3D ARCNN: An Asymmetric Residual CNN for Decreasing False Positive Rate of Lung Nodules Detection

Bowen Liu; Hong Song; Qiang Li; Yucong Lin; Jian Yang

doi:10.1109/BIBM55620.2022.9994973

3D ARCNN: An Asymmetric Residual CNN for Decreasing False Positive Rate of Lung Nodules Detection

Bowen Liu, Hong Song^*, Qiang Li, Yucong Lin^*, Jian Yang

^*Corresponding author for this work

Beijing Institute of Technology

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

2 Citations (Scopus)

Abstract

Lung cancer is with the highest morbidity and mortality, and early detection of cancerous changes is essential to reduce the risk of death. To achieve this, it is necessary to reduce the false positive rate of detection. In this paper, we propose a novel asymmetric residual network, called 3D ARCNN, to reduce false positive rate of lung nodules detection. 3D ARCNN consists of asymmetric convolutional and multilayer cascaded residual network structures. To solve the problem of deep neural network with large amounts of parameters and poor reproduction ability, the proposed model uses asymmetric convolution to reduce model parameters and enhance the generalization ability of the model. In addition, the model uses an internally cascaded multi-stage residual to prevent the gradient vanishing and exploding problems of deep networks. Experiments are performed on the public dataset LUNA16. Our method achieved high detection sensitivity of 91.6%, 92.7%, 93.2% and 95.8% at 1, 2, 4 and 8 false positives per scan, respectively, which got an average CPM index of 0.912. Experimental results show that the proposed 3D ARCNN is very useful for reducing the false positive rate of lung nodules in the clinic.

Original language	English
Title of host publication	Proceedings - 2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022
Editors	Donald Adjeroh, Qi Long, Xinghua Shi, Fei Guo, Xiaohua Hu, Srinivas Aluru, Giri Narasimhan, Jianxin Wang, Mingon Kang, Ananda M. Mondal, Jin Liu
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	1644-1647
Number of pages	4
ISBN (Electronic)	9781665468190
DOIs	https://doi.org/10.1109/BIBM55620.2022.9994973
Publication status	Published - 2022
Event	2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022 - Las Vegas, United States Duration: 6 Dec 2022 → 8 Dec 2022

Publication series

Name	Proceedings - 2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022

Conference

Conference	2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022
Country/Territory	United States
City	Las Vegas
Period	6/12/22 → 8/12/22

Keywords

Asymmetric convolution
False positive reduction
Multi-layer cascade
Residual network

UN SDGs

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

Access to Document

10.1109/BIBM55620.2022.9994973

Cite this

Liu, B., Song, H., Li, Q., Lin, Y., & Yang, J. (2022). 3D ARCNN: An Asymmetric Residual CNN for Decreasing False Positive Rate of Lung Nodules Detection. In D. Adjeroh, Q. Long, X. Shi, F. Guo, X. Hu, S. Aluru, G. Narasimhan, J. Wang, M. Kang, A. M. Mondal, & J. Liu (Eds.), Proceedings - 2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022 (pp. 1644-1647). (Proceedings - 2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/BIBM55620.2022.9994973

Liu, Bowen ; Song, Hong ; Li, Qiang et al. / 3D ARCNN : An Asymmetric Residual CNN for Decreasing False Positive Rate of Lung Nodules Detection. Proceedings - 2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022. editor / Donald Adjeroh ; Qi Long ; Xinghua Shi ; Fei Guo ; Xiaohua Hu ; Srinivas Aluru ; Giri Narasimhan ; Jianxin Wang ; Mingon Kang ; Ananda M. Mondal ; Jin Liu. Institute of Electrical and Electronics Engineers Inc., 2022. pp. 1644-1647 (Proceedings - 2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022).

@inproceedings{5ad85a30171947d8905a7d8f386432d7,

title = "3D ARCNN: An Asymmetric Residual CNN for Decreasing False Positive Rate of Lung Nodules Detection",

abstract = "Lung cancer is with the highest morbidity and mortality, and early detection of cancerous changes is essential to reduce the risk of death. To achieve this, it is necessary to reduce the false positive rate of detection. In this paper, we propose a novel asymmetric residual network, called 3D ARCNN, to reduce false positive rate of lung nodules detection. 3D ARCNN consists of asymmetric convolutional and multilayer cascaded residual network structures. To solve the problem of deep neural network with large amounts of parameters and poor reproduction ability, the proposed model uses asymmetric convolution to reduce model parameters and enhance the generalization ability of the model. In addition, the model uses an internally cascaded multi-stage residual to prevent the gradient vanishing and exploding problems of deep networks. Experiments are performed on the public dataset LUNA16. Our method achieved high detection sensitivity of 91.6%, 92.7%, 93.2% and 95.8% at 1, 2, 4 and 8 false positives per scan, respectively, which got an average CPM index of 0.912. Experimental results show that the proposed 3D ARCNN is very useful for reducing the false positive rate of lung nodules in the clinic.",

keywords = "Asymmetric convolution, False positive reduction, Multi-layer cascade, Residual network",

author = "Bowen Liu and Hong Song and Qiang Li and Yucong Lin and Jian Yang",

note = "Publisher Copyright: {\textcopyright} 2022 IEEE.; 2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022 ; Conference date: 06-12-2022 Through 08-12-2022",

year = "2022",

doi = "10.1109/BIBM55620.2022.9994973",

language = "English",

series = "Proceedings - 2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "1644--1647",

editor = "Donald Adjeroh and Qi Long and Xinghua Shi and Fei Guo and Xiaohua Hu and Srinivas Aluru and Giri Narasimhan and Jianxin Wang and Mingon Kang and Mondal, {Ananda M.} and Jin Liu",

booktitle = "Proceedings - 2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022",

address = "United States",

}

Liu, B, Song, H, Li, Q, Lin, Y & Yang, J 2022, 3D ARCNN: An Asymmetric Residual CNN for Decreasing False Positive Rate of Lung Nodules Detection. in D Adjeroh, Q Long, X Shi, F Guo, X Hu, S Aluru, G Narasimhan, J Wang, M Kang, AM Mondal & J Liu (eds), Proceedings - 2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022. Proceedings - 2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022, Institute of Electrical and Electronics Engineers Inc., pp. 1644-1647, 2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022, Las Vegas, United States, 6/12/22. https://doi.org/10.1109/BIBM55620.2022.9994973

3D ARCNN: An Asymmetric Residual CNN for Decreasing False Positive Rate of Lung Nodules Detection. / Liu, Bowen; Song, Hong; Li, Qiang et al.
Proceedings - 2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022. ed. / Donald Adjeroh; Qi Long; Xinghua Shi; Fei Guo; Xiaohua Hu; Srinivas Aluru; Giri Narasimhan; Jianxin Wang; Mingon Kang; Ananda M. Mondal; Jin Liu. Institute of Electrical and Electronics Engineers Inc., 2022. p. 1644-1647 (Proceedings - 2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022).

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

TY - GEN

T1 - 3D ARCNN

T2 - 2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022

AU - Liu, Bowen

AU - Song, Hong

AU - Li, Qiang

AU - Lin, Yucong

AU - Yang, Jian

PY - 2022

Y1 - 2022

N2 - Lung cancer is with the highest morbidity and mortality, and early detection of cancerous changes is essential to reduce the risk of death. To achieve this, it is necessary to reduce the false positive rate of detection. In this paper, we propose a novel asymmetric residual network, called 3D ARCNN, to reduce false positive rate of lung nodules detection. 3D ARCNN consists of asymmetric convolutional and multilayer cascaded residual network structures. To solve the problem of deep neural network with large amounts of parameters and poor reproduction ability, the proposed model uses asymmetric convolution to reduce model parameters and enhance the generalization ability of the model. In addition, the model uses an internally cascaded multi-stage residual to prevent the gradient vanishing and exploding problems of deep networks. Experiments are performed on the public dataset LUNA16. Our method achieved high detection sensitivity of 91.6%, 92.7%, 93.2% and 95.8% at 1, 2, 4 and 8 false positives per scan, respectively, which got an average CPM index of 0.912. Experimental results show that the proposed 3D ARCNN is very useful for reducing the false positive rate of lung nodules in the clinic.

AB - Lung cancer is with the highest morbidity and mortality, and early detection of cancerous changes is essential to reduce the risk of death. To achieve this, it is necessary to reduce the false positive rate of detection. In this paper, we propose a novel asymmetric residual network, called 3D ARCNN, to reduce false positive rate of lung nodules detection. 3D ARCNN consists of asymmetric convolutional and multilayer cascaded residual network structures. To solve the problem of deep neural network with large amounts of parameters and poor reproduction ability, the proposed model uses asymmetric convolution to reduce model parameters and enhance the generalization ability of the model. In addition, the model uses an internally cascaded multi-stage residual to prevent the gradient vanishing and exploding problems of deep networks. Experiments are performed on the public dataset LUNA16. Our method achieved high detection sensitivity of 91.6%, 92.7%, 93.2% and 95.8% at 1, 2, 4 and 8 false positives per scan, respectively, which got an average CPM index of 0.912. Experimental results show that the proposed 3D ARCNN is very useful for reducing the false positive rate of lung nodules in the clinic.

KW - Asymmetric convolution

KW - False positive reduction

KW - Multi-layer cascade

KW - Residual network

UR - http://www.scopus.com/inward/record.url?scp=85146653104&partnerID=8YFLogxK

U2 - 10.1109/BIBM55620.2022.9994973

DO - 10.1109/BIBM55620.2022.9994973

M3 - Conference contribution

AN - SCOPUS:85146653104

T3 - Proceedings - 2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022

SP - 1644

EP - 1647

BT - Proceedings - 2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022

A2 - Adjeroh, Donald

A2 - Long, Qi

A2 - Shi, Xinghua

A2 - Guo, Fei

A2 - Hu, Xiaohua

A2 - Aluru, Srinivas

A2 - Narasimhan, Giri

A2 - Wang, Jianxin

A2 - Kang, Mingon

A2 - Mondal, Ananda M.

A2 - Liu, Jin

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 6 December 2022 through 8 December 2022

ER -

Liu B, Song H, Li Q, Lin Y, Yang J. 3D ARCNN: An Asymmetric Residual CNN for Decreasing False Positive Rate of Lung Nodules Detection. In Adjeroh D, Long Q, Shi X, Guo F, Hu X, Aluru S, Narasimhan G, Wang J, Kang M, Mondal AM, Liu J, editors, Proceedings - 2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022. Institute of Electrical and Electronics Engineers Inc. 2022. p. 1644-1647. (Proceedings - 2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022). doi: 10.1109/BIBM55620.2022.9994973

3D ARCNN: An Asymmetric Residual CNN for Decreasing False Positive Rate of Lung Nodules Detection

Abstract

Publication series

Conference

Keywords

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this