ACC of 88.2%. The ACC of the final voting model using gray and binary snapshots achieves 88.6% and 86.4%, respectively. Our specially designed CNNs outperform the typical off-the-shelf CNNs and the pre-trained CNNs with fine tuning. The class-discriminative regions of COPD are mainly located at central airways; however, regions in HC are scattering and located at peripheral airways. It is feasible to identify COPD using snapshots of 3D lung airway tree extracted from CT images via deep CNN. The CNNs can represent the abnormalities of airway tree in COPD and make accurate CT-based diagnosis of COPD.

KW - Airway

KW - Chronic obstructive pulmonary disease (COPD)

KW - Computed tomography (CT)

KW - Convolutional neural networks

KW - Deep learning

KW - Image classification

UR - http://www.scopus.com/inward/record.url?scp=85081920923&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85081920923&partnerID=8YFLogxK

U2 - 10.1109/ACCESS.2020.2974617

DO - 10.1109/ACCESS.2020.2974617

M3 - Article

AN - SCOPUS:85081920923

VL - 8

SP - 38907

EP - 38919

JO - IEEE Access

JF - IEEE Access

M1 - 9000819

ER -

Du R, Qi S, Feng J, Xia S, Kang Y, Qian W et al. Identification of COPD from Multi-View Snapshots of 3D Lung Airway Tree via Deep CNN. IEEE Access. 2020;8:38907-38919. 9000819. doi: 10.1109/ACCESS.2020.2974617