Segmentation and Uncertainty Measures of Cardiac Substrates within Optical Coherence Tomography Images via Convolutional Neural Networks

Ziyi Huang; Yu Gan; Theresa Lye; Darnel Theagene; Spandana Chintapalli; Simeran Virdi; Andrew Laine; Elsa Angelini; Christine P. Hendon

doi:10.1109/ISBI45749.2020.9098495

Segmentation and Uncertainty Measures of Cardiac Substrates within Optical Coherence Tomography Images via Convolutional Neural Networks

Ziyi Huang, Yu Gan, Theresa Lye, Darnel Theagene, Spandana Chintapalli, Simeran Virdi, Andrew Laine, Elsa Angelini, Christine P. Hendon

Department of Biomedical Engineering

Stevens Institute of Technology

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

Abstract

Segmentation of human cardiac tissue has a great potential to provide critical clinical guidance for Radiofrequency Ablation (RFA). Uncertainty in cardiac tissue segmentation is high because of the ambiguity of the subtle boundary and intra-linter-physician variations. In this paper, we proposed a deep learning framework for Optical Coherence Tomography (OCT) cardiac segmentation with uncertainty measurement. Our proposed method employs additional dropout layers to assess the uncertainty of pixel-wise label prediction. In addition, we improve the segmentation performance by using focal loss to put more weights on mis-classified examples. Experimental results show that our method achieves high accuracy on pixel-wise label prediction. The feasibility of our method for uncertainty measurement is also demonstrated with excellent correspondence between uncertain regions within OCT images and heterogeneous regions within corresponding histology images.

Original language	English
Title of host publication	ISBI 2020 - 2020 IEEE International Symposium on Biomedical Imaging
Publisher	IEEE Computer Society
Pages	1958-1961
Number of pages	4
ISBN (Electronic)	9781538693308
DOIs	https://doi.org/10.1109/ISBI45749.2020.9098495
State	Published - Apr 2020
Event	17th IEEE International Symposium on Biomedical Imaging, ISBI 2020 - Iowa City, United States Duration: Apr 3 2020 → Apr 7 2020

Publication series

Name	Proceedings - International Symposium on Biomedical Imaging
Volume	2020-April

Conference

Conference	17th IEEE International Symposium on Biomedical Imaging, ISBI 2020
Country/Territory	United States
City	Iowa City
Period	4/3/20 → 4/7/20

ASJC Scopus subject areas

Biomedical Engineering
Radiology Nuclear Medicine and imaging

Keywords

Cardiac tissue imaging
Convolutional neural networks
Optical coherence tomography
Semantic segmentation

Access to Document

10.1109/ISBI45749.2020.9098495

Cite this

Huang, Z., Gan, Y., Lye, T., Theagene, D., Chintapalli, S., Virdi, S., Laine, A., Angelini, E., & Hendon, C. P. (2020). Segmentation and Uncertainty Measures of Cardiac Substrates within Optical Coherence Tomography Images via Convolutional Neural Networks. In ISBI 2020 - 2020 IEEE International Symposium on Biomedical Imaging (pp. 1958-1961). Article 9098495 (Proceedings - International Symposium on Biomedical Imaging; Vol. 2020-April). IEEE Computer Society. https://doi.org/10.1109/ISBI45749.2020.9098495

@inproceedings{e3cf4153cd1f404d9d9f1e450c3e9fa6,

title = "Segmentation and Uncertainty Measures of Cardiac Substrates within Optical Coherence Tomography Images via Convolutional Neural Networks",

abstract = "Segmentation of human cardiac tissue has a great potential to provide critical clinical guidance for Radiofrequency Ablation (RFA). Uncertainty in cardiac tissue segmentation is high because of the ambiguity of the subtle boundary and intra-linter-physician variations. In this paper, we proposed a deep learning framework for Optical Coherence Tomography (OCT) cardiac segmentation with uncertainty measurement. Our proposed method employs additional dropout layers to assess the uncertainty of pixel-wise label prediction. In addition, we improve the segmentation performance by using focal loss to put more weights on mis-classified examples. Experimental results show that our method achieves high accuracy on pixel-wise label prediction. The feasibility of our method for uncertainty measurement is also demonstrated with excellent correspondence between uncertain regions within OCT images and heterogeneous regions within corresponding histology images.",

keywords = "Cardiac tissue imaging, Convolutional neural networks, Optical coherence tomography, Semantic segmentation",

author = "Ziyi Huang and Yu Gan and Theresa Lye and Darnel Theagene and Spandana Chintapalli and Simeran Virdi and Andrew Laine and Elsa Angelini and Hendon, {Christine P.}",

note = "Publisher Copyright: {\textcopyright} 2020 IEEE.; 17th IEEE International Symposium on Biomedical Imaging, ISBI 2020 ; Conference date: 03-04-2020 Through 07-04-2020",

year = "2020",

month = apr,

doi = "10.1109/ISBI45749.2020.9098495",

language = "English",

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

publisher = "IEEE Computer Society",

pages = "1958--1961",

booktitle = "ISBI 2020 - 2020 IEEE International Symposium on Biomedical Imaging",

}

Huang, Z, Gan, Y, Lye, T, Theagene, D, Chintapalli, S, Virdi, S, Laine, A, Angelini, E & Hendon, CP 2020, Segmentation and Uncertainty Measures of Cardiac Substrates within Optical Coherence Tomography Images via Convolutional Neural Networks. in ISBI 2020 - 2020 IEEE International Symposium on Biomedical Imaging., 9098495, Proceedings - International Symposium on Biomedical Imaging, vol. 2020-April, IEEE Computer Society, pp. 1958-1961, 17th IEEE International Symposium on Biomedical Imaging, ISBI 2020, Iowa City, United States, 4/3/20. https://doi.org/10.1109/ISBI45749.2020.9098495

Segmentation and Uncertainty Measures of Cardiac Substrates within Optical Coherence Tomography Images via Convolutional Neural Networks. / Huang, Ziyi; Gan, Yu; Lye, Theresa et al.
ISBI 2020 - 2020 IEEE International Symposium on Biomedical Imaging. IEEE Computer Society, 2020. p. 1958-1961 9098495 (Proceedings - International Symposium on Biomedical Imaging; Vol. 2020-April).

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

TY - GEN

T1 - Segmentation and Uncertainty Measures of Cardiac Substrates within Optical Coherence Tomography Images via Convolutional Neural Networks

AU - Huang, Ziyi

AU - Gan, Yu

AU - Lye, Theresa

AU - Theagene, Darnel

AU - Chintapalli, Spandana

AU - Virdi, Simeran

AU - Laine, Andrew

AU - Angelini, Elsa

AU - Hendon, Christine P.

PY - 2020/4

Y1 - 2020/4

N2 - Segmentation of human cardiac tissue has a great potential to provide critical clinical guidance for Radiofrequency Ablation (RFA). Uncertainty in cardiac tissue segmentation is high because of the ambiguity of the subtle boundary and intra-linter-physician variations. In this paper, we proposed a deep learning framework for Optical Coherence Tomography (OCT) cardiac segmentation with uncertainty measurement. Our proposed method employs additional dropout layers to assess the uncertainty of pixel-wise label prediction. In addition, we improve the segmentation performance by using focal loss to put more weights on mis-classified examples. Experimental results show that our method achieves high accuracy on pixel-wise label prediction. The feasibility of our method for uncertainty measurement is also demonstrated with excellent correspondence between uncertain regions within OCT images and heterogeneous regions within corresponding histology images.

AB - Segmentation of human cardiac tissue has a great potential to provide critical clinical guidance for Radiofrequency Ablation (RFA). Uncertainty in cardiac tissue segmentation is high because of the ambiguity of the subtle boundary and intra-linter-physician variations. In this paper, we proposed a deep learning framework for Optical Coherence Tomography (OCT) cardiac segmentation with uncertainty measurement. Our proposed method employs additional dropout layers to assess the uncertainty of pixel-wise label prediction. In addition, we improve the segmentation performance by using focal loss to put more weights on mis-classified examples. Experimental results show that our method achieves high accuracy on pixel-wise label prediction. The feasibility of our method for uncertainty measurement is also demonstrated with excellent correspondence between uncertain regions within OCT images and heterogeneous regions within corresponding histology images.

KW - Cardiac tissue imaging

KW - Convolutional neural networks

KW - Optical coherence tomography

KW - Semantic segmentation

UR - http://www.scopus.com/inward/record.url?scp=85085862382&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85085862382&partnerID=8YFLogxK

U2 - 10.1109/ISBI45749.2020.9098495

DO - 10.1109/ISBI45749.2020.9098495

M3 - Conference contribution

T3 - Proceedings - International Symposium on Biomedical Imaging

SP - 1958

EP - 1961

BT - ISBI 2020 - 2020 IEEE International Symposium on Biomedical Imaging

PB - IEEE Computer Society

T2 - 17th IEEE International Symposium on Biomedical Imaging, ISBI 2020

Y2 - 3 April 2020 through 7 April 2020

ER -

Huang Z, Gan Y, Lye T, Theagene D, Chintapalli S, Virdi S et al. Segmentation and Uncertainty Measures of Cardiac Substrates within Optical Coherence Tomography Images via Convolutional Neural Networks. In ISBI 2020 - 2020 IEEE International Symposium on Biomedical Imaging. IEEE Computer Society. 2020. p. 1958-1961. 9098495. (Proceedings - International Symposium on Biomedical Imaging). doi: 10.1109/ISBI45749.2020.9098495

Segmentation and Uncertainty Measures of Cardiac Substrates within Optical Coherence Tomography Images via Convolutional Neural Networks

Abstract

Publication series

Conference

ASJC Scopus subject areas

Keywords

Access to Document

Other files and links

Fingerprint

Cite this