A Novel Lesion Segmentation Algorithm based on U-Net Network for Tuberculosis CT Image

Shaoyue Wen; Jing Liu; Wenge Xu

doi:10.1109/ICCAIS52680.2021.9624633

A Novel Lesion Segmentation Algorithm based on U-Net Network for Tuberculosis CT Image

Shaoyue Wen
, Jing Liu^*
, Wenge Xu

^*Corresponding author for this work

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

7 Citations (Scopus)

Abstract

Lung CT images provide several essential information for lung disease diagnosis and lung surgery. However, the traditional detection method through manual segmentation is laborious and time-consuming. This paper presents automatic tuberculosis (TB) lesion segmentation method based on U-Net neural network for detecting TB. In addition, we combined an edge detection algorithm called canny edge detector with this network to get a more accurate TB lesion boundary. This method is trained on two split databases with 3576 lung CT images obtained by data enhancement on 447 discontinuous lung CT images. The results show that the proposed approach is validated for complex TB lesions with a high dice coefficient (91.2%).

Original language	English
Title of host publication	10th International Conference on Control, Automation and Information Sciences, ICCAIS 2021 - Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	909-914
Number of pages	6
ISBN (Electronic)	9781665440295
DOIs	https://doi.org/10.1109/ICCAIS52680.2021.9624633
Publication status	Published - 2021
Externally published	Yes
Event	10th International Conference on Control, Automation and Information Sciences, ICCAIS 2021 - Xi'an, China Duration: 14 Oct 2021 → 17 Oct 2021

Publication series

Name	10th International Conference on Control, Automation and Information Sciences, ICCAIS 2021 - Proceedings

Conference

Conference	10th International Conference on Control, Automation and Information Sciences, ICCAIS 2021
Country/Territory	China
City	Xi'an
Period	14/10/21 → 17/10/21

UN SDGs

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

SDG 3 Good Health and Well-being

Keywords

Image processing
Lesions segmentation
Tuberculosis
Unet

Access to Document

10.1109/ICCAIS52680.2021.9624633

Cite this

Wen, S., Liu, J., & Xu, W. (2021). A Novel Lesion Segmentation Algorithm based on U-Net Network for Tuberculosis CT Image. In 10th International Conference on Control, Automation and Information Sciences, ICCAIS 2021 - Proceedings (pp. 909-914). (10th International Conference on Control, Automation and Information Sciences, ICCAIS 2021 - Proceedings). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICCAIS52680.2021.9624633

Wen, Shaoyue ; Liu, Jing ; Xu, Wenge. / A Novel Lesion Segmentation Algorithm based on U-Net Network for Tuberculosis CT Image. 10th International Conference on Control, Automation and Information Sciences, ICCAIS 2021 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2021. pp. 909-914 (10th International Conference on Control, Automation and Information Sciences, ICCAIS 2021 - Proceedings).

@inproceedings{4c4a7767ee3d4408bc78c631a6a95df7,

title = "A Novel Lesion Segmentation Algorithm based on U-Net Network for Tuberculosis CT Image",

abstract = "Lung CT images provide several essential information for lung disease diagnosis and lung surgery. However, the traditional detection method through manual segmentation is laborious and time-consuming. This paper presents automatic tuberculosis (TB) lesion segmentation method based on U-Net neural network for detecting TB. In addition, we combined an edge detection algorithm called canny edge detector with this network to get a more accurate TB lesion boundary. This method is trained on two split databases with 3576 lung CT images obtained by data enhancement on 447 discontinuous lung CT images. The results show that the proposed approach is validated for complex TB lesions with a high dice coefficient (91.2\%).",

keywords = "Image processing, Lesions segmentation, Tuberculosis, Unet",

author = "Shaoyue Wen and Jing Liu and Wenge Xu",

note = "Publisher Copyright: {\textcopyright} 2021 IEEE.; 10th International Conference on Control, Automation and Information Sciences, ICCAIS 2021 ; Conference date: 14-10-2021 Through 17-10-2021",

year = "2021",

doi = "10.1109/ICCAIS52680.2021.9624633",

language = "English",

series = "10th International Conference on Control, Automation and Information Sciences, ICCAIS 2021 - Proceedings",

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

pages = "909--914",

booktitle = "10th International Conference on Control, Automation and Information Sciences, ICCAIS 2021 - Proceedings",

}

Wen, S, Liu, J & Xu, W 2021, A Novel Lesion Segmentation Algorithm based on U-Net Network for Tuberculosis CT Image. in 10th International Conference on Control, Automation and Information Sciences, ICCAIS 2021 - Proceedings. 10th International Conference on Control, Automation and Information Sciences, ICCAIS 2021 - Proceedings, Institute of Electrical and Electronics Engineers Inc., pp. 909-914, 10th International Conference on Control, Automation and Information Sciences, ICCAIS 2021, Xi'an, China, 14/10/21. https://doi.org/10.1109/ICCAIS52680.2021.9624633

A Novel Lesion Segmentation Algorithm based on U-Net Network for Tuberculosis CT Image. / Wen, Shaoyue; Liu, Jing; Xu, Wenge.
10th International Conference on Control, Automation and Information Sciences, ICCAIS 2021 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2021. p. 909-914 (10th International Conference on Control, Automation and Information Sciences, ICCAIS 2021 - Proceedings).

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

TY - GEN

T1 - A Novel Lesion Segmentation Algorithm based on U-Net Network for Tuberculosis CT Image

AU - Wen, Shaoyue

AU - Liu, Jing

AU - Xu, Wenge

PY - 2021

Y1 - 2021

N2 - Lung CT images provide several essential information for lung disease diagnosis and lung surgery. However, the traditional detection method through manual segmentation is laborious and time-consuming. This paper presents automatic tuberculosis (TB) lesion segmentation method based on U-Net neural network for detecting TB. In addition, we combined an edge detection algorithm called canny edge detector with this network to get a more accurate TB lesion boundary. This method is trained on two split databases with 3576 lung CT images obtained by data enhancement on 447 discontinuous lung CT images. The results show that the proposed approach is validated for complex TB lesions with a high dice coefficient (91.2%).

AB - Lung CT images provide several essential information for lung disease diagnosis and lung surgery. However, the traditional detection method through manual segmentation is laborious and time-consuming. This paper presents automatic tuberculosis (TB) lesion segmentation method based on U-Net neural network for detecting TB. In addition, we combined an edge detection algorithm called canny edge detector with this network to get a more accurate TB lesion boundary. This method is trained on two split databases with 3576 lung CT images obtained by data enhancement on 447 discontinuous lung CT images. The results show that the proposed approach is validated for complex TB lesions with a high dice coefficient (91.2%).

KW - Image processing

KW - Lesions segmentation

KW - Tuberculosis

KW - Unet

UR - https://www.scopus.com/pages/publications/85123978055

U2 - 10.1109/ICCAIS52680.2021.9624633

DO - 10.1109/ICCAIS52680.2021.9624633

M3 - Conference Proceeding

AN - SCOPUS:85123978055

T3 - 10th International Conference on Control, Automation and Information Sciences, ICCAIS 2021 - Proceedings

SP - 909

EP - 914

BT - 10th International Conference on Control, Automation and Information Sciences, ICCAIS 2021 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 10th International Conference on Control, Automation and Information Sciences, ICCAIS 2021

Y2 - 14 October 2021 through 17 October 2021

ER -

Wen S, Liu J, Xu W. A Novel Lesion Segmentation Algorithm based on U-Net Network for Tuberculosis CT Image. In 10th International Conference on Control, Automation and Information Sciences, ICCAIS 2021 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2021. p. 909-914. (10th International Conference on Control, Automation and Information Sciences, ICCAIS 2021 - Proceedings). doi: 10.1109/ICCAIS52680.2021.9624633

A Novel Lesion Segmentation Algorithm based on U-Net Network for Tuberculosis CT Image

Abstract

Publication series

Conference

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this