UG-Net: Corneal Endothelial Cell Segmentation Based on Uncertainty Estimation and Soft Spatial Attention

Yinglin Zhang; Wei Wang; Biao Wang; Zicao  Cai; Dave Towey; Ruibin Bai; Risa Higashita; Jiang Liu

doi:10.1109/ISBI53787.2023.10230682

UG-Net: Corneal Endothelial Cell Segmentation Based on Uncertainty Estimation and Soft Spatial Attention

Yinglin Zhang, Wei Wang, Biao Wang, Zicao Cai, Dave Towey, Ruibin Bai, Risa Higashita, Jiang Liu

School of Computer Science

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

Abstract

Corneal endothelial cell segmentation of the microscope image is critical for clinical parameters quantification. However, the low-contrast regions are ambiguous and hard to segment. The uncertainty has been proved to be effective for detecting the ambiguous regions. Besides, spatial attention can guide the model to focus on the region of interest during the training process. This paper proposes an Uncertainty Guided Network (UG-Net) for corneal endothelial cell segmentation based on uncertainty estimation and spatial attention. We first use Bayesian approximation to obtain aleatoric and epistemic uncertainty maps. Then, two uncertainty maps are utilized to guide the model to focus on the low-contrast regions with uncertainty-based soft spatial attention. Experimental results show that the proposed method performs better than other state-of-the-art methods.

Original language	English
Title of host publication	2023 IEEE International Symposium on Biomedical Imaging, ISBI 2023
Publisher	IEEE Computer Society
ISBN (Electronic)	9781665473583
DOIs	https://doi.org/10.1109/ISBI53787.2023.10230682
Publication status	Published - 2023
Event	20th IEEE International Symposium on Biomedical Imaging, ISBI 2023 - Cartagena, Colombia Duration: 18 Apr 2023 → 21 Apr 2023

Publication series

Name	Proceedings - International Symposium on Biomedical Imaging
Volume	2023-April
ISSN (Print)	1945-7928
ISSN (Electronic)	1945-8452

Conference

Conference	20th IEEE International Symposium on Biomedical Imaging, ISBI 2023
Country/Territory	Colombia
City	Cartagena
Period	18/04/23 → 21/04/23

Keywords

Corneal Endothelial Cell
Deep Learning
Segmentation
Uncertainty Estimation

ASJC Scopus subject areas

Biomedical Engineering
Radiology Nuclear Medicine and imaging

Access to Document

10.1109/ISBI53787.2023.10230682

Cite this

Zhang, Y., Wang, W., Wang, B., Cai, Z., Towey, D., Bai, R., Higashita, R., & Liu, J. (2023). UG-Net: Corneal Endothelial Cell Segmentation Based on Uncertainty Estimation and Soft Spatial Attention. In 2023 IEEE International Symposium on Biomedical Imaging, ISBI 2023 (Proceedings - International Symposium on Biomedical Imaging; Vol. 2023-April). IEEE Computer Society. https://doi.org/10.1109/ISBI53787.2023.10230682

@inproceedings{daa0bb786c0443a785be24659dc3d9b6,

title = "UG-Net: Corneal Endothelial Cell Segmentation Based on Uncertainty Estimation and Soft Spatial Attention",

abstract = "Corneal endothelial cell segmentation of the microscope image is critical for clinical parameters quantification. However, the low-contrast regions are ambiguous and hard to segment. The uncertainty has been proved to be effective for detecting the ambiguous regions. Besides, spatial attention can guide the model to focus on the region of interest during the training process. This paper proposes an Uncertainty Guided Network (UG-Net) for corneal endothelial cell segmentation based on uncertainty estimation and spatial attention. We first use Bayesian approximation to obtain aleatoric and epistemic uncertainty maps. Then, two uncertainty maps are utilized to guide the model to focus on the low-contrast regions with uncertainty-based soft spatial attention. Experimental results show that the proposed method performs better than other state-of-the-art methods.",

keywords = "Corneal Endothelial Cell, Deep Learning, Segmentation, Uncertainty Estimation",

author = "Yinglin Zhang and Wei Wang and Biao Wang and Zicao Cai and Dave Towey and Ruibin Bai and Risa Higashita and Jiang Liu",

note = "Funding Information: This work was supported in part by Guangdong Provincial Department of Education (2020ZDZX3043), Guangdong Provincial Key Laboratory (2020B121201001), and Shenzhen Natural Science Fund (JCYJ20200109140820699 and the Stable Support Plan Program 20200925174052004). Funding Information: This work was supported in part by Guangdong Provincial Department of Education (2020ZDZX3043), Guangdong Provincial Key Laboratory (2020B121201001), and Shen-zhen Natural Science Fund (JCYJ20200109140820699 and the Stable Support Plan Program 20200925174052004) Publisher Copyright: {\textcopyright} 2023 IEEE.; 20th IEEE International Symposium on Biomedical Imaging, ISBI 2023 ; Conference date: 18-04-2023 Through 21-04-2023",

year = "2023",

doi = "10.1109/ISBI53787.2023.10230682",

language = "English",

series = "Proceedings - International Symposium on Biomedical Imaging",

publisher = "IEEE Computer Society",

booktitle = "2023 IEEE International Symposium on Biomedical Imaging, ISBI 2023",

address = "United States",

}

Zhang, Y, Wang, W, Wang, B, Cai, Z, Towey, D , Bai, R, Higashita, R & Liu, J 2023, UG-Net: Corneal Endothelial Cell Segmentation Based on Uncertainty Estimation and Soft Spatial Attention. in 2023 IEEE International Symposium on Biomedical Imaging, ISBI 2023. Proceedings - International Symposium on Biomedical Imaging, vol. 2023-April, IEEE Computer Society, 20th IEEE International Symposium on Biomedical Imaging, ISBI 2023, Cartagena, Colombia, 18/04/23. https://doi.org/10.1109/ISBI53787.2023.10230682

UG-Net: Corneal Endothelial Cell Segmentation Based on Uncertainty Estimation and Soft Spatial Attention. / Zhang, Yinglin; Wang, Wei; Wang, Biao et al.
2023 IEEE International Symposium on Biomedical Imaging, ISBI 2023. IEEE Computer Society, 2023. (Proceedings - International Symposium on Biomedical Imaging; Vol. 2023-April).

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

TY - GEN

T1 - UG-Net

T2 - 20th IEEE International Symposium on Biomedical Imaging, ISBI 2023

AU - Zhang, Yinglin

AU - Wang, Wei

AU - Wang, Biao

AU - Cai, Zicao

AU - Towey, Dave

AU - Bai, Ruibin

AU - Higashita, Risa

AU - Liu, Jiang

N1 - Funding Information: This work was supported in part by Guangdong Provincial Department of Education (2020ZDZX3043), Guangdong Provincial Key Laboratory (2020B121201001), and Shenzhen Natural Science Fund (JCYJ20200109140820699 and the Stable Support Plan Program 20200925174052004). Funding Information: This work was supported in part by Guangdong Provincial Department of Education (2020ZDZX3043), Guangdong Provincial Key Laboratory (2020B121201001), and Shen-zhen Natural Science Fund (JCYJ20200109140820699 and the Stable Support Plan Program 20200925174052004) Publisher Copyright: © 2023 IEEE.

PY - 2023

Y1 - 2023

N2 - Corneal endothelial cell segmentation of the microscope image is critical for clinical parameters quantification. However, the low-contrast regions are ambiguous and hard to segment. The uncertainty has been proved to be effective for detecting the ambiguous regions. Besides, spatial attention can guide the model to focus on the region of interest during the training process. This paper proposes an Uncertainty Guided Network (UG-Net) for corneal endothelial cell segmentation based on uncertainty estimation and spatial attention. We first use Bayesian approximation to obtain aleatoric and epistemic uncertainty maps. Then, two uncertainty maps are utilized to guide the model to focus on the low-contrast regions with uncertainty-based soft spatial attention. Experimental results show that the proposed method performs better than other state-of-the-art methods.

AB - Corneal endothelial cell segmentation of the microscope image is critical for clinical parameters quantification. However, the low-contrast regions are ambiguous and hard to segment. The uncertainty has been proved to be effective for detecting the ambiguous regions. Besides, spatial attention can guide the model to focus on the region of interest during the training process. This paper proposes an Uncertainty Guided Network (UG-Net) for corneal endothelial cell segmentation based on uncertainty estimation and spatial attention. We first use Bayesian approximation to obtain aleatoric and epistemic uncertainty maps. Then, two uncertainty maps are utilized to guide the model to focus on the low-contrast regions with uncertainty-based soft spatial attention. Experimental results show that the proposed method performs better than other state-of-the-art methods.

KW - Corneal Endothelial Cell

KW - Deep Learning

KW - Segmentation

KW - Uncertainty Estimation

UR - http://www.scopus.com/inward/record.url?scp=85172153567&partnerID=8YFLogxK

U2 - 10.1109/ISBI53787.2023.10230682

DO - 10.1109/ISBI53787.2023.10230682

M3 - Conference contribution

AN - SCOPUS:85172153567

T3 - Proceedings - International Symposium on Biomedical Imaging

BT - 2023 IEEE International Symposium on Biomedical Imaging, ISBI 2023

PB - IEEE Computer Society

Y2 - 18 April 2023 through 21 April 2023

ER -

UG-Net: Corneal Endothelial Cell Segmentation Based on Uncertainty Estimation and Soft Spatial Attention

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this